amats 2011 model user guide: documentation and validation...amats 2011 model user guide:...

AMATS 2011 Model User Guide:

Documentation and Validation

Prepared for

AMATS 4700 Elmore Road

Anchorage, AK 99507

December 2012

6 Hutton Centre Dr., Suite 700

Santa Ana, CA 92707

TBG121212233731SCO I

Contents

Section Page

Acronyms and Abbreviations ...............................................................................................................III

1.0 Introduction ............................................................................................................................... 1-1 1.1 Document Organization............................................................................................... 1-1 1.2 Background .................................................................................................................... 1-2 1.3 Updated Model Capabilities Summary ..................................................................... 1-4

1.3.1 Land Use Development, Zone Definition, and Network Updates ........... 1-4 1.3.2 Private School Trips ......................................................................................... 1-4 1.3.3 Transit Model Updates .................................................................................... 1-4 1.3.4 Modeling Tolls and Road User Charges ....................................................... 1-5 1.3.5 Growth Based Travel Forecasts ..................................................................... 1-5

2.0 Updated Model Capabilities - Detailed Processes and Parameters ................................ 2-1 2.1 Expansion of Modeling Area ....................................................................................... 2-1 2.2 Refinement of Traffic Analysis Zone System ............................................................ 2-2 2.3 Network Additions and Updates ............................................................................... 2-3 2.4 Regional and TAZ Development Assumptions ....................................................... 2-4 2.5 Transit Model Updates ................................................................................................. 2-4 2.6 Modeling Tolls and Road User Charges .................................................................... 2-5 2.7 Growth Based Travel Forecasts................................................................................... 2-6

3.0 Model Validation ...................................................................................................................... 3-1

4.0 Traffic Assignment and Volumes .......................................................................................... 4-1 4.1 Cordon Crossings ......................................................................................................... 4-1 4.2 Facility Type Comparisons .......................................................................................... 4-4 4.3 Comparisons of Statistical Performance .................................................................... 4-5 4.4 Individual Link Performance ...................................................................................... 4-6

5.0 Summary and Conclusions ..................................................................................................... 5-1

6.0 Bibliography .............................................................................................................................. 6-1

CONTENTS

II TBG121212233731SCO

Tables

2-1 Value of Time Equivalents by Market Segment .................................................................... 2-5

4-1 Summary of 2007 Average Daily Cordon Counts/Volumes and Differences .................. 4-3

4-2 Summary of 2007 Average Weekday Facility Class Counts/Volumes and Differences .................................................................................................................................. 4-5

4-3 Percent Difference Targets for Daily Volumes for Individual Links .................................. 4-6

Figures

2-1 AMATS Model Expansion ........................................................................................................ 2-2

2-2 AMATS Model Zone Refinement ............................................................................................ 2-3

4-1 Maximum Desirable Deviation in Total Screenline Volumes .............................................. 4-2

4-2 AMATS model Screenlines ....................................................................................................... 4-3

4-3 Comparison of All Available Measured and Modeled 2007 Weekday Traffic Volumes ....................................................................................................................................... 4-6

Appendices A Anchorage Travel Model Calibration and Validation Final Report (February 2005) B Economic Forecast Disaggregation C Model Inventory D Variable Dictionary AMATS Model Socioeconomic Database E Special Generators F Hotel/Motel Visitor Model G Screenlines

TBG121212233731SCO III

Acronyms and Abbreviations

MOA Municipality of Anchorage

AMATS Anchorage Metropolitan Area Transportation Solutions

MSB Matanuska-Susitna Borough

ADOT Alaska Department of Transportation

CTPP Census Transportation Planning Package

TAZ transportation (traffic) analysis zone

ISER Institute of Social and Economic Research

CBD Central Business District

VOT value of time

RMSE Root Mean Squared Error

pwtmb

Sticky Note

Capitalize "T" in Transportation? and capitalize value of time? Also, it appears that the font in Root Mean Squared Error is different. Please update as all others are capitalized.

pwtmb

Highlight

pwtmb

Highlight

pwtmb

Highlight

TBG121212233731SCO 1-1

1.0 Introduction

This document summarizes the development, calibration, and validation of the updated travel demand model for the Municipality of Anchorage (MOA) urbanized area (consisting of the Anchorage Bowl and Eagle River). The travel demand model is the primary tool of the Anchorage Metropolitan Area Transportation Solutions (AMATS) planning organization in its analysis and continuing development of the area’s Metropolitan Transportation Plan.

The AMATS model that is being updated is the 2002 AMATS model, which was validated for use in the 2025 AMATS Long-Range Transportation Plan (CH2M HILL, 2005b). The 2011 AMATS model was completed as part of the ongoing effort to develop an updated travel forecasting toolset to support the 2035 Metropolitan Transportation Plan.

The 2002 AMATS model has been updated to incorporate models procedures and software developed independently in the course of the recent project evaluation of the Seward Highway to Glenn Highway Connection. Work undertaken as part of the Seward Highway to Glenn Highway Connection project resulted in enhancements and extensions to the existing travel modeling capabilities, and these are documented in this report. The updates to the 2002 AMATS model result in a more refined regional travel model for use in MOA/AMATS transportation planning projects.

This report provides information on processing logic, parameters, and other changes made to support new model capabilities in a regional travel demand modeling context. It should be noted that while new parameters and modeling processes were developed and the modeling area was updated, the formulation and basic parameters of the model are unchanged from the 2002 AMATS model, which was calibrated using the Anchorage Household Travel Survey in 2002 (NuStats, 2002).

The purpose of this document is to specifically describe new capabilities, parameters, and associated data requirements as well as describe how those model capabilities have been incorporated into the pre-existing model framework. The updated model output was validated against 2007 weekday traffic counts, and results of that validation are presented in the validation section of this report.

The updated travel demand model will be referred to as the AMATS model for the remainder of this report.

1.1 Document Organization This document is organized into six sections and seven appendices.

Section 1: An executive summary and overview of the new model capabilities from a system and policy perspective. The basic modeling process and its general capabilities are also briefly described.

1.0 INTRODUCTION

1-2 TBG121212233731SCO

Section 2: A more complete overview of the updated model, its operation, and structure. Multiple subsections are included to describe structure, operation, parameters, and data requirements of model elements discussed.

Section 3: Documentation of the model validation.

Section 4: Documentation of approach, assumptions, and results of the model validation process using all available 2007 traffic count data.

Section 5: A discussion of the summary and conclusions for the updated model validation.

Section 6: Bibliography.

Appendix A: The 2002 AMATS Travel Model Calibration and Validation Report as an available reference to the full model and its submodels, including model installation.

Appendix B: Economic forecast disaggregation activity is documented as work completed for the Seward Highway to Glenn Highway Connection project and used as a basis of the updated model.

Appendix C: The updated model inventory.

Appendix D: Socioeconomic database dictionary. This appendix is an update to the 2002 AMATS model variable dictionary included in Appendix A of the 2002 AMATS Travel Model Calibration and Validation Report.

Appendix E: Special generators. This appendix is an update to the 2002 AMATS model special generator appendix (Appendix D) of the 2002 AMATS Travel Model Calibration and Validation Report.

Appendix F: Hotel/Motel visitor model. This appendix is an update to the 2002 AMATS model hotel/motel visitor model appendix (Appendix E) of the 2002 AMATS Travel Model Calibration and Validation Report. Updates have been made to the hotel/motel database, including average occupancy rates, to update the model to 2007 conditions.

Appendix G: Screenlines. This appendix is an update to the 2002 AMATS model screenline appendix (Appendix F) of the 2002 AMATS Travel Model Calibration and Validation Report. Model updates that were made for this process have resulted in a revalidated base year (2007) screenline analysis.

1.2 Background Travel modeling in the MOA is supported by the AMATS large-scale, multi-modal four-step travel model. The Anchorage Travel Demand Model (2002 AMATS model) was estimated using the 2002 Anchorage Household Travel Survey (NuStats, 2002) and is documented in the Anchorage Travel Model Calibration and Validation Final Report prepared by CH2M HILL (2005a) for the Municipality of Anchorage Traffic Department as part of the Seward Highway to Glenn Highway Connection project. This report is provided as Appendix A of this report.

The elements of the 2002 AMATS model include an auto ownership model, land use (and mixed land use) based trip generation, household disaggregation submodels, application of urban form factors, direct modeling of highway and transit service, calibrated mode split and auto

1.0 INTRODUCTION

TBG121212233731SCO 1-3

occupancy models, multimodal multiple-time-of-day period highway assignment procedures, and a model feedback loop to reflect impacts of road congestion of area travel patterns.

The 2002 AMATS model provides a strong foundation and appropriate departure point for addressing the needs of this model update. The range and types of policies which were considered for the Seward Highway to Glenn Highway Connection project along with the geographic extent of the project’s influence area required enhancement to the model in order to fully support the project’s requirements for travel forecasts and performance evaluation.

The AMATS model takes advantage of data resources describing the household characteristics and travel characteristics of the MOA and Matanuska-Susitna Borough (MSB) population, as well as supplemental datasets assembled from government agencies, the business communities, and other sources. These sources were gathered for the model development, and only the traffic counts were updated for the model update.

Primary resources used in the development of the new model include the following:

Alaska Department of Labor Wage & Salary Employment Inventories (ES-202–Unemployment Insurance Dataset)

MOA and Alaska Department of Transportation (ADOT) Road Characteristics Inventories

MOA Planning Department Demographics and Residential Use Inventories

United States Military Base Personnel and Active Armed Forces Data

2000 United States Census and Census Transportation Planning Package (CTPP)

2001 People Mover Transit Rider Survey

2002 Anchorage Household Travel Survey

2002 Student Enrollment and Public School Attendance Database

2002 People Mover Transit Rider Counts

2007 MOA and ADOT Traffic Counts

2007 People Mover Transit Schedules and Route Maps

These data were used to create a model of weekday travel in and around the MOA. The AMATS model can be used to test a range of assumptions regarding changing land use patterns, new and updated transportation projects and facilities, new or modified public policies regarding land use, and transportation services and facilities. The 2007 MOA and ADT traffic counts, and the 2007 People Mover transit schedules and route maps were used in support of this most recent model update process.

The AMATS model design and more recent updates were developed in consultation with, and input from, local transportation experts and the community. It is structured to be responsive to the expected range of questions and issues that the MOA expects to face regarding its transportation system through 2035.

As mentioned above, models and their coefficients and other parameters developed for the 2002 AMATS model are unchanged. Section 1.3 discusses the rationale and policy evaluation capabilities added as part of the update process.

1.0 INTRODUCTION

1-4 TBG121212233731SCO

1.3 Updated Model Capabilities Summary Arising out of the Seward Highway to Glenn Highway Connection travel modeling needs (and associated model enhancement process), a number of new and improved functions were incorporated into the AMATS model structure. Following completion of that project, the MOA and AMATS requested these capabilities be permanently incorporated into the regional travel model. These capabilities are designed to address a range of issues and policy options under consideration as part of the Seward Highway to Glenn Highway Connection project. The capabilities listed here also measurably extend the geographic and policy analysis capabilities of the regional travel model.

1.3.1 Land Use Development, Zone Definition, and Network Updates As part of the Seward Highway to Glenn Highway Connection project, the traffic analysis zone (TAZ) system was refined, and the boundaries of the modeling area were extended to include the MSB. This was done to improve representation of trips between the Anchorage Bowl and the MSB and also to provide better representation of local access and through collector streets in northeast Anchorage.

As part of the AMATS model update, new forecasts prepared for 2035 by the Institute of Social and Economic Research (ISER) were disaggregated to the updated zone system and applied to the AMATS model. This updated ISER economic forecast was the forecast used in the Seward Highway to Glenn Highway Connection project. The ISER report is available at the ISER website and is titled Economic and Demographic Projects for Alaska and Greater Anchorage 2010-2035, dated December 2009 (ISER, 2009). Appendix B is the final technical memorandum titled Anchorage and Matanuska-Susitna Borough 2035 Land Use Allocation and Forecast dated July 2010.

As part of this data update, networks (both highway and transit) were updated to include any new projects and transit services in the MOA and fully detailed highway and transit network descriptions were added for the MSB.

1.3.2 Private School Trips As a part of the model update, private and public school trips and school locations were described separately to account for their different trip distribution patterns. The 2002 AMATS school model described all school trips as one purpose. The primary purpose for stratifying public and private school trips are the mode of travel used for each. Primary modes of travel for public school trips are walk, public school bus, public transit, and automobile. The primary mode of travel for private school trips are automobile trips. Automobile trips are often trips linked to a work trip, in which a parent drives their child to school on their way to work.

1.3.3 Transit Model Updates The current transit speed model was supplemented to include classes for Central Business District (CBD) and exclusive bus lane transit operations. The 2002 AMATS model did not include dedicated bus lanes in the CBD, and that functionality was added to this update for transit planning purposes.

Also, the capability to model movements by automobile to and from park-n-ride facilities has been added and is further documented in Section 2 of this report.

1.0 INTRODUCTION

TBG121212233731SCO 1-5

1.3.4 Modeling Tolls and Road User Charges A major change in the AMATS model is the addition of utilizing generalized costs for impedances in demand modeling. From a policy perspective, this allows socio-economic markets to be modeled separately using stratified incremental costs and allows destination choice, mode choice, and route choice to reflect the impact of tolls or other road user charges on traveler behavior.

The handling of tolls and road user charges in the AMATS model is further documented in Section 2 of this report.

1.3.5 Growth Based Travel Forecasts The 2002 AMATS model used a completely synthetic travel demand modeling procedure. This means that models are calibrated based on known existing traveler behavior, and these models are applied to estimate total trips, resultant trip patterns, and vehicle and transit volumes for a given model year. Typically, adequate performance of the model is demonstrated by producing a synthetic forecast of existing conditions and comparing that to observed daily traffic volumes. Performance standards are considered to be met when statistical measures show acceptable correspondence.

In support of analyses required by the Seward Highway to Glenn Highway Connection project, it was concluded that higher standards of model versus count validation were required to adequately forecast corridor-specific, time-of-day and specific vehicle type volumes. To achieve this, a process known as adaptive assignment was used. Adaptive assignment uses known traffic volumes to adjust baseline triptables, and conforms the trip pattern estimates generated by the demand models to produce link volume estimates consistent to observed traffic count data. Using adaptive assignment, model validation becomes statistical analysis of how well the conformed tables can reproduce the counts used in adjustment.

For the model update process, an adaptive assignment based triptable was used to generate a baseline triptable of person trips. This baseline table, along with the difference between forecast year and base year synthetic tables, was used to generate future year forecasts.

The advantage of this process is the ability to achieve better correspondence with known existing conditions (traffic count data). Since the process does not redistribute the base increment of trips, studies where major shifts in travel and land use patterns are anticipated may not want to utilize this approach. However, as implemented, the approach does retain the full capability to model changes in mode choice and route choice with a more refined representation of both time-of-day and different vehicle categories. This capability has been implemented as an optional model process in the updated AMATS model. Details of model operation and data requirements are provided and further discussed in Section 2 of this report.

TBG121212233731SCO 2-1

2.0 Updated Model Capabilities - Detailed Processes and Parameters

The 2002 AMATS model was expanded and modified to better address future planning projects and policies in the MOA and the greater region (including the MSB). The purpose of the updated model capabilities and enhancements are to:

Improve and expand representation of transit Expand the study area to allow impacts of development immediately north of the MOA (i.e.,

the MSB) to be accounted for

Provide ability to model impacts of road user charges on travel patterns and traffic volumes

Address the long-term horizon year (2035) for the project

Support required inputs to detailed simulations of traffic in the project area

Enhance the underlying base calibration

This list of enhancements was based on likely forecasting requirements associated with the Seward Highway to Glenn Highway Connection project and the demands of focused modeling in a specific corridor. For corridor-level modeling, particularly when leading up to specific recommendations for facility and operational improvements, travel models typically are refined and revalidated to demonstrate the ability to better represent existing conditions in the vicinity of the corridor under study.

A comprehensive list of the model inventory (model structure) is included in Appendix C. The comprehensive model inventory includes all model updates and enhanced capabilities discussed in this section.

2.1 Expansion of Modeling Area The AMATS modeling area was expanded to encompass the defined urbanized area of the MSB. Figure 2-1 shows both the current and updated model boundaries. The 2002 AMATS model boundaries are shown in the AMATS Network (yellow) and encompasses the Anchorage Bowl and Chugiak-Eagle River. The updated AMATS model boundaries expand upon the 2002 AMATS model boundaries, and include the MSB Network Expansion area from the figure (orange).

2.0 UPDATED MODEL CAPABILITIES - DETAILED PROCESSES AND PARAMETERS

2-2 TBG121212233731SCO

FIGURE 2-1: AMATS MODEL EXPANSION

AMATS = Anchorage Metropolitan Area Transportation Solutions MSB = Matanuska-Susitna Borough

2.2 Refinement of Traffic Analysis Zone System Expansion of the study area involved expanding the current AMATS model study area to include the urbanized and planned urbanized portions of the MSB. The basic description was taken from other models:

MSB transportation demand model developed for the Mat-Su Borough Long-Range Transportation Plan (HDR, 2007)

Knik Arm Bridge and Toll Authority (KABATA) (HDR, 2006)

MSB Network Expansion

AMATS Network


TBG121212233731SCO 2-3

Some enhancements to that description were made. Adding the MSB involved expanding the highway and transit networks, adding traffic analysis zones, and defining the accesses for those zones to the network. Traffic count information was also added to support adaptive assignment and the resulting model validation. A total of 168 traffic analysis zones, 525 highway links (including 212 centroid links), and 8 transit routes were added to represent the MSB.

Figure 2-1 shows shaded areas representing the 2002 AMATS model and the area added for the updated AMATS model. Network and zone attributes were added and updated to support all AMATS model functions.

The TAZ structure within the Anchorage Bowl was also refined (with the addition of 24 zones) in the northeast part of the Anchorage Bowl to improve representation of access and refine the traffic forecasts. Figure 2-2 shows color coded zones that were updated for the AMATS model.

FIGURE 2-2: AMATS MODEL ZONE REFINEMENT

2.3 Network Additions and Updates The addition of the MSB to the transportation network, and the change in future year to 2035 required network updates. Figure 2-1 shows new roads and connections added to the transportation network to represent the MSB. These came from the Mat-Su Borough Long-Range Transportation Plan (HDR, 2007).


2-4 TBG121212233731SCO

2.4 Regional and TAZ Development Assumptions As part of the process of updating the model horizon year to 2035, new forecasts of regional economic activity were developed. These forecasts were then disaggregated to the level of TAZs for incorporation into the travel model. The forecasts were developed as an independent activity by the ISER, which is the primary source of economic forecasts in the MOA and MSB regions. Documentation of regional economic development assumptions is provided in the ISER report, Economic and Demographic Projections for Alaska and Greater Anchorage 2010-2035, developed for economic growth projections used for the Seward Highway to Glenn Highway Connection project (ISER, 2009).

Forecast disaggregation was performed as part of the Seward Highway to Glenn Highway Connection project activity and is documented separately and included as Appendix B. As part of this process, new TAZ socio-economic assumptions were developed for existing conditions for zones added in MSB.

2.5 Transit Model Updates Two updates were made to modeling transit network and paths. First, categories were added to the transit speed model to represent exclusive bus lane operation and running in mixed traffic in the CBD. Both special operation definitions are specified by time period (7:00-9:00 AM, 3:00-6:00 PM and off-peak) of operation. These are implemented as follows:

Exclusive Bus Lane: If the variable CLASS_X in the link record is 1 and if bus lane operation is allowed (USE_BUS_LANE in the period mode table is 1) then the base speed of buses will be equivalent to the highway free flow speed adjusted by stop spacing and stop dwell.

CBD Buses Running In Mixed Traffic: If the variable CBDTransit in the link record is 1 then use the CBDTransit value in the period mode table for stop spacing. Bus speed is determined by the base operating speed in the period mode table for the given highway facility class adjusted by assumed CBD stop spacing and assumed stop dwell.

In addition to the enhancements to the bus speed model, a new park-n-ride access model to account for park-n-ride facilities in Eagle River and at Huffman Road near the New Seward Highway was developed. Using TransCAD’s native functionality to find access paths using the underlying highway network description, the model was implemented by adding to two variables to node/intersection records and modifying the transit path finding script to include paths from park-n-rides. The node description was modified by adding these variables:

PNRServiceArea – Nonzero values designate park-n-ride locations. The value is the assumed service area radius in miles.

PNRName – Used to report the number of zones having feasible paths to the park-n-ride in the summary report.


TBG121212233731SCO 2-5

2.6 Modeling Tolls and Road User Charges The entire model chain was extensively modified to support generalized cost impedances. Generalized cost impedances allow representation of the network link travel time as a composite value of time that includes the equivalent value of time for direct monetary charges. As part of this update, to generate sensitivity to the area population’s economic strata, income tertiles generated by the household income disaggregation model and used in trip generation were carried through the entire model chain to traffic assignment. This creates separate path and travel mode share based on income market segment.

As part of this process, different equivalent values of time (VOT) were generated for each income tertile and for commercial travel. These values of time were applied consistently through demand modeling and for traffic assignment to generate separate market segment based forecasts. Table 2-1 shows the value of time equivalents by market segment, as assumed in 2002 dollars.

TABLE 2-1 Value of Time Equivalents by Market Segment

Segment Minutes per $ $ per Hour $ per Minute

Low Income (<$40,000 per Year) 5.00 12.00 0.200

Medium Income ($40,000 to $70,000 per Year) 4.75 12.63 0.211

High Income (>$70,000 per Year) 4.32 13.89 0.232

Commercial 2.29 26.15 0.436

Personal Time Average 12.84 0.214

Note: Assumed Values of Time are in 2002 dollars.

Operationally, the generalized cost model is implemented in three ways, affecting Trip Distribution, Mode Choice, and Highway Assignment:

The highway network used for path building was augmented to include variables and their calculation representing time period and market segment specific generalized costs for traversing a given link. For each directional link, these variables are calculated using the input direct link charge (in dollars) time and the median value of time (TIMEDIST), and then adjusted by a market segment correction factor.

“COST” fields in the link database represent the average private and commercial vehicle charges by direction of travel and time of day. “Market segment correction factors” adjust those values up or down to represent low or high income groups. These factors can be accessed and edited via the Global Parameters panel.

Both trip distribution and mode choice models have been modified to utilize market segment specific impedance tables to generate trip patterns and mode shares respectively. This process is largely transparent to the model user.

The traffic assignment procedures have been modified to include value of time as a parameter input. This has also necessitated utilization of a multi-class assignment procedure to assign trips both by VOT market segment and by vehicle class (single-occupant versus

pwtmb

Sticky Note

Jennifer will provide explanation of Market Segment correction in comment response to Teresa (earlier question).


2-6 TBG121212233731SCO

multiple-occupant). The latter was implemented to address evaluate of policies employing carpool and/or managed lane strategies.

2.7 Growth Based Travel Forecasts The other major update to the AMATS model was the addition of functionality to create adjusted triptables using adaptive assignment (OD Estimation) and specialized post processing and to directly utilize those base triptables in future forecasting. Recognizing that this process may not be appropriate for all forecasts needs, a two-path model process was implemented, which allows use of the standard synthetic forecasting procedures or the newer growth factoring process.

Steps involved in each of these are described below. It should be noted that all other updates to the travel model described above are applied regardless of which model process path is used. Specific functions have been defined and implemented to generate the base tables. These can be accessed as (“>” references submenus):

Utilities/Processors>OD Estimation> Reestimate [TP] Hwy Triptable (where TP=AM, PM, or off-peak)

This function invokes the AMATS traffic assignment routine in adaptive assignment mode. Prior to this, fields in the link database for automobile ground counts (suffixed with COUNT) and truck ground counts (suffixed with TRKCT) for each time period must be populated (Note: This is the current configuration implemented in the script. Available data may require script changes). Following population of count data, a highway binary network is built using the link database.

The standard full model for base or existing conditions must be run to generate required triptable inputs to the estimation process. Once these steps are completed, this function can be run. The output will be vehicle triptables for trucks, single occupant and multiple occupant autos ready for input to the person triptable re-estimation process.

Note: It is recommended that the correspondence between input count data and adjusted volumes be reviewed before applying the output triptables. Adaptive assignment is highly sensitive to the extent of coverage and the distribution of available count data.

Utilities/Processors>OD Estimation> Allocate Reestimated [TP] Triptables

To convert the vehicle triptables resulting from adaptive assignment to person trips (the form of the output of the trip distribution and time period factoring steps), the outputs of the synthetic mode choice model are used to recreate market segment and modal shares. Since the adjusted tables provide value for auto person trip interchanges (by using an average occupancy for auto trips), distribution of trips to modes and market segments can be calculated by measuring their relative market shares. This provides the adjusted tables in categories and a format where their values can be used as direct inputs to the mode choice models. These values are applied post-time-of-day factoring because one of the major functions of the adaptive assignment process is to readjust time-of-day shares.


TBG121212233731SCO 2-7

Once these person triptables are re-estimated, zone-to-zone total person trips are summarized to be reinput to the mode share estimation process. Data in this format can be combined with estimates of parallel trip growth to generate future forecasts.

The use of adaptive assignment requires invoking special functions to combine the base triptables representing existing trips with estimates of future trip growth to obtain total expected future travel for mode choice and traffic assignment input. To do this, triptables for both existing and the future condition under study must be available. The existing triptables are generated using the pure synthetic procedure (the AMATS model structure using naming and file management techniques to maintain separate synthetic and adjusted triptables for this process). Once these tables are available, the function Time of Day/Mode Choice>Growth Factoring> Growth Factor Trips can be invoked to calculate interchange growth and merge the result with the baseline triptables. These triptables can then be input to mode share estimation eventually resulting in traffic assignment.

As mentioned above, this represents one alternate path through the model process. Standard future synthetic output will be generated if the model chain is run without the Growth Factor steps. Filenames throughout are compatible.

These functions have been added primarily to supply the model user with options and more robust functionality to answer a wider range of planning and policy questions. Major new functionality includes:

the ability to vary assumptions regarding travel networks and development for the MSB and measure those impacts on the rest of the region

the ability to test and evaluate road pricing and other revenue generation and fee-based congestion management schemes

the capacity to utilize adaptive assignment to develop a high degree of correspondence between observed and modeled traffic for time periods and vehicle types

This functionality has been implemented within the AMATS model interface and architecture to promote ease of use and reduce additional training required.

TBG121212233731SCO 3-1

3.0 Model Validation

The Anchorage Transportation Demand Model is the base for travel modeling for regional travel modeling projects (including the recent 2035 Metropolitan Transportation Plan). All travel models need to be validated prior to being used to produce future-year forecasts. Model validation is defined as the process by which base year model results are compared to known sources of data such as traffic counts and transit ridership data.

Validation involves testing the model's predictive capabilities by replicating observed conditions. Although absolute criteria for assessing the validity of all model systems cannot be precisely defined, the Federal Highway Administration (FHWA) has developed a set of target values that serve as standards for evaluating the relative performance of traffic models. These standards are contained in the Model Validation and Reasonableness Checking Manual, prepared for the Transportation Model Improvement Program, FHWA, February 1997 (TMIP et al, 1997).

FHWA standards have been applied to the Anchorage Transportation Demand Model, and the results are documented in the report entitled, Anchorage Travel Model Calibration and Validation Final Report (CH2M HILL, 2005a), prepared for the Municipality of Anchorage Traffic Department. The validation process demonstrated that the Regional Transportation Model accurately simulated traffic volumes and transit usage in the Year 2002. This document shows that the updated model (with a base year of 2007) is performing at, or better than, the validation for the base year of 2002, and thus is performing per FHWA targets.

TBG121212233731SCO 4-1

4.0 Traffic Assignment and Volumes

One of the most important indicators of model performance is the direct measurement of how well the travel model, in an attempt to replicate base (in this case, 2007) conditions can mimic existing traffic conditions as indicated by field counts of traffic volumes throughout the network. Such model validation is industry standard practice and a necessary prerequisite to using the model to forecast expected futures. To guide travel modelers in establishing minimum standards for the travel model development process, FHWA has published a set of guidelines for acceptable performance.

To validate the AMATS model, four different types of model performance were measured. These included the following:

Measured versus modeled aggregate traffic counts for selected cordons or cutline crossings

Percent difference in volume between measured and modeled stratified by facility type

Correlation measures between measured and modeled volumes for all available count locations

Comparison of measured and modeled volumes at individual locations.

The count data used for comparison was all day weekday traffic counts for 2002. Sources were both ADOT and MOA counts supplied for this purpose and the ADOT Volumes map for the Anchorage and the Eagle River areas. These observed volumes were compared to the sum of the AM, PM, and off-peak period outputs for the 2002 Base travel model alternative utilizing trip distribution feedback. The results of each of these comparisons along with the corresponding Federal guidelines are presented below.

4.1 Cordon Crossings Measurement of traffic across cordons or screenlines provides an indication of how well the model performs in replicating major trip patterns and movements throughout the network. Cordons typically encompass all facilities that serve the same definable travel corridor. This allows for the fact that the model may not perfectly represent competition between parallel facilities. Federal guidelines for acceptable performance are based on a sliding scale that ranges from acceptable deviations of 65 percent for 5000 vehicles to 15 percent for 200,000 of more vehicles (Figure 4-1).

For evaluating the AMATS model’s performance, 28 separate cordons were defined. These ranged from specific east-west and north-south crossings in the downtown area to longer cordons (which, in some cases, bisect the entire Anchorage Bowl). The cordons used and their associated identifiers are shown in Figure 4-2.

4.0 TRAFFIC ASSIGNMENT AND VOLUMES

4-2 TBG121212233731SCO

FIGURE 4-1: MAXIMUM DESIRABLE DEVIATION IN TOTAL SCREENLINE VOLUMES

Once the cordons were defined, special procedures were written to automatically generate inputs to a set of spreadsheets to report their performance both in tabular and graphic form. The resulting spreadsheet reports which list and display counted and modeled traffic volumes for all links intersected by each cordon are provide as Appendix G. Table 4-1 itemizes performance of each cordon along with the measured deviation and the corresponding maximum deviation based on the Federal guideline. (Diff equals actual difference–Max Diff equals maximum desirable difference based on Federal guideline).

Based on Table 4-1, all cordons meet or exceed Federal guidelines. Examination of graphs showing the relative facility counts and volumes for each cordon (Appendix G, Screenlines), shows that, in almost all cases, the relative magnitude of traffic volumes on specific facilities in a given corridor (i.e., for a given cordon) correlates well with observed count data.


4-3

FIGURE 4-2: AMATS MODEL SCREENLINES

TABLE 4-1 Summary of 2007 Average Daily Cordon Counts/Volumes and Differences

ID Cordon Count Volume (At Count

Locations)

Difference (Count – Volume)

Percent Difference

101 N of Tudor Rd - Minnesota to Muldoon 139,094 158,520 (19,426) -14%

201 N of Dimond Av - Minnesota to Birch 139,899 157,550 (17,651) -13%

301 S of O'Malley - C St to Hillside 28,629 34,485 (5,856) -20%

401 S of Glenn Hwy - Ingra to Muldoon 57,463 67,518 (10,055) -17%

501 W of Muldoon - Tudor Rd to Glenn Hwy - - -

601 W of Boniface - Tudor Rd to Davis 59,665 69,506 (9,841) -16%

602 W of Birch - Rabbit Creek to Abbott 24,288 25,118 (830) -3%

701 E of Lake Otis - Tudor Rd to Commercial Rd 41,493 55,525 (14,032) -34%

702 E of Lake Otis - De Armoun to Dowling 25,818 30,014 (4,196) -16%

801 E of New Seward - Rabbit Creek to 3rd 2,648 2,637 11 0%

901 S of Int'l Airport - Minnesota to Lake Otis 156,295 182,177 (25,882) -17%

1001 S of 92nd/Abbott - Minnesota to Birch 85,181 91,563 (6,382) -7%

2001 N of Eagle River - Glenn Hwy 25,639 27,782 (2,143) -8%

2002 N of Eagle River Rd Access - Glenn to Birchwood 4,256 4,240 16 0%

2002

2003

Source: MOA

200520082101

201

101

702

2010

301

2017

901

801

602

701

601

1001

501

2014

2016

220021002013

20072006

2020

401

Source: MOA

Seward Highway

101

201

301

401

501

601

602

701

702

801

901

1001

2001

2005

2006

2007

2008

2010

2013

2014

2016

2017

2020

2100

2101

2200

Anchorage Bowl Chugiak - Eagle River

2002

2003


4-4 TBG121212233731SCO

TABLE 4-1 Summary of 2007 Average Daily Cordon Counts/Volumes and Differences

ID Cordon Count Volume (At Count

Locations)

Difference (Count – Volume)

Percent Difference

2003 S of Hiland - Glenn Hwy 46,740 53,005 (6,265) -13%

2005 N of 3rd St - C St to Port Access - - -

2006 W of Gambell -3rd St to 16th 30,445 26,690 3,755 12%

2007 S of 9th - L St to Medfra 16,152 23,190 (7,038) -44%

2008 E of Ingra - 3rd to 15th 54,841 60,810 (5,969) -11%

2010 N of Fireweed/Northern Lights - Minnesota to Muldoon 125,762 142,778 (17,016) -14%

2013 E of Northwood - Northern Lights to Int'l Airport 22,234 22,355 (121) -1%

2014 N of Int'l Airport - Spenard to Lake Otis 146,845 162,786 (15,941) -11%

2016 W of Minnesota - Raspberry to Klatt 29,291 32,125 (2,834) -10%

2017 S of Dimond/Abbott - Minnesota to Hillside 91,277 98,541 (7,264) -8%

2020 S End of Study Area - Seward Hwy - - -

2100 W of A - Ship Creek to Tudor 92,477 100,521 (8,044) -9%

2101 N of Ship Creek - C St to Muldoon 4,934 6,740 (1,806) -37%

2200 E of Ingra - 3rd to Tudor 140,498 153,674 (13,176) -9%

4.2 Facility Type Comparisons The second validation test performed on the AMATS model was to compare the percentage of measured versus modeled traffic for different facility types. This measures how well assumed and calculated speeds in the model are distributing traffic between the different types of facilities present in the model. As with cordons, FHWA has developed guidelines to measure what is considered to be acceptable performance for a travel model.

The facility types specified for comparison in the Federal guidelines are Freeways, Major Arterials, Minor Arterials, and Collectors. To create a basis for comparison with the defined facility classes, the AMATS model designations were grouped into these four categories using the following rules.

Freeways were designed as Freeways (CLASS=1) Expressways and major arterials were designated as Major Arterials (CLASS=2, 3) Minor arterials and frontage roads were designated as Minor Arterials (CLASS=4) Collectors and local roads were designated as Collectors (CLASS=5, 6)


4-5

Counts and volumes for each category were summarized and then the aggregate difference was compared to the corresponding Federal guideline. This summary is presented below in Table 4-2.

TABLE 4-2 Summary of 2007 Average Weekday Facility Class Counts/Volumes and Differences

Facility Class Count Volume Difference (Count – Volume)

Percent Difference

Freeway 993,166 1,013,511 (20,345) -2.05%

Major Arterial 4,660,042 4,787,115 (127,072) -2.73%

Minor Arterial 701,015 716,187 (15,172) -2.16%

Collector 186,083 183,709 2,374 1.28%

All Classes 6,641,742 6,808,895 (167,153) -2.52%

As the table indicates, measured and modeled volumes for freeways and major arterials are well within the identified targets. Particularly for the most significant facilities, this indicates good correspondence with traffic count information. This conclusion is reinforced by the results of statistical analysis presented below.

4.3 Comparisons of Statistical Performance While both the cordon and facility class comparisons use a subset of the count information, the third performance evaluation approach used measures the overall statistical performance of the model by comparing all available counts to corresponding model volumes. Figure 4-3 is a scattergram, which shows all measured and modeled volumes pairs. Fitting a linear function to this dataset yields an R2 or correlation coefficient of .9233, which exceeds the Federal guideline of .880. The second statistical measure used is root mean squared error (RMSE). The RMSE for the same dataset is 27.84. This RMSE value is comparable with findings for other regional travel models (no Federal guidelines are proscribed).


4-6 TBG121212233731SCO

FIGURE 4-3: COMPARISON OF ALL AVAILABLE MEASURED AND MODELED 2007 WEEKDAY TRAFFIC VOLUMES

4.4 Individual Link Performance Federal guidelines are also provided for percent difference targets between measured and modeled traffic volumes for individual links based on total link volume. These targets are listed in Table 4-3.

TABLE 4-3 Percent Difference Targets for Daily Volumes for Individual Links

Annual Average Daily Traffic Maximum Desirable Deviation

<1,000 0.60

1,000 - 2,500 0.47

2,500 – 5,000 0.36

5,000 – 10,000 0.29

10,000 – 25, 000 0.25

25,000 – 50,000 0.22

>50,000 0.21

Using these guidelines, all count locations’ specific performance was evaluated. Of the 655 individual counts, 233 exceed the Federal targets. However, when corrections are made for directional imbalances in the model due to count information being non-directional, this number is reduced to 80. When only those locations with a daily count of greater than 2500 are considered, the number of locations exceeding Federal targets is 53 or about 8.1 percent of all count locations. In general, this indicates that the AMATS model is replicating existing traffic behavior not only at the system-wide, facility group, and cordon levels, but also at the level of the individual road segment.

R² = 0.9233

0

5000

10000

15000

20000

25000

30000

35000

0 5000 10000 15000 20000 25000 30000 35000

20

07

Co

un

t

Model Volume

pwtmb

Sticky Note

Some folks may not know what the red line indicates or the outliers (observed vs forecasted data, R2). I suggest a key.

TBG121212233731SCO 5-1

5.0 Summary and Conclusions

A new and more robust travel demand model has been calibrated and validated for the Anchorage metropolitan area. This memorandum documents the new model structure, inputs, calibration and validation. The new model structure is significantly more robust and sophisticated than the travel model previously used in the Municipality of Anchorage (the 2002 AMATS model).

The new model structure incorporates a more refined TAZ system, including an extended modeling area inclusive of the MSB. The new model structure also includes a separation of private and public school trips, transit model updates for park-n-ride facilities and CBD bus lane capabilities, toll modeling and road user charges, and growth based travel forecasts.

The details of the model calibration are documented in this report. The new travel model validation is complete and demonstrated to be in compliance with FHWA guidelines.

Model-estimated vehicle volumes compared to observed traffic counts for each of 28 different cordons across road links are each within FHWA guideline criteria.

A statistical linear regression fit of model-estimated vehicle volumes compared to all available observed traffic counts throughout the Anchorage road network shows an R2

correlation coefficient of 0.9233, again bettering the FHWA guideline criteria of 0.880.

In sum, AMATS now has an up-to-date, context-sensitive, powerful, proven, and fully integrated travel demand modeling and analysis toolkit to address transportation planning and analysis needs.

APPENDIX A

TBG121212233731SCO A-1

Appendix A: Anchorage Travel Model Calibration and Validation Final Report

(February 2005)

APPENDIX A

TBG121212233731SCO A-2

This page intentionally left blank.

ANC/TP41158.DOC/050460017 FINAL DRAFT

F i n a l D r a f t

Anchorage Travel Model Calibration and Validation

Prepared for

Anchorage Metropolitan Transportation Solutions (AMATS)

February 2005

301 West Northern Lights Boulevard

Suite 601 Anchorage, Alaska 99503

ANC/TP41158.DOC/050460017 III FINAL DRAFT

Contents

Section Page

Introduction.......................................................................................................................................I-1 Background ........................................................................................................................... I-1 Document Organization ...................................................................................................... I-2 Important Definitions–Trips, Trip Origins and Destinations, Trip Productions and Trip Attractions............................................................................................................. I-2

1 Travel Model Overview .................................................................................................... 1-1

1.1 Basic Process.................................................................................................................. 1-1 1.2 Traveler Market (Household) Segmentation............................................................ 1-2 1.3 Trip Purpose Stratification .......................................................................................... 1-2 1.4 Time of Day Periods..................................................................................................... 1-3 1.5 Travel Mode Stratification........................................................................................... 1-3 1.6 Transportation Networks ............................................................................................ 1-4

2 Travel Model Structure ..................................................................................................... 2-1

2.1 Land Use and Demographic Inputs........................................................................... 2-1 2.2 Land Use/Density Inputs ........................................................................................... 2-2 2.3 Household Segmentation ............................................................................................ 2-2 2.4 Trip Distribution........................................................................................................... 2-5 2.5 Time of Day ................................................................................................................... 2-5 2.6 Mode Choice ................................................................................................................. 2-6 2.7 Traffic Assignment ....................................................................................................... 2-6 2.8 Summary........................................................................................................................ 2-7

3 Traveler Market (Household) Segmentation ................................................................ 3-1

3.1 Household Size ............................................................................................................. 3-1 3.2 Household Income ....................................................................................................... 3-3 3.3 Workers per Household .............................................................................................. 3-4 3.4 Auto Ownership ........................................................................................................... 3-6 3.5 Children per Household.............................................................................................. 3-7

4 Trip Generation .................................................................................................................. 4-1

4.1 Home Based Trip Production Models ....................................................................... 4-1 4.2 Non-Home Based Trip Production Models .............................................................. 4-2 4.3 Commercial Vehicle (Truck) Trips............................................................................. 4-3 4.4 Trip Attraction Models ................................................................................................ 4-3 4.5 Special Generators/Visitors ........................................................................................ 4-4 4.6 Other Special Considerations ..................................................................................... 4-6 4.7 Validation Results......................................................................................................... 4-6

CONTENTS

IV ANC/TP41158.DOC/050460017 FINAL DRAFT

5 Trip Distribution.................................................................................................................5-1 6 Time of Day Factoring........................................................................................................6-1 7 Mode Choice ........................................................................................................................7-1

7.1 Home Based Work Mode Choice................................................................................7-3 7.2 Home Based Shop Mode Choice.................................................................................7-3 7.3 Home Based School Mode Choice..............................................................................7-4 7.4 Home Based Other Mode Choice ...............................................................................7-5 7.5 Non-Home Based Trip Mode Choice.........................................................................7-5 7.6 Validation Results .........................................................................................................7-6

8 Traffic Assignment and Volumes....................................................................................8-1

8.1 Cordon Crossings..........................................................................................................8-1 8.2 Facility Type Comparisons ..........................................................................................8-3 8.3 Comparisons of Statistical Performance ....................................................................8-5 8.4 Individual Link Performance ......................................................................................8-6

9 Summary and Conclusions ...............................................................................................9-1 Appendix

A Variable Dictionary–Anchorage Travel Model Socioeconomic Database B Model Feedback Process C Commercial Vehicle Travel Model D Special Generators E Hotel/Motel Visitor Model F Screenlines Figure

2-1 Model Overview Flow Chart............................................................................................ 2-3 3-1 Average Household Size vs. % Total Households by Size Group.............................. 3-2 3-2 Zonal/Regional Income vs. % Total Households by Income Group ......................... 3-3 3-3 Average Workers/Household vs. % Total Households by Group............................ . 3-5 4-1 Survey vs. Modeled Trips by Community Council Area............................................. 4-7 5-1 Survey vs. Model Trip Length Frequencies (Home based work) ................................ 5-3 5-2 Survey vs. Model Trip Length Frequencies (Home based shop) ................................. 5-3 5-3 Survey vs. Model Trip Length Frequencies (Home based school) .............................. 5-4 5-4 Survey vs. Model Trip Length Frequencies (Home based other) ................................ 5-4 5-5 Survey vs. Model Trip Length Frequencies (Non-home based work) ........................ 5-5 5-6 Survey vs. Model Trip Length Frequencies (Non-home based nonwork) ................. 5-5 8-1 Maximum Desirable Deviation in Total Screenline Volumes....................................... 8-2 8-2 Anchorage Bowl Screenlines ............................................................................................. 8-3 8-3 Statistical Comparison of All Available Measured and Modeled 2002 Weekday

Traffic Volumes ................................................................................................................... 8-5

CONTENTS

ANC/TP41158.DOC/050460017 V FINAL DRAFT

Table

3-1 Comparison of Household Size Model Results (2002) and 2000 Census –Proportion of Household Size Group (Totals for Anchorage Area) ............................3-2

3-2 Comparison of Household Income Model Results (2002) and 2000 Census – Proportion of Household Income Groups (Totals for Anchorage Area).....................3-4

3-3 Comparison of Household Worker Model Results (2002) and 2000 Census –Proportion of Household Worker Groups (Totals for Anchorage Area) ....................3-5

3-4 Auto Ownership Model Variables and Coefficients ......................................................3-6 3-5 Comparison of Household Auto Ownership Model Results (2002) and 2000

Census – Proportion of Household Auto Ownership Groups (Totals for Anchorage Area)..................................................................................................................3-7

4-1 Home Based Work Trip Rates (Workers/HH by Income Group................................4-1 4-2 Home Based Shopping Trip Rates (Household Size by Income Group ......................4-1 4-3 Home Based School Trip Rates (Household Size by Children/HH) ...........................4-2 4-4 Home Based Other Trip Rates (Household Size by Income Group)............................4-2 4-5 Special Generators...............................................................................................................4-5 4-6 Total Trips by Trip Purpose (Survey-Model Comparison) .......................................... .4-6 5-1 Average Trip Length (in Minutes) by Trip Purpose.......................................................5-2 6-1 AM Peak Period Trip Purpose and Directional Factors.................................................6-1 6-2 PM Peak Period Trip Purpose and Directional Factors .................................................6-1 6-3 Off Peak Period Trip Purpose and Directional Factors..................................................6-2 7-1 Complete List of Mode Choice Variables (All Purposes) ..............................................7-2 7-2 Home Based Work Mode Choice Model Variables and Coefficients ..........................7-3 7-3 Home Based Shop Mode Choice Model Variables and Coefficients ...........................7-4 7-4 Home Based School Mode Choice Model Variables and Coefficients.........................7-4 7-5 Home Based Other Mode Choice Model Variables and Coefficients ..........................7-5 7-6 Non-Home Based Work Mode Choice Model Variables and Coefficients .................7-6 7-7 Non-Home Based Non-work Mode Choice Model Variables and Coefficients.........7-6 7-8 Survey vs. Model Mode Shares for Home Based Work Trips.......................................7-7 7-9 Survey vs. Model Mode Shares for Home Based Shop Trips........................................7-7 7-10 Survey vs. Model Mode Shares for Home Based School Trips.....................................7-8 7-11 Survey vs. Model Mode Shares for Home Based Other Trips ......................................7-8 7-12 Survey vs. Model Mode Shares for Non-Home Based Work Trips .............................7-9 7-13 Survey vs. Model Mode Shares for Non-Home Based Non-work Trips .....................7-9 7-14 Survey vs. Model Mode Shares for All Purposes .........................................................7-10 8-1 Summary of 2002 Weekday Cordon Counts/Volumes and Differences ....................8-4 8-2 Summary of 2002 Weekday Facility Class Counts/Volumes and Differences...........8-5 8-3 Percent Difference Targets for Daily Volumes for Individual Links ...........................8-6

CONTENTS

VI ANC/TP41158.DOC/050460017 FINAL DRAFT

ANC/TP41158.DOC/050460017 I-1 FINAL DRAFT

Introduction

This document summarizes the development, calibration, and validation of a new travel demand model for the Anchorage, Alaska urbanized area. The travel demand model is a primary tool of the Anchorage Metropolitan Area Transportation Solutions (AMATS) planning organization in its analysis and continuing development of the area’s Long Range Transportation Plan.

The report summarizes the features of the new travel demand model, its internal submodels and linkages structure, data inputs, relevant major assumptions, and, the tests that were performed to validate and measure the performance of the model in reproducing current travel behavior and transportation system usage.

Background The design of the travel demand model takes advantage of new data resources describing the household characteristics and travel characteristics of the Anchorage area population, as well as supplemental datasets assembled from government agencies, the Anchorage business community and other sources.

Primary resources used in the development of the new model include the following:

• 2002 Anchorage Household Travel Survey;

• 2001 PeopleMover Transit Rider Survey;

• 2000 US Census and Census Transportation Planning Package (CTPP);

• 2002 Student Enrollment and Public School Attendance Database

• U.S. Military Base Personnel and Active Armed Forces Data

• 2002 PeopleMover Transit Rider Counts

• 2002 Municipality of Anchorage (MOA) and Alaska Department of Transportation (ADOT) Traffic Counts;

• MOA Planning Department Demographics and Residential Use Inventories;

• Alaska Department of Labor Wage & Salary Employment Inventories (ES-202–Unemployment Insurance Dataset)

• MOA and ADOT Road Characteristics Inventories

• PeopleMover Transit Schedules and Route Maps.

These data were used to create a new model of weekday travel in and around the Anchorage area. The new travel demand model can be used to test a range of assumptions regarding changing land use patterns; new and updated transportation projects and

INTRODUCTION

I-2 ANC/TP41158.DOC/050460017 FINAL DRAFT

facilities; and, new or modified public policies regarding land use, and transportation services and facilities.

The new travel model design was developed in consultation and with input from local transportation experts and the general community. It is structured to be a responsive as possible to the expected range of questions and issues that Anchorage expects to face regarding its transportation system through 2025.

Document Organization The documentation is divided into nine sections. The first section, Travel Model Overview, discusses the basic features, capabilities and dimensions of the model. The second section, Travel Model Structure, provides more information on the actual modeling process and further details some of the topics in Travel Model Overview. The third section, Traveler Market (Household) Segmentation¸ specifically documents the set of models developed to estimate residential market segmentation and provides data on their ability to reproduce 2002 actual conditions. The fourth section, Trip Generation, documents the models used to estimate likelihood of trips either being produced and/or attracted to specific developments and geographic locations. The fifth section, Trip Distribution, discusses the models which were developed to predict general trip patterns and how well they match 2002 known conditions. The sixth section, Time of Day, illustrates the basis used for segmenting daily travel estimates into time period specific estimates. The seventh section, Mode Choice, documents the models and the assumptions used to translate general estimates of travel into those, which are specific to mode of travel. This also details tests and comparisons of model prediction results to 2002 actual travel mode shares. The eighth section, Trip Assignments and Volumes, compares the output of the model in terms of specific traffic volumes for different roads and corridors to independent traffic count data collected in the field. This comparison is evaluated against accepted industry guidelines for travel demand model performance. The ninth section summarizes the results, findings and recommendations arising out of the model calibration and validation work.

Important Definitions–Trips, Trip Origins and Destinations, Trip Productions and Trip Attractions Common nomenclature is used within the travel modeling community to characterize travel and trip-making. Several definitions are defined below:

Trip: A trip is a one-way journey from a point of origin to a destination for a specific purpose, i.e., a home to work journey. Unless otherwise specified, trip means a single person journey. Trips include walking and bicycling (referred to as non-motorized trips) when these are the primary modes of travel. Motorized trips include driving or riding as a passenger in various vehicles – autos, trucks, buses, school buses, etc.

Sometimes trips will be defined as a mode journey, as in a vehicle trip which is used to represent a vehicle driver journey or a transit trip which applies to a transit passenger journey.

Trip Origins and Destinations: Each trip has two end points -- an origin and a destination.

INTRODUCTION

ANC/TP41158.DOC/050460017 I-3 FINAL DRAFT

Trip Productions and Attractions: It is often useful to relate trips to the trip-maker and his/her residence location. Accordingly, all trips which start or end at the trip-maker’s home are defined as Home-Based Trip Productions at the home-end and as Home-Based Trip Attractions at the non-home trip end. For trips which neither start or end at the trip-maker’s residence, the origin is defined as the Trip Production and the destination is defined as the Trip Attraction.

INTRODUCTION

I-4 ANC/TP41158.DOC/050460017 FINAL DRAFT

ANC/TP41158.DOC/050460017 1-1 FINAL DRAFT

SECTION 1

Travel Model Overview

The Anchorage area travel model is designed to provide forecasts of prospective usage of the transportation system given a wide variety of assumptions about the available network and service and service cost policies that may be assumed for future conditions. To do this, salient characteristics of the traveling population and the available services must be measured and then translated into statistical models and abstract representation. The goal here is to capture, as well as possible, the significant determinants of traveler response to changes in the system and policies.

1.1 Basic Process The framework for this process is travel activity during a typical weekday. To provide effective models that are sensitive to the range of differences in the characteristics of both the travelers and their travel opportunities, the problem is broken down into specific travel market segments. This section of the report discusses the key market segmentation definitions and assumptions used in the Anchorage model. A brief introduction to the modeling process itself is provided here; the model process is presented in significantly more detail in the remainder of the report.

To estimate travel demand, the modeling process is broken into six basic steps. These are:

• Household Disaggregation–Estimation of the socio-economic groups or market segments that prospective travelers belong to…

• Trip Generation–Estimation of expected tripmaking rates for each of these segments for a given set of tripmaking purposes…

• Trip Distribution–Estimation of expected origin-destination patterns for the defined trip purposes and by a set of time periods…

• Time of Day Segmentation–Estimation of the shares of origin-destination trips expected for defined time periods and stratified by trip purpose …

• Modal Choice–Estimation of the shares of origin-destination trips expected for separate travel modes for each market segment, trip purpose and time period…

• Trip Assignment–Estimation of the expected routes (by mode) for the mode specific interchange sets defined by Modal Choice…

Understanding how these steps are applied requires a picture of how all of the above segments are defined.

TRAVEL MODEL OVERVIEW

1-2 ANC/TP41158.DOC/050460017 FINAL DRAFT

1.2 Traveler Market (Household) Segmentation For the Anchorage travel demand model, several separate and distinct submodels were developed to segment households in each subarea (i.e. traffic analysis zone) into five different types of groups. These groups provide breakdowns of the data into different household sizes, range of household incomes, as well as, groups representing different numbers of workers per household and auto ownership groups. The specific defined segments are listed below.

• Household Size–one, two, three, four or more persons • Household Income–less than $40K, at least $40K, but less than $70K, $70K or more • Auto Ownership–zero, one, two, three or more vehicles • Workers in Household–zero, one, two, three or more workers • Children in Household–zero, one, two, three or more children

With the exception of the auto ownership classes, all market segment proportions are derived from the estimated average value for the traffic analysis zone.

This segmentation is used to estimate trip generation and mode choice components of the travel demand model. Cross products of segmentation pairs are used to identify more specific market segments. For instance, all two-person households which own two autos. The specific combinations of segments used for the trip generation and mode choice models are identified and described in those sections of this document.

1.3 Trip Purpose Stratification Travel models recognize that trips made for different purposes have different characteristics. To improve accuracy of predictive models, specific models are estimated to represent each defined trip purpose. In the case of the Anchorage, such stratification is used in the trip generation, trip distribution, time-of-day factoring and mode choice submodels. Application of trip purpose stratification recognizes that the basic rates of trips produced at origin locations and attracted to specific destinations; the expected length of trip; the time of day a given trip is made; and, the relative likelihood of using a particular travel mode (e.g. driving an auto versus being a passenger versus riding the bus) are all influenced by the purpose of the trip.

The Anchorage travel model has six person trip purposes and two additional commercial vehicle (i.e., truck) trip purposes. These six person trip purposes are:

• Trips between home and work (Home based Work–HBW) • Trips between home and shopping (Home based Shop–HBS) • Trips between home and school (Home based School–HBSch) • Trips between home and any other type of destination (Home based Other–HBO) • Trips between work and other places besides work (Non-home based Work–NHBW) • Trips between two non-home/non-work locations (Non-home based

Non-work–NHBNW)

Two separate trip purposes represent light and heavy commercial truck movements.



1.4 Time of Day Periods Another consideration in travel demand model design is determining what periods of the day will be separately forecast. Modeling traffic by time of day has the advantage of linking traffic volume estimates more closely to determinants such as road capacity, transit service availability, likely times for trips of specific purposes to be made, as well as, such influences as time dependent road use restrictions (e.g. peak period high occupancy vehicle restrictions) and road operation (e.g. reversible lanes). Modeling time periods also provides more directly usable information for road and traffic engineering applications; transit planning; and, for estimating vehicular emissions and pollutants.

The Anchorage travel model stratifies travel into three weekday time periods. These are:

• AM Peak Period (7:00 AM-9:00 AM); • PM Peak Period (3:00 PM-6:00 PM); • Off Peak Period (12:00 AM-7:00 AM and 9:00 AM-3:00 PM and 6:00 PM-12:00AM)

Time period stratification is applied after trip distribution and before mode choice. . The traffic assignment module (which places expected traffic on the most likely routes given road congestion) computes adjusted travel times due to traffic congestion during each time period. These adjusted travel time results are fed back to trip distribution to recalculate origin-destination patterns. Trips for all time periods are distributed to each time period. Then, after each cycle of trip distribution, each time-period-specific trip distribution output is factored by it’s time-of-day proportion and directionality percentages to yield trips occurring in each time period. The actual proportion of trips in a given time period is calculated after each cycle of trip distribution.

1.5 Travel Mode Stratification The mode choice models utilize market stratification and trip purpose stratification to predict traveler choice of travel mode for every trip. Travel mode in turn is used to define what routes and services are utilized for the ubiquitous transportation network and it’s specific modal network components.

The Anchorage mode choice model considers the following modal options:

• Auto drive alone • Auto drive with passenger • Auto passenger • Public Transit/School Bus Transit • Walking • Bicycling

In the case of transit modes, all trip purposes except home-base school are estimated as public transit trips. Home based school transit trips are estimated as school bus trips.



1.6 Transportation Networks The key to understanding how estimated travel demand effects the condition of the transportation system is the definition of the networks that represent linkages between different area locations. In travel models, most often this is represented as a set of interconnected links that approximate the physical layout of various area transportation facilities including highways, streets, exclusive transit facilities, railroads, sidewalks, trails, etc. Expected speeds of travel along these links is used to estimate how long it may take to get between two places via a given mode and route. Bus travel times consider such things as waiting for a bus, transferring between buses, walking to and from the bus stop, and, or course, time spent riding the bus. The objective in defining these networks is to represent how the traveler will see travel options and choose between them.

As noted earlier, the Anchorage travel demand model’s process is applied separately for three different time periods. The model maintains information about potential paths for in each time period. Potential trip attributes such as auto travel times, frequency of bus services and alternate bus routing patterns are all stratified by time of day. Travel by walking and travel by bicycle are assumed at constant average speeds for all three time periods.

The Anchorage travel model transit service, includes a computational submodel that adjusts the speeds and travel times of buses operating in mixed flow traffic with automobiles, based on the level of congestion and operating speeds experienced by autos. This procedure maintains consistency in mode attributes and allows simulation of such effects as traveling through congested areas versus having an independent bus lane to be effectively represented.


SECTION 2

Travel Model Structure

The Anchorage travel demand model consists of a series of processes that progressively build up forecasts of travel activity. Major inputs to these processes include the following:

• Land use and socio-economic data representing area development

• Multimodal networks representing travel options and connections between different part of the area

Figure 2-1 is a flowchart illustrating the detailed sequencing and linking of the various steps used in the travel model. This section focuses on providing a broad overview of the key elements of those steps in the context of the overall process. The subsequent sections will detail assumptions, calibration and validation of each of these steps.

2.1 Land Use and Demographic Inputs Besides the networks representing multimodal transportation facilities available to prospective travelers, another primary input to the travel modeling process is an inventory or forecast of the distribution and make up of land uses, households, population and employment in the study area. In assembling inventories or developing forecasts of these inputs, attention must be paid to providing adequate detail to support the input requirements of the travel demand model. This means both (1) having estimates at the traffic analysis zone level to provide adequate geographic specificity and (2) categorizing data such as types of employment in enough detail to support input requirements. The Anchorage model design requires the following data inputs and associated breakdown for each traffic analysis zone. A detailed description of the data fields is provided in Appendix A–Variable Dictionary Anchorage Travel Model Soicioecnomic Database.

• Traffic Analysis Zone ID • Total Population • Number of Households • Median Household Income (2002 Dollars) • Average Number of Workers/Household • Enrollment of Schools in Zone • Employment Categories

− Agricultural Employment − Mining Employment − Construction Employment − Manufacturing Employment − Transportation/Public Utilities Employment − Wholesale Trade Employment − Retail Trade Employment − Finance, Insurance and Real Estate Employment

TRAVEL MODEL STRUCTURE


− Service Employment − Government Employment − University Employment − School Employment − Medical/Health Services Employment

• Work Trip Parking Costs (2002 Dollars) • Other Trip parking Cost (2002 Dollars)

Categorization of employment data was based on employer reported North American Industrial Classification System (NAICS) codes to allocate the 2002 employment inventory. Future forecasts use the employment type stratification of the AMATS Land Use Allocation model directly.

2.2 Land Use/Density Inputs The new travel model provides sensitivity to environmental factors that influence tripmaking in general and, specifically, determination of mode choice. To do this, indicators of characteristics such as pedestrian friendliness, mix of land uses and similar measures are calculated based on the spatial structure of the travel model network and land use assumptions. Specifically, the values calculated in this manner are used in the auto ownership model and in all the mode choices models (specific variables used and associated coefficients are documented in Mode Choice (Section 7). These values are calculated for each traffic analysis zone.

• Street intersection density (per square mile); • Ratio of employment to residents (per zone); • Employment density (per square half mile and mile); • Residential density (per square half mile and mile);

2.3 Household Segmentation Households are segmented into five different primary groupings representing their membership in the household size, number of worker, median income, auto ownership and number of children classes described in Travel Model Overview (Section 1). Depending on the model calculations being performed, each of these groups may or may not be treated separately.

The trip generation models use the following household segmentations:

• Household size and number of workers are used together to estimate home based work productions

• Household size and median income class are used together to estimate both home based shopping and home based other productions

• Household size and number of children are used together to estimate home based school trip productions



The mode choice model uses the following household segmentations:

• Separate models for each household size and number of workers category are used together to estimate home based work mode shares

• Separate models for each household size and median income range category are used together to estimate home based shopping and home based other mode shares

In all cases, except the auto ownership and children per household models, the segmentation models use the zonal average value as the input and based on 2002 Census data (both SF1 tables from the general census and CTPP(Census Transportation Planning Package) tabulations), stratify that average into the segments required by the downstream models. For the auto ownership model, a multinomial logit formulation was used. For the children per household model, a cross classification formulation was used. All models are fully documented in Traveler Market (Household) Segmentation (Section 3) of this document.

2.4 Trip Distribution Separate trip distribution models were calibrated for each of the six person travel trip purposes and for the two commercial vehicle trip categories. All trip distribution formulations except home based school are based on the gravity model. Utilizing data from the Anchorage Household Travel Survey, accessibility/deterrence functions were developed for all person based trip purposes except home based school. For medium and heavy duty truck trips, the functions used in the previous Anchorage model (based on FHWA Quick Response Freight Traffic Estimation methods) were utilized.

For home based school, a Fratar growth factoring model was used. This model uses a combination of a “seed” table of known existing origins and destinations and a parallel set of updated trip productions and attractions to “grow” interchanges in the seed table to match the updated production and attraction values. The process iterates to expand interchange values while using the independently determined productions and attractions for each traffic analysis zone as control totals.

The output of the trip distribution model is stratified by both trip purpose and time of day. Time of day stratification is achieved by inputting the time period specific network travel times and creating three person trip tables for each of the eight trip purposes (six for person travel and two for commercial vehicle/truck travel). The process of estimating the number of trips for each trip purpose in each time period, which occurs after trip distribution is completed applies time-of-day factors and directional flow factors to the daily trip distributions to output three time-period partitioned trip tables for each trip purpose.

The trip purpose specific trip distribution functions are documented in the Trip Distribution (Section 5) of this document.

2.5 Time of Day Trip estimates from the trip distribution step are used to calculate the number of trips occurring in each of three modeled time periods for all eight purposes. If congested times are used to estimate each of the time periods, origin-destination patterns for parallel (e.g.



AM peak period home based work vs. PM peak period home based work) can vary significantly. The purpose of this step in travel model processing is to produce a set of triptables which correctly represent the relative directional orientation and the proportion of trips occurring in each modeled period. This is done by multiplying each trip table by household travel survey derived factors indicating the percentage of trips from home vs. trips to home for home based trip purposes (the factor for the non-home based work purpose is based on similar survey information for that activity. Non-home based non-work is always 50/50); then, factoring the resulting table by the total percentage of trips for a given trip purpose occurring in the period. These factors and their application are fully documented in the Time of Day (Section 6) of this document.

2.6 Mode Choice The mode choice models estimate shares of all trips by mode. They comprise the most complex step in travel model processing. For the Anchorage model, mode choice is performed separately for traveler market segments, trip purpose and modeled time period. All non-commercial trip purposes are specified as multinomial logit models which estimate mode shares for all modes identified in Travel Mode Stratification in the Travel Model Overview section (Section 1). Coefficients and performance of these models is fully documented in Mode Choice (Section 7).

Operationally, mode share estimates are obtained for each market segment and time period for each home based trip purpose. The segments within each time period are combined to be input to the corresponding trip assignment (either vehicle trips to highway or transit riders to transit routes). For non-home based trip purposes, mode shares are calculated in aggregate for a given time period as input to the traffic assignment process. For commercial trips (medium and heavy duty trucks), these are already estimates of vehicle trips, so no mode choice takes place.

2.7 Traffic Assignment The last step in the travel modeling process is assigning the mode specific and time period specific estimates of origin-destination interchanges to specific routes in the travel model network. The model software provides capabilities to do this for both highway and transit trips. (For transit, only public transit trips are assigned. School bus rider trips are not currently assigned to the network.)

Public transit trips are assigned to the transit system network routes which are represented by stops, travel times, and waiting delays to traverse the minimum “cost” route between a given origin and destination via public bus or other transit mode. This assignment is sensitive to conditions in the public transit network during the assignment’s time period (associated road congestion, bus frequencies by route, etc.). The model output is rider volumes by route and specific segment. This data can be examined in aggregate to determine transit riders in specific corridors or areas of the city.

Highway vehicle trips are assigned using a mathematical algorithm that is sensitive to the relationship between traffic volume and available road capacity. The algorithm uses this relationship to: 1) estimate expected congestion related delays; 2) considering that delay



recalculate the “best” routes between origins and destinations; 3) reassign traffic to available routes given updates to route times based on delay. This cycle theoretically (equilibrium is actually estimated using a sub-optimal heuristic that approximates an equilibrium state) continues until the condition is reached that no group of travelers can improve their origin-destination travel time by choosing an alternate route. At his point, the system is said to be in equilibrium. Either the approximation of an equilibrium condition or a maximum number of iterations to reach this condition ends the assignment process. As referenced above, highway assignments in the Anchorage model are made for three specific time periods.

Based on final assignment speeds, the estimated link travel times in the network are updated. These new travel times can then be reinput to the trip distribution process to show impacts of expected congestion on traveler destination and mode choices as part of the travel model feedback process. The cycle of trip distribution, time of day factoring, mode choice and traffic assignment can then be repeated until either the travel times input to the traffic assignment process and those output from the assignment process reach equilibrium of a maximum number of iterations is completed. This process uses a mathematical model termed Method of Successive Averages to achieve equilibrium and ensure closure between input speeds and volumes and those output as the final iteration. This model equilibrium feedback process is described more fully in Appendix B–Model Feedback Process.

The formulation used to determine overall model equilibrium is discussed in detail in Traffic Assignment and Volumes (Section 8). That section also includes discussion of the process of comparing measured to modeled traffic volumes.

2.8 Summary The purpose of this section has been to provide a roadmap to the remainder of the discussions and an overview of formulations of specific models that make up the Anchorage travel model; and, to broadly explain the process used to apply those models. The remainder of the document discusses the specific models and the measurement of their performance.


SECTION 3

Traveler Market (Household) Segmentation

This section provides detailed documentation of the models developed to segment households into different groups or market segments. The following three major data sources were utilized:

• 2000 US Census Transportation Planning Package • 2000 US Census SF1 Short Form Statistical Tabulations • 2002 Anchorage Household Travel Survey

Each of the models which were developed is discussed below.

3.1 Household Size Median household size information must be translated to numbers of households falling into each size group or segment. These segments support disaggregate estimation of auto ownership, cross classification of trip production rates (for home based shop, home based school and home based other) and disaggregate estimation of mode choice (for the same trip purposes). To estimate the proportions of different sized households in each TAZ, median values from the 2000 U.S. Census data are correlated with corresponding percentages of each of the groups. This analysis was performed for both the complete census data sample (100% of the area population) at the block group level and for traffic analysis zones using the Census Transportation Planning Package subset (approximately a 6% sample of the area population). Using these two sources, curves were developed relating the proportion of one person, two person, three person and four or more person households to a range of median values. These curves are illustrated in the graphic below.

The number of households in each TAZ is multiplied times these proportions to determine the number of houses in each category. For future conditions, zonal population and households estimates are used to derive mean values and the corresponding size segment for each category in each zone.

Figure 3-1 illustrates the family of curves indicating the relationship between average household size and the proportion of households falling into each group for that average.

TRAVELER MARKET (HOUSEHOLD) SEGMENTATION


Household Size Model

0.00

0.20

0.40

0.60

0.80

1.00

1.1

1.4

1.7

2.0

2.3

2.6

2.9

3.2

3.5

3.8

4.1

Median Houshold Size

Gro

up P

ropo

rtio

n

1 Person

2 Person

3 Person

4 Person

Figure 3-1. Average Household Size vs. % Total Households by Size Group

Aggregating the estimates of household size groups for the Anchorage model’s study area, yields the following proportions (Table 3-1) for each group and an overall average for the study area of 2.711. Comparable proportions obtained by aggregating the parallel Census (100% sample) information are also shown in Table 3-1 (average is 2.745). The household size model estimates an average household size of 4.966 for households of four of more persons compared to an actual value of 4.966 based on 2000 U.S. Census information. (Note that while the U.S. Census does include military housing, the estimates from the Anchorage household size model do not.)

TABLE 3-1 Comparison of Household Size Model Results (2002) and 2000 Census – Proportion of Household Size Group (Totals for Anchorage Area)

Household Category Model Total

Model Group/ Total Census Total

Census Group/ Total

One Person 23805 .250 22142 .234

Two Person 28772 .300 30155 .318

Three Person 17371 .183 16942 .179

Four or More 25088 .264 25583 .270

Totals 95036 1.00 94822 1.00



3.2 Household Income The second model developed was similar to the household size model. This model predicts the proportion of household falling into low, medium and high household income groups based on median income for each traffic analysis zone. Median household income must be translated to numbers of households falling into specified income groups to support disaggregate estimation of auto ownership, cross classification of trip production rates (for home based shop and home based other trip purposes) and disaggregate estimation of mode choice (for home based work, home based shop and home based other trip purposes). To estimate the proportions of income groups in each TAZ, median values from the 2000 U.S. Census are correlated with corresponding percentages of the groups. This analysis was performed for both the complete census data sample (100% of the area population) at the block group level and for traffic analysis zones using the Census Transportation Planning Package subset (approximately 6% sample of the area population). Using these two sources, curves were developed relating the proportion of low income (< $40,000/year), medium income ($40,000/year to $69,999/year), and high income ($70,000/year or more) households to a range of median values. These curves are illustrated in the graphic below.

Total households are multiplied times these proportions to determine the number of houses in each category. For future conditions, estimation of zonal median incomes will impact model results by changing the size of the market segment associated with each category in a zone.

Figure 3-2 illustrates the family of curves indicating the relationship between the median income for a given traffic analysis zone relative to the median income in the study area (using this ratio allows all estimates to be made relative to the range of household incomes in Anchorage) and the proportion of households falling into each group for that average.

Household Income Model

0.000

0.200

0.400

0.600

0.800

1.000

0.1

0.4

0.7

1.0

1.3

1.6

1.9

2.2

Zonal/Regional Income

Gro

up P

ropo

rtio

n

Low Income

Medium Income

High Income

Figure 3-2. Zonal/Regional Income vs. % Total Households by Income Group



Aggregating the estimates of household income groups for the Anchorage model’s study area, yields the following proportions (Table 3-2) for each group. Comparable proportions obtained by aggregating the parallel 2000 U.S. Census (100% sample) information are also shown in Table 3-2. It should be noted that while the census does include military housing, the estimates from the household income model do not.

TABLE 3-2 Comparison of Household Income Model Results (2002) and 2000 Census – Proportion of Household Income Groups (Totals for Anchorage Area)

Household Category Model Total Model Group/

Total Census Total Census Group/

Total

Low Income (>$40000/Yr) 35074 .369 32027 .337

Medium Income ($40001-$70000/Yr) 29284 .308 31587 .332

High Income (>$70000/Yr) 30677 .323 31466 .331

Totals 95036 1.00 95080 1.00

As would be expected, comparisons of the household income model to 2000 U.S. Census data shows greater discrepancies then the household size model. This is likely due to the differences in the years being compared and also in the separate survey reporting categories used to report median zonal income.

3.3 Workers per Household The average number of workers per household is used to generate households falling into worker classes to support disaggregate estimation of auto ownership, estimate of home based work trip productions and estimation of home based work mode. To estimate the proportions of worker classes in each TAZ, average workers per household values from the 2000 U.S. Census are correlated with corresponding percentages of the groups. This analysis was performed for both the complete census data sample (100% of the area population) at the block group level and for traffic analysis zones using the Census Transportation Planning Package subset (approximately 6% sample of the area population). Because the CTPP dataset provides better categorization of the workers per household variable and due to logical inconsistencies in the Census (100%) sample information, curves were developed based exclusively on the CTPP data relating the proportion of zero worker, one worker, two worker, and three or more worker households to a range of average values. These curves are illustrated in Figure 3-3 below.

Total households are multiplied times these proportions to determine the number of households in each category. For future conditions, estimation of the zonal average of number of workers per household will impact model results by changing the size of the market segment associated with each category in a zone.



Household Worker Model

0.0000.2000.4000.600

0.8001.0001.200

0.0

0.4

0.8

1.2

1.6

2.0

2.4

2.8

3.2

Average Workers/Household

Gro

up P

ropo

rtio

n

0 Worker

1 Worker

2 Worker

3+ Worker

Figure 3-3. Average Workers/Household vs. % Total Households by Group

Aggregating the estimates of household worker groups for the Anchorage model’s study area, yields the following proportions (Table 3-3) for each group. Comparable proportions obtained by aggregating the parallel 2000 U.S. Census information are also shown in Table 3-3. It should be noted that while the census does include military housing, the estimates from the household worker model does not.

TABLE 3-3 Comparison of Household Worker Model Results (2002) and 2000 Census – Proportion of Household Worker Groups (Totals for Anchorage Area)

Household Category Model Total Model Group/

Total Census Total Census Group/

Total

Zero Workers 14204 .149 15165 .160

One Worker 38282 .403 38805 .408

Two Workers 36550 .385 35283 .371

Three or More 6000 .063 5655 .059

Totals 95036 1.00 95046 1.00



3.4 Auto Ownership The Anchorage travel model includes a model for estimating auto ownership groups. These ownership groups and the resulting estimate of autos per person per TAZ are used as an input to mode choice modeling. Vehicle (auto) ownership group shares were determined using a multinomial logit model formulation. This estimation process was chosen to allow more direct forecasting of auto ownership for future years based on primary forecasting variables. The auto ownership model forecasts group proportions for 0, 1, 2 and 3 or more auto owning households. The variables used for input include household size, number of workers in household, residential density and land use mix of the residence zone. The model is applied separately for each household income group.

The logit formulation calculates segment shares based on the proportion of the total of all shares each share represents. These shares are expressed the natural exponent of an empirically derived function. The general formulation is as follows:

Share(ui)=eui/ eut

where:

Share(ui) = calculated share of some group i

eui = the natural exponent of the derived utility ui

eut = sum of all eu for all possible options

The auto ownership model uses the following variables in its utility expression:

• Household size • Number of workers per household • Income segment (low, medium or high) • Density of street intersections within ½ mile • Index of zonal mix of residential and employment uses

The following table (Table 3-4) lists the associated coefficients used in the auto ownership model:

TABLE 3-4 Auto Ownership Model Variables and Coefficients

Variable Zero Autos One Auto Two Autos

Constant -21.0651 4.1442 2.5154

Household Size -1.23182 -1.37421 -0.36378

No of Workers -1.39570 -0.93408 -0.48255

Income Group 22.34620 (high) 18.92600

(med) -11.39630 (low) 2.05728 (high) 0.37583 (med) -1.80951 (low)

0.64969 (high) 0.09619 (med) -0.40170 (low)

Intersection Density 0.03418 0.01294 0.00016

Mixed Use Index 0.00199 0.00007 0.00002

For households having 3 or more vehicles, the utility value is 0. This value provides the base reference for scaling the utilities of all other groups.



Aggregating the estimates of the sizes of household auto ownership groups for the Anchorage model’s study area, yields the following proportions (Table 3-5) for each group. Comparable proportions obtained by aggregating the parallel 2000 U.S. Census information are also shown in Table 3-5. It should be noted that while the census does include military housing, the estimates from the household auto ownership model do not.

TABLE 3-5 Comparison of Household Auto Ownership Model Results (2002) and 2000 Census – Proportion of Household Auto Ownership Groups (Totals for Anchorage Area)

Household Category Model

Model Group/ Total Census Census Group/ Total

Zero Autos 7317 .077 6013 .063

One Auto 33643 .354 33107 .349

Two Autos 39725 .418 39689 .418

Three or More 14351 .151 16035 .169

Totals 95036 1.00 95114 1.00

3.5 Children per Household To support estimation of Home Based School trip productions, a cross classification model was developed to correlate the number of school aged children in a household to household size and number of workers in the household.


SECTION 4

Trip Generation

4.1 Home Based Trip Production Models Trip generation models and rates are based on information from the 2002 Anchorage Household Travel Survey. Cross-classification methods were used to estimate the number of origins (or productions) associated with each of the four home based trip purposes. These models were developed by cross classifying the trips per household by trip purpose with associated market segments. Review of the associated rates and their ability to replicate totals from the travel survey were used as criteria for both selecting the cross classification dimensions associated with each trip purpose and for determination of the number of cross classification categories. Another criterion in this process was the feasibility of developing reasonable estimation models to predict membership in the cross classification categories.

Categorical rates for home base work, home based shop, home based school and home based other trip purposes are presented in the four tables (Table 4-1 through 4-4) below.

TABLE 4-1 Home Based Work Trip Rates (Workers/HH by Income Group)

No of Workers/HH Low Income Medium Income High Income

0 0.170 0.170 0.170

1 1.170 1.220 1.390

2 2.350 2.440 2.780

3 or more 2.870 2.980 3.430

TABLE 4-2 Home Based Shopping Trip Rates (Household Size by Income Group)

Persons/HH Low Income Medium Income High Income

1 0.42 0.47 0.47

2 0.55 0.64 0.64

3 0.57 0.71 0.88

4 or more 0.80 1.08 1.30

TRIP GENERATION


TABLE 4-3 Home Based School Trip Rates (Household Size by Children/HH)

Persons/HH 1 Child 2 Children 3 or more Children

1 0.00 0.00 0.00

2 0.96 0.00 0.00

3 1.18 2.74 0.00

4 or more 2.40 2.35 4.36

TABLE 4-4 Home Based Other Trip Rates (Household Size by Income Group)

Persons/HH Low Income Medium Income High Income

1 1.21 1.27 1.27

2 1.94 2.47 2.47

3 2.04 2.59 3.12

4 or more 4.06 5.44 5.71

These trip generation rates were applied and evaluated by comparing results obtained by their application the expanded totals from the household travel survey. These tests and an analysis of the outcomes is present in the Validation Results section at the end of Trip Generation (Section 4).

4.2 Non-Home Based Trip Production Models In addition to the four home based trip purposes, two additional trip purposes are used to describe the full range of non-commercial trips. The two purposes used for non-commercial trips are non-home based work and non-home based non-work. This distinction was applied to separate out those trips made from a work location to supplemental destinations such as midday trips to visit other businesses for work or personal reasons from other non-work related non-home based trips. This separation of work-based and non-work based trips has been shown to improve the quality of small area forecasts of number, length and time of day of non-home based trips.

The number of these non-home trips is more likely to be influenced by the characteristics of the location of their origin rather than the socio-economic characteristics of the traveler. Because of this, production models correlate these trips to inputs that describe the origin location such as the number of employees at a location. Typically, non-home based trip generation models are specified using such formulations as linear regression. In case of the

TRIP GENERATION


Anchorage model, the production models for both non home-based work and non-work are specified as linear regression equations. The equations are presented below.

For non-home based work, the final estimated equation was:

NHBW Productions=0.809*Total Employment

For non-home based non-work, the final estimated equation was:

NHBNW Productions= 0.222*Population+

4.000*Adjusted1 Retail Employment+ 0.487*Other Employment+ 0.299*Number of Households

4.3 Commercial Vehicle (Truck) Trips The commercial and truck trip generation model previously developed by AMATS was reused. This model is based on a set of standard equations for trip generation developed by the Federal Highway Administration as part of the Quick Response Freight Estimation Method. The FHWA methodology specifies a set of procedures to, first, apply and, then, adjust the generic models based on available local data. These procedural steps were followed and applied as part of AMATS’ previous calibration process. Full documentation of this model and calibration process is provided as Appendix C–Commercial Vehicle Travel Model of this report. (Note that further refinements in truck trip generation rates were formulated for Anchorage special generators).

4.4 Trip Attraction Models To complete the trip generation model set, a set of models were developed to identify where the trip productions were bound–that is, the trip attraction end of each trip. As in the case of non-home based trip generation, these trip attraction models are based on the zonal characteristics of potential destination zones and are typically formulated as linear regressions. Information from the 2002 Anchorage Household Travel Survey regarding the destinations of travelers by trip purposes was tallied for each TAZ, together with the information from the land use, employment, and demographic inventories. Regression analysis was applied to this composite TAZ dataset to develop trip attraction equations for each trip purpose. These equations are presented below.

Home based work attractions:

HBW Attractions=0.994*Total Employment

Home based shopping attractions:

HBS Attractions=2.314* Adjusted1 Retail Employment+

3.080*Construction Employment

Home based school attractions:

HBSCH Attractions=2.091*Number of Students

TRIP GENERATION


Home based other attractions:

HBO Attractions=0.484*Population+

3.344* Adjusted1 Retail Employment+ 0.773*Health Services Employment+ 1.576*Educational Employment+ 0.481*Service (inc FIRE2) Employment+ 3.256*Construction Employment

Non-home based work attractions:

NHBW Attractions=2.327*Government Employment+

3.287* Educational Employment+ 0.746* Service (inc FIRE2) Employment+ 1.285* Industrial3 Employment

Non-home based non-work attractions:

NHBNW Attractions=0.154*Population+

3.830* Adjusted1 Retail Employment+ 3.619*Educational Employment+ 0.592*Service (inc FIRE2) Employment+ 1.704*Construction Employment

4.5 Special Generators/Visitors In addition to the trip generation based directly on zonal households and employment, the AMATS travel model incorporates the following two additional trip generation elements:

1. Estimates of trips to and from a range of special uses (e.g. airports, universities and military bases) that cannot be adequately accounted for using the standard methods described above.

2. Estimates of trips made by non-resident visitors.

These two models are described below.

The special generators model specifies tripmaking rates and the distribution of trips into the defined trip purposes. In most cases, all special generator trips are attractions (i.e. non-home destinations of home based or other trips being made). However, some uses also generate productions such as residential students at a college or university. The AMATS travel model design also incorporates truck trip generation for some of the special uses. Table 4-5 lists specific uses that were incorporated as special generators in the travel model. A complete listing of the associated trip rates, trip rate basis and purpose/production-attraction breakdowns is provided in Appendix D–Special Generators.

TRIP GENERATION


TABLE 4-5 Special Generators

Elmendorf AFB-Post Rd Gate

Ft Richardson

Ft Richardson (Trucks only)

Ship Creek (Port Area)

Ship Creek (Port Trucks only)

Columbia Hospital

Columbia Hospital (Trucks only)

University of Alaska

University of Alaska (Trucks only)

Alaska Pacific University

Alaska Pacific University (Trucks only)

Providence Medical Center

Providence Medical Center (Trucks only)

Loussac Public Library

Alaska Native Hospital

Alaska Native Hospital (Trucks only)

Stevens International Airport

Stevens International Airport (Trucks only)

Air National Guard Base

Eagle River Landfill

Out of Area Trips (North)

Out of Area Trips (South)

Elmendorf AFB-Gov't Hill Gate

Elmendorf AFB-Boniface Gate

The visitor model predicts the number of trips made by visitors staying at area hotels and motels. There are three inputs to the model, as follows:

• An inventory of the hotels/motels; their locations in the model area; and the number of guest rooms

• An estimate of expected hotel occupancy based on type of property and location (currently, the travel model uses three different levels of occupancy)

• An estimated rate of person trips per occupied room (these trips are treated as home based other productions at the hotel/motel sites)

The 2002 hotel/motel inventory and assumed occupancy for each property is provided in Appendix E–Hotel/Motel Visitor Model. Based on Institute of Transportation Engineers studies and other trip generation research, the trip rate used is 8.3 trips per occupied room.

TRIP GENERATION


4.6 Other Special Considerations Early results from the trip generation process indicated that significant underprediction of total trips was occurring in the vicinity of “big box” (e.g. Wal-Mart) stores. To correct this, these zones were identified as SPECIAL RETAIL ZONES (SRZ’s) and trip rates were adjusted accordingly from national research data. Recent work in evaluating differences between trip generation for big box retail versus other retail, retail employment-related attractions for the home based shopping and home based other trip purposes was examined. This resulted in increasing home based shopping and home based other trip purposes attractions rates for SRZ’s by multiplying times 2.644.

4.7 Validation Results After completing specification of all trip generation models, the models were scripted and then applied to the 2002 socio-economic database to test their ability to replicate existing travel. Results of the trip generation process were compared by trip purpose for the Anchorage area. The comparison is not exact because of influences of the special generator, external and visitor trips. Some of these trips are only partially or are not represented in the survey.

Table 4-6 shows the comparison of survey vs. model generated total trips by trip purpose.

TABLE 4-6 Total Trips by Trip Purpose (Survey-Model Comparison)

Trip Purpose Survey Model Survey/Model

Home based Work 172600 175500 0.983

Home based Shop 71400 70000 1.020

Home based School 103800 101300 1.024

Home based Other 287500 320000 0.898

Non-home based Work 107100 111300 0.962

Non-home based Non-work 184600 210300 0.878

All Areas 927000 988400 0.938

The conclusion that can be drawn from this is that approximately 6% of daily person tripmaking is externally or visitor based.

A second comparison was made to examine the geographical distribution of trip generation. Total surveyed trips (expanded) by community council area were plotted against total modeled trips for the same areas and the correlation coefficient between the parallel data points was calculated. This plot is shown as Figure 4-1 below.

TRIP GENERATION


Survey vs Model Trips

R2 = 0.9455

0

50000

100000

150000

200000

250000

300000

0 50000 100000 150000 200000 250000 300000

Survey Trips by Community Council

Mod

el T

rips

by C

omm

unity

Cou

ncil

FIGURE 4-1. SURVEY VS. MODELED TRIPS BY COMMUNITY COUNCIL AREA

The results show that the trip generation and attractions models developed for Anchorage correlate well on a aggregate basis to observed data from the household survey. For reference, a R2 (correlation coefficient) of one point zero (1.0) indicates a perfect match. Zero indicates no discernable relationship between two groups of data.

TRIP GENERATION



SECTION 5

Trip Distribution

Trip distribution models use the estimates of trips produced and trips attracted in an area as inputs to creating patterns of origins and their associated destinations. The Anchorage travel model utilizes two types of trip distribution models. First, a gravity model which utilizes a trip length frequency distribution from local survey data to attempt to replicate current trip patterns based on matching the associated trip length distribution for each trip purpose. Trip lengths are specified in minutes to account for delays encountered, differential speeds of different travel modes, etc. To ensure that survey vs. model trip characteristics are comparable, the starting point for this process is to use the calculated paths in the travel model network to estimate trip lengths for both survey and model trips.

The second method is a growth or “Fratar” model. Unlike a gravity model, the Fratar model does not consider how network operating characteristics may impact travel behavior. The Fratar model starts from an existing condition table of trip interchange data and “grows” those interchanges based on newly input zonal totals for productions and attractions. Fratar models are typically used where either network conditions over time are expected to remain reasonably stable or where destination choice is not a function of the characteristics of the origin-destination path chosen. In the Anchorage case, Fratar distribution was used for the home based school trip purpose only. The gravity model distribution is used for all other trip purposes.

The gravity model is calibrated mathematically by, given the input data, attempting to match each one minute interval (1 to 60 minutes) with the associated proportion of survey trips of that length. In the case of the Anchorage model, this produced discontinuous relationships due to the “lumpiness” of the survey data. The original curves were smoothed to provide a continuous function. Although the trip distribution models are executed on a time period specific basis, only one daily average curve is used for each trip purpose. For this reason, only daily trip frequency distributions are considered here.

It should be noted that inputs to the trip distribution process are stratified by time period. When the model is run using feedback to the distribution process, inputs to distribution include link travel times stratified by time of day period. Distribution for each time period is calculated separately. Because the Anchorage model uses distribution feedback, the trip distribution curves had to be readjusted to consider the feedback effects. Table 5-1 shows the comparison between the survey and travel model average trip lengths by trip purpose and overall.

TRIP DISTRIBUTION


TABLE 5-1 Average Trip Length (in Minutes) by Trip Purpose

Trip Purpose Survey Model Survey/Model

Home based Work 12.038 13.354 0.902

Home based Shop 9.566 9.600 0.996

Home based School 9.016 5.578 1.616

Home based Other 9.657 10.638 0.908

Non-home based Work 8.448 7.426 1.138

Non-home based Non-work 9.037 8.554 1.056

All Areas 9.863 10.219 0.965

Longer trip lengths for home based work trips and home based other trips in the model are largely due to the inclusion of through (trips which begin or end outside of the study area) trips in the statistics. Using the corresponding trip lengths but excluding through trips, the home based work and home based other average lengths are about 12.2 and 8.3 minutes respectively.

The significantly shorter trip length for home based school trips is based on “growing” the ASD attendance database; not on a trip length based model. The likely cause of the difference is underreporting in the survey of the shorter, mostly non-motorized, trips. These would be trips made by children walking or riding their bikes to school whose travel was not captured in the survey. Given the incidental nature of many non-home based work trips, it’s likely that differences in the survey and model average trip lengths can be attributed to a similar cause.

Figures 5-1 through 5-6 detail the relationship between the surveyed trip distribution and modeled trip distribution for each trip purpose.

TRIP DISTRIBUTION


HBW Daily Trip Distribution

0.00

2000.00

4000.00

6000.00

8000.00

10000.00

12000.00

14000.00

16000.00

1 6 11 16 21 26 31 36 41 46 51 56

Minutes

No

of T

rips

Model Daily

Survey Daily

Figure 5-1. Survey vs. Model Trip Length Frequencies (Home based work)

HBS Daily Trip Distribution

0.00

1000.00

2000.00

3000.00

4000.00

5000.00

6000.00

7000.00

8000.00

9000.00

1 6 11 16 21 26 31 36 41 46 51 56

Minutes

No

of T

rips

Model Daily

Survey Daily

Figure 5-2. Survey vs. Model Trip Length Frequencies (Home based shop)

TRIP DISTRIBUTION


HBSch Daily Trip Distribution

0.00

5000.00

10000.00

15000.00

20000.00

25000.00

1 6 11 16 21 26 31 36 41 46 51 56

Minutes

No

of T

rips

Model Daily

Survey Daily

Figure 5-3. Survey vs. Model Trip Length Frequencies (Home based school)

HBO Daily Trip Distribution

0.00

5000.00

10000.00

15000.00

20000.00

25000.00

30000.00

35000.00

1 6 11 16 21 26 31 36 41 46 51 56

Minutes

No

of T

rips

Model Daily

Survey Daily

Figure 5-4. Survey vs. Model Trip Length Frequencies (Home based other)

TRIP DISTRIBUTION


NHBW Daily Trip Distribution

0.00

2000.00

4000.00

6000.00

8000.00

10000.00

12000.00

1 6 11 16 21 26 31 36 41 46 51 56

Minutes

No

of T

rips

Model Daily

Survey Daily

Figure 5-5. Survey vs. Model Trip Length Frequencies (Non-home based work)

NHBNW Daily Trip Distribution

0.00

5000.00

10000.00

15000.00

20000.00

25000.00

1 6 11 16 21 26 31 36 41 46 51 56

Minutes

No

of T

rips

Model Daily

Survey Daily

Figure 5-6. Survey vs. Model Trip Length Frequencies (Non-home based non-work)

Generally, the graphs show a good correspondence between the survey trip distributions and the model trip distributions. The difference in absolute magnitudes of the non-home

TRIP DISTRIBUTION


based non-work distribution is a reflection of the model trips which start or end outside of the study area; these trips are not included in the household survey trip set.


SECTION 6

Time of Day Factoring

Time of day factoring is the straightforward process of converting the trip distribution outputs to a set of triptables that correctly represent the percent of trips in the AM, PM, and off peak periods and their directional orientation. To perform this step, these factors were developed directly from the 2002 household travel survey. The three sets of time period factors are provided in the tables below.

TABLE 6-1 AM Peak Period Trip Purpose and Directional Factors

Trip Purpose % Trip in Period % Home Origins % Home Destinations

Home based work 25.7 95.5 4.5

Home based shop 2.5 79.4 20.6

Home based school 43.2 97.2 2.8

Home based other 14.4 79.8 20.2

Non-home based worka 14.3 81.5 18.5

Non-Home based non-work 7.6 50.0 50.0

Trucks 6.0 50.0 50.0

aFor Non-home based work trips, “% Home Origins” is the percent of trips with a work origin, “% Home Destinations” is percent of trips with a work destination.

TABLE 6-2 PM Peak Period Trip Purpose and Directional Factors








Trucks 25.0 50.0 50.0


TIME OF DAY FACTORING


TABLE 6-3 Off Peak Period Trip Purpose and Directional Factors








Trucks 69.0 50.0 50.0


After the person trip-tables output by the trip distribution model are factored into tables representing directional trips by time period, this data is input into the mode choice models to determine shares of tripmaking by specific travel modes.


SECTION 7

Mode Choice

Mathematically, the mode choice models are the most complex models used in the Anchorage travel model process. Also, in order to measure competition between travel modes, they also have the largest input data requirements.

Mode choice modeling in the Anchorage model is stratified in three dimensions:

• Trip purposes are modeled separately

• Within each trip purpose, time periods are modeled separately

• Within each time period and trip purpose, specific traveler markets are modeled separately for the home based trip purposes

Mode choice models were separately defined and calibrated for each of the six person based trip purposes (not trucks). The impact of time periods is represented through supplying information regarding network conditions (e.g. road speeds, bus service frequencies) that is specific to the time period in question. Traveler market segments are represented through calibrated parameters used in the trip-purpose-specific mode choice model. Operationally, separate mode choice model runs are made for each segment (within a given trip purpose and time period) and then these results are combined to represent the time period totals. This is shown in the Model Overview flowchart (Figure 2-1). In the Anchorage model, mode choice considers the following modal options:

• Auto drive alone • Auto drive with passenger • Auto passenger • Public Transit/School Bus Transit • Walking • Bicycling

In the case of transit modes, all trip purposes except home base school estimate public transit trips. Home based school estimates school bus trips.

Table 7-1 lists the variables used in the mode choice models and their description. (Note: the variables, which define market segments are defined and listed separately in the following section.

MODE CHOICE


TABLE 7-1 Complete List of Mode Choice Variables (All Purposes)

Variable Models Using Description

In-vehicle Travel Time (IVTT)

All Time spent traveling in or on a vehicle. Typically valued lower than time spent waiting for or accessing a mode.

Out-of--vehicle Travel Time (OVTT)

All Time spent waiting for or accessing a mode. (Walk travel time for the walk mode is not considered to be OVTT).

Autos Available per Person (AUTOPP)

HBW, HBS, HBO Measured at the TAZ level as zonal auto ownership divided by zonal population

No of Workers per Household (NWHH)

HBW, HBSch Measured at the TAZ level as zonal workers divided by zonal households

Total Employment within 1 mile (D_TEMP1)

HBW,HBO,NHBW Measured with GIS as the sum of zonal employment within 1 mile of the center of the destination zone adjusted by the percentage of the adjacent zones within the 1 mile radius

Total Employment within ½ mile (D_TEMPH)

HBS, NHBNW Measured with GIS as the sum of zonal employment within ½ mile of the center of the destination zone adjusted by the percentage of the adjacent zones within the ½ mile radius

Non-retail Employment within 1/2 mile (D_NREMPH)

HBS Measured with GIS as the sum of zonal non-retail employment within ½mile of the center of the destination zone adjusted by the percentage of the adjacent zones within the ½ mile radius

Average Size per Household (HHSIZE)

HBSch Measured at the TAZ level as zonal population divided by zonal households

Average No of Children per Household (AVCHH)

HBSch Measured at the TAZ level as zonal number of children divided by zonal households

Origin Walk Node Density (O_WALKDENS)

HBSch, NHBW, NHBNW

Measured with GIS as the sum of connections (nodes) within ½ mile of the center of the origin zone from detailed network map (not the model network)

Destination Walk Node Density (D_WALKDENS)

HBSch Measured with GIS as the sum of connections (nodes) within ½ mile of the center of the destination zone from detailed network map (not the model network)

Origin Zone Retail to Total Employment Ratio (O_RERATIOH)

NHBW Measured with GIS as the sum of zonal retail employment within ½ mile of the center of the origin zone divided by corresponding total employment and adjusted by the percentage of the adjacent zones within the ½ mile radius

Destination Zone Retail to Total Employment Ratio (D_RERATIOH)

NHBW Measured with GIS as the sum of zonal retail employment within ½ mile of the center of the destination zone divided by corresponding total employment and adjusted by the percentage of the adjacent zones within the ½ mile radius

Origin Zone Parking Cost (O_PARKCOST)

NHBW, NHBNW Assumed monthly parking charge for origin zone (in 2002 dollars)

Destination Zone Parking Cost (D_PARKCOST)

NHBW, NHBNW Assumed monthly parking charge for destination zone (in 2002 dollars)

Downtown Zone (CBD)

NHBNW Identifies trip either starts or ends in downtown zone

MODE CHOICE


7.1 Home Based Work Mode Choice The home based work mode choice model is run separately for four defined market segments. These include the following:

• Low income-one worker households • Low income-two or more worker households • Medium and above income- one worker households • Medium and above income-two or more worker households

Table 7-2 shows the list of variables and the associated coefficients used in the home based work mode choice models. Coefficients provided for the specific market groups listed above are only used in the models applied for those specific market segments (these variables and coefficients appear last and are underlined).

TABLE 7-2 Home Based Work Mode Choice Model Variables and Coefficients

Variable Drive Alone Drive w/

Passenger Auto

Passenger Public Transit Walking Bicycling

Model Constant -1.9766 -1.6963 0.6550 -0.8822 -5.8168

IVTT -0.0804 -0.0804 -0.0804 -0.0804 -0.0804 -0.0804

OVTT -0.1206 -0.1206 -0.1206

AUTOPP -0.5522 -1.1232 -2.1585

NWHH 0.2161

Ln(D_TEMP1) 0.3604

1 Worker Household

-0.5197

Low Income Household

2.9013 0.9052

7.2 Home Based Shop Mode Choice The home based shop mode choice model is run separately for four defined market segments. These are:

• Low income-three or more person households • Low income-one and two person households • Medium and above income- three or more person households • Medium and above income- one and two person households

Table 7-3 shows the list of variables and the associated coefficients used in the home based shop mode choice models. Coefficients provided for the specific market groups listed above

MODE CHOICE


are only used in the models applied for those specific market segments (these variables and coefficients appear last and are underlined).

TABLE 7-3 Home Based Shop Mode Choice Model Variables and Coefficients

Variable Drive Alone

Drive w/ Passenger

Auto Passenger

Public Transit Walking Bicycling

Model Constant -0.7321 -0.1081 -2.1997 -2.7804 -3.3619

IVTT -0.0693 -0.0693 -0.0693 -0.0693 -0.0693 -0.0693

OVTT -0.1040 -0.1040 -0.1040

AUTOPP -0.4342 -4.6678 -1.9654

D_NREMPH 0.0008

Ln(D_TEMPH) 0.5641

3 or More Person HH

0.7126 1.0733

Low Income Household

1.6626

7.3 Home Based School Mode Choice The home based school mode choice model is produces estimates of mode share for the dive alone, drive with passenger, auto passenger, walking, bicycling and school bus. This model does not estimate public transit mode share and no market segment stratification is used.

Table 7-4 shows the list of variables and the associated coefficients used in the home based school mode choice models.

TABLE 7-4 Home Based School Mode Choice Model Variables and Coefficients


Drive w/ Passenger

Auto Passenger School Bus Walking Bicycling

Model Constant 6.8889 4.0282 2.1906 2.4959 5.5937

IVTT -0.0001 -0.0001 -0.0001 -0.0001 -0.0001 -0.0001

Trip Mileage -0.0575

HHSIZE 0.8414 1.6953 1.3040 1.3801 1.6914

NWHH 0.5194 -0.5041 -1.0928

AVCHH 0.0459 -0.1652 0.0409 0.1801 -0.3482

O_WALKDENS -0.0058 -0.0086 0.0060 0.0066

D_WALKDENS 0.0213 0.0128 0.0259 0.0164

MODE CHOICE


7.4 Home Based Other Mode Choice The home based other mode choice model is run separately for two defined market segments. These are:

• Three or more person households • One and two person households

Table 7-5 shows the list of variables and the associated coefficients used in the home based other mode choice models. Coefficients provided for the specific market groups listed above are only used in the models applied for those specific market segments (these variables and coefficients appear last and are underlined).

TABLE 7-5 Home Based Other Mode Choice Model Variables and Coefficients


Drive w/ Passenger

Auto Passenger


Model Constant -0.2192 0.4757 -2.7682 -2.8514 -2.3935

IVTT -0.0716 -0.0716 -0.0716 -0.0716 -0.0716 -0.0716

OVTT -0.1075 -0.1075 -0.1075

AUTOPP -0.4686 -0.8712 -3.2475 -1.6513

D_TEMP1 0.0003 0.0004

3 or More Person HH

0.6910 1.0074

7.5 Non-Home Based Trip Mode Choice The non-home based trip mode choice models is produces estimates of mode share for trips with one end at work and the other end at some location other than the traveler’s home and where the trip both originate at and is destined to a “non home-non work” location. The mode choice models for non-home based trips rely on knowledge of the characteristics of the origin and destination locations, not of the characteristics of the traveler.

The tables below shows the list of variables and the associated coefficients used in the two non-home based trip mode choice models.

MODE CHOICE


TABLE 7-6 Non-Home Based Work Mode Choice Model Variables and Coefficients


Drive w/ Passenger

Auto Passenger


Model Constant -1.9430 -1.6368 -1.6974 -3.4902 -7.4196

IVTT -0.0669 -0.0669 -0.0669 -0.0669 -0.0669 -0.0669

OVTT -0.1003 -0.1003 -0.1003

O_WALKDENS 0.0004

O_RERATIO 1.8802

D_RERATIO 1.8802

Ln(D_TEMP1) 0.0734

O_PARKCOST -0.0064 -0.0064 -0.0064

D_PARKCOST -0.0064 -0.0064 -0.0064

TABLE 7-7 Non-Home Based Non-work Mode Choice Model Variables and Coefficients


Drive w/ Passenger

Auto Passenger


Model Constant -0.6652 -0.1704 -3.3072 -3.8852 -6.1466

IVTT -0.0565 -0.0565 -0.0565 -0.0565 -0.0565 -0.0565

OVTT -0.0847 -0.0847 -0.0847

O_WALKDENS 0.0010

CBD 0.7528 0.7528 0.7528 0.7528 0.7528 0.7528

Ln(D_TEMPH) 0.2357

O_PARKCOST 0.0107 0.0107 0.0107

D_PARKCOST 0.0107 0.0107 0.0107

7.6 Validation Results Following the calibration and final adjustment of the mode choice models, the models were applied to forecast 2002 mode choice using the 2002 socio-economic database, travel networks and related assumptions. These forecasts were compared to the mode shares from the household travel survey. The survey data for public transit was also checked and adjusted on the basis of transit rider counts from 2003 to improve the allocation of trips by time of day. Table 7-8 through 7-14 presents the results of this comparison. Information

MODE CHOICE


included is number of survey vs. model trips by mode by trip purpose and for the system as a whole. Each table presents the following for each trip purpose:

• Survey and model trip totals • Ratio between the two totals • Absolute difference between the two totals • Resulting difference of estimated mode share

TABLE 7-8 Survey vs. Model Mode Shares for Home Based Work Trips

Mode Survey Model Ratio Diff Share Diff

DA 143527 142283 0.991 1244 0.006

DP 9768 9637 0.987 131 0.001

AP 11143 10996 0.987 147 0.001

TR 4564 4609 1.010 -45 0.000

WALK 5043 6219 1.233 -1176 -0.007

BIKE 1683 1837 1.092 -154 -0.001

AUTO 153295 151920 0.991 1375 0.007

TOTAL 175728 175581 0.999 147 0.000

TABLE 7-9 Survey vs. Model Mode Shares for Home Based Shop Trips


DA 36014 34847 0.968 1167 0.002

DP 13920 13481 0.968 439 0.000

AP 18102 17523 0.968 579 0.001

TR 1625 1565 0.963 60 0.000

WALK 2022 2143 1.060 -121 -0.003

BIKE 407 432 1.061 -25 -0.001

AUTO 49934 48328 0.968 1606 0.002

TOTAL 72090 69991 0.971 2099 0.000

MODE CHOICE


TABLE 7-10 Survey vs. Model Mode Shares for Home Based School Trips


DA 0 7 0.000 -7 0.000

DP 31863 32008 1.005 -145 0.006

AP 31064 31660 1.019 -596 0.001

SB 20773 22004 1.059 -1231 -0.007

WALK 14763 15072 1.021 -309 0.000

BIKE 3996 4069 1.018 -73 0.000

AUTO 31863 32015 1.005 -152 0.006

TOTAL 102459 104820 1.023 -2361 0.000

TABLE 7-11 Survey vs. Model Mode Shares for Home Based Other Trips


DA 118049 132311 1.121 -14262 -0.002

DP 59580 65806 1.105 -6226 0.002

AP 84248 92436 1.097 -8188 0.005

TR 3251 3002 0.923 249 0.002

WALK 18643 21915 1.176 -3272 -0.004

BIKE 3423 4565 1.334 -1142 -0.002

AUTO 177629 198117 1.115 -20488 -0.001

TOTAL 287194 320035 1.114 -32841 0.000

Note: Model Totals include Hotel Visitor Model Trips

MODE CHOICE


TABLE 7-12 Survey vs. Model Mode Shares for Non-Home Based Work Trips


DA 77965 82575 1.059 -4610 0.001

DP 8267 8756 1.059 -489 0.000

AP 9188 9730 1.059 -542 0.000

TR 500 583 1.166 -83 0.000

WALK 10453 11323 1.083 -870 -0.002

BIKE 327 255 0.780 72 0.001

AUTO 86232 91331 1.059 -5099 0.002

TOTAL 106700 113222 1.061 -6522 0.000

TABLE 7-13 Survey vs. Model Mode Shares for Non-Home Based Non-work Trips


DA 85847 102078 1.189 -16231 0.006

DP 33488 40704 1.215 -7216 -0.002

AP 46365 56354 1.215 -9989 -0.003

TR 750 955 1.273 -205 0.000

WALK 7571 9295 1.228 -1724 -0.001

BIKE 726 888 1.223 -162 0.000

AUTO 119335 142782 1.196 -23447 0.004

TOTAL 174747 210274 1.203 -35527 0.000

MODE CHOICE


TABLE 7-14 Survey vs. Model Mode Shares for All Purposes


DA 461402 494101 1.071 -32699 0.005

DP 156886 170392 1.086 -13506 -0.001

PA 200110 218699 1.093 -18589 -0.002

TR 10691 10714 1.002 -23 0.001

SB 20773 22004 1.059 -1231 0.000

WALK 58495 65967 1.128 -7472 -0.003

BIKE 10562 12046 1.141 -1484 -0.001

AUTO 618288 664493 1.075 -46205 0.004

TOTAL 918919 993923 1.082 -75004 0

Examining the predicted vs. observed mode shares indicates that given similar assumptions regarding socio-economic conditions and networks, the model has the ability to reproduce the observed mode choice behavior.


SECTION 8

Traffic Assignment and Volumes

One of the most important indicators of model performance is the direct measurement of how well the travel model, in an attempt to replicate base (in this case, 2002) conditions can mimic existing traffic conditions as indicated by field counts of traffic volumes throughout the network. Such model validation is industry standard practice and a necessary prerequisite to using the model to forecast expected futures. To guide travel modelers in establishing minimum standards for the travel model development process, the Federal Highway Administration has published a set of guidelines for acceptable performance.

To validate the Anchorage model, four different types of model performance were measured. These included the following:

• Measured versus modeled aggregate traffic counts for selected cordons or cutline crossings

• Percent difference in volume between measured and modeled stratified by facility type

• Correlation measures between measured and modeled volumes for all available count locations

• Comparison of measured and modeled volumes at individual locations.

The count data used for comparison was all day weekday traffic counts for 2002. Sources were both ADOT and MOA counts supplied for this purpose and the ADOT Volumes map for the Anchorage and the Eagle River areas. These observed volumes were compared to the sum of the AM, PM and off peak period outputs for the 2002 Base travel model alternative utilizing trip distribution feedback. The results of each of these comparisons along with the corresponding Federal guidelines are presented below.

8.1 Cordon Crossings Measurement of traffic across cordons or screenlines provides an indication of how well the model performs in replicating major trip patterns and movements throughout the network. Cordons typically encompass all facilities that serve the same definable travel corridor. This allows for the fact that the model may not perfectly represent competition between parallel facilities. Federal guidelines for acceptable performance are based on a sliding scale which ranges from acceptable deviations of 65% for 5000 vehicles to 15% for 200,000 of more vehicles (Figure 8-1).

TRAFFIC ASSIGNMENT AND VOLUMES


0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

5 15 25 45 65 85 105 125 145 165 185 200 250

Volume(1000's)

Dev

iatio

n

Figure 8-1. Maximum Desirable Deviation in Total Screenline Volumes

For evaluating the Anchorage model’s performance, 23 separate cordons were defined. These ranged from specific east-west and north-south crossings in the downtown area to longer cordons which, in some cases, bisect the entire Anchorage Bowl. The cordons used and their associated identifiers are shown in Figure 8-2.

Once the cordons were defined, special procedures were written to automatically generate inputs to a set of spreadsheets to report their performance both in tabular and graphic form. The resulting spreadsheet reports which list and display counted and modeled traffic volumes for all links intersected by each cordon are provide as Appendix F–Screenlines. Table 8-1 itemizes performance of each cordon along with the measured deviation and the corresponding maximum deviation based on the Federal guideline. (Diff equals actual difference–Max Diff equals maximum desirable difference based on Federal guideline).

Based on Table 8-1, all cordons meet or exceed Federal guidelines. Of the 23, only six cordon have deviations greater than 0.10 and no screenlines have deviations greater than 0.20. Also, there does not seem to be any discernable pattern linking the cordons with a greater than 0.10 deviation. They are distributed throughout the study area. This is an indicator that there is no systematic bias in the model process.

Examination of graphs showing the relative facility counts and volumes for each cordon (Appendix F–Screenlines), shows that, in almost all cases, the relative magnitude of traffic volumes on specific facilities in a given corridor (i.e. for a given cordon) correlates well with observed count data.



Figure 8-1



TABLE 8-1 Summary of 2002 Weekday Cordon Counts/Volumes and Differences

ID Cordon Count Volume Diff Max Diff

101 N of Tudor Rd–Minnesota to Muldoon 251582 213846 0.150 0.150

201 N of Dimond Av–Minnesota to Birch 152853 148502 0.028 0.201

301 S of O'Malley–C St to Hillside 73446 69282 0.057 0.288

401 S of Glenn Hwy–Ingra to Muldoon 133621 117008 0.124 0.217

501 W of Muldoon–Tudor Rd to Glenn Hwy 104740 100093 0.044 0.245

601 W of Boniface–Tudor Rd to Davis 144484 146529 0.014 0.213

602 W of Birch–Rabbit Creek to Abbott 25010 26068 0.042 0.416

701 E of Lake Otis–Tudor Rd to Commercial Rd 193863 189001 0.025 0.172

702 E of Lake Otis–De Armoun to Dowling 57459 61801 0.076 0.317

801 E of New Seward–Rabbit Creek to 3rd 295053 279182 0.054 0.123

901 S of Int'l Airport–Minnesota to Lake Otis 212495 197652 0.070 0.162

1001 S of Dimond–Minnesota to Birch 117861 97897 0.169 0.232

2001 N of Eagle River–Glenn Hwy 24599 24443 0.006 0.418

2002 N of Eagle River Rd Access–Glenn to Birchwood 37470 35613 0.050 0.368

2003 S of Hiland–Glenn Hwy 48224 56701 0.176 0.338

2005 N of 3rd St–C St to Port Access 19071 19809 0.039 0.448

2006 W of Gambell -3rd St to 16th 72766 62875 0.136 0.305

2007 S of 9th–L St to Medfra 112614 109349 0.029 0.311

2008 E of Ingra–3rd to 15th 76077 81407 0.070 0.284

2010 N of Fireweed/Northern Lights–Minnesota to M ldoon

229702 217208 0.054 0.153

2013 E of Northwood–Northern Lights to Int'l Airport 83372 80889 0.030 0.273

2014 N of Int'l Airport–Spenard to Lake Otis 226517 217426 0.040 0.154

2016 W of Minnesota–Raspberry to Klatt 68604 56806 0.172 0.362

2020 S End of Study Area–Seward Hwy 8960 8754 0.023 0.538

Combined Cordon Totals 2770443 2618141 0.055

Note: Differences >0.10 are highlighted

8.2 Facility Type Comparisons The second validation test performed on the Anchorage model was to compare the percentage of measured versus modeled traffic for different facility types. This measures how well assumed and calculated speeds in the model are distributing traffic between the different types of facilities present in the model. As with cordons, FHWA has developed guidelines to measure what is considered to be acceptable performance for a travel model.



The facility types specified for comparison in the Federal guidelines are Freeways, Major Arterials, Minor Arterials and Collectors. To create a basis for comparison with the defined facility classes, the Anchorage model designations were grouped into these four categories using the following rules.

• Freeways were designed as Freeways (CLASS=1); • Expressways and major arterials were designated as Major Arterials (CLASS=2,3); • Minor arterials and frontage roads were designated as Minor Arterials (CLASS=4); • Collectors and local roads were designated as Collectors (CLASS=5,6).

Counts and volumes for each category were summarized and then the aggregate difference was compared to the corresponding Federal guideline. This summary is presented below in Table 8-2.

TABLE 8-2 Summary of 2002 Weekday Facility Class Counts/Volumes and Differences

Facility Class Count Volume Diff Max Diff

Freeway 1049059 1066200 1.016 0.070

Major Arterial 4696423 4390255 0.935 0.100

Minor Arterial 1511410 1236461 0.818 0.150

Collector 442589 318774 0.720 0.250

All Classes 7699481 7011690 0.911

As the table indicates, measured and modeled volumes for freeways and major arterials are well within the identified targets. For minor arterials and collectors the model is slightly underpredicting but is within 5% of the targets in both cases. Particularly for the most significant facilities, this indicates good correspondence with traffic count information. This conclusion is reinforced by the results of statistical analysis presented below.

8.3 Comparisons of Statistical Performance While both the cordon and facility class comparisons use a subset of the count information, the third performance evaluation approach used measures the overall statistical performance of the model by comparing all available counts to corresponding model volumes. Figure 8-3 is a scattergram, which shows all measured and modeled volumes pairs. Fitting a linear function to this dataset yields an R2 or correlation coefficient of .9043, which exceeds the Federal guideline of .880. The second statistical measure used is root-mean-square-error (RMSE). The RMSE for the same dataset is 29.15. This RMSE value is comparable with findings for other regional travel models (no Federal guidelines are proscribed).



Counts/Volumes

R2 = 0.9043

0

10000

20000

30000

40000

50000

60000

70000

0 10000 20000 30000 40000 50000 60000

Daily Traffic Counts

Trav

el M

odel

Vol

umes

Figure 8-3. STATISTICAL COMPARISON OF ALL AVAILABLE MEASURED AND MODELED 2002 WEEKDAY TRAFFIC VOLUMES

8.4 Individual Link Performance Federal guidelines are also provided for percent difference targets between measured and modeled traffic volumes for individual links based on total link volume. These targets are listed in Table 8-3.

TABLE 8-3 Percent Difference Targets for Daily Volumes for Individual Links

Average Annual Daily Traffic Maximum Desirable Deviation

<1,000 .60

1,000-2,500 .47

2,500-5,000 .36

5,000-10,000 .29

10,000-25,000 .25

25,000-50,000 .22

>50,000 .21



Using these guidelines, all count locations’ specific performance was evaluated. Of the 655 individual counts, 233 exceed the Federal targets. However, when corrections are made for directional imbalances in the model due to count information being non-directional, this number is reduced to 80. When only those locations with a daily count of greater than 2500 are considered, the number of locations exceeding Federal targets is 53 or about 8.1% of all count locations. In general, this indicates that the Anchorage model is replicating existing traffic behavior not only at the systemwide, facility group, and cordon levels, but also at the level of the individual road segment.


SECTION 9

Summary and Conclusions

A new and more robust travel demand model has been calibrated and validated for the Anchorage metropolitan area. This memorandum documents the new model structure, inputs, calibration and validation.

This new Anchorage travel demand model incorporates up-to-date travel behavior relationships for Anchorage residents (2002 Household Travel Survey) and the latest available demographic and employment attributes (2000 U. S. Census and 2002 Alaska Department of Labor employment database) for the metropolitan area.

The new model structure is significantly more robust and sophisticated than the travel model previously used in Anchorage. It incorporates dynamic relationships and sensitivity to land use development densities and patterns, household composition and demographic characteristics, transportation system connectivity, cost and level of service performance, and traveler decision-making regarding mode choice selection from available options.

In addition to traditional travel model functionality and features, the new Anchorage travel model interfaces seamlessly with the Anchorage Land Use Allocation Model. It also now incorporates a rich and powerful set of over two dozen performance evaluation capabilities, display outputs and metrics. These evaluation tools are entirely new -- and critically important to assist in assessing alternative transportation scenarios and related transportation system planning, operations, and design analysis.

Additionally, the entire model tool set is integrated in a user-friendly, menu-driven software environment which is far easier to use and maintain. Technical documentation is internally embedded for each model software application step.

The details of the model calibration are documented in this report.

The new travel model validation is complete and demonstrated to be in compliance with Federal Highway Administration federal guidelines.

• Socio-economic and demographic attributes of households are consistent with 2000 U.S. Census and State/local data

• Model-estimated person trip generation matches the 2002 Anchorage Household Travel Survey results

• Model-estimated transit passenger trips are within 1 percent of People Mover’s reported bus riders for 2002

• Model-estimated vehicle volumes compared to observed traffic counts for each of 23 different “cordons” across road links are each within FHWA guideline criteria. For all cordons combined, actual vs. model-estimated traffic volumes are within 5.5%

SUMMARY AND CONCLUSIONS


• A statistical linear regression fit of model-estimated vehicle volumes compared to all available observed traffic counts throughout the Anchorage road network shows a R2 correlation coefficient of 0.9043, again bettering the FHWA guideline criteria of 0.880.

In sum, the Anchorage Metropolitan Area Transportation Solutions (AMATS) organization now has an up-to-date, context-sensitive, powerful, proven and fully integrated travel demand modeling and analysis toolkit to address transportation planning and analysis needs.


Appendix A Variable Dictionary–Anchorage Travel Model

Socioeconomic Database

ANC/TP41158.DOC/050460017 A-1 FINAL DRAFT

APPENDIX A

Variable Dictionary–Anchorage Travel Model Socioeconomic Database

(Note: Does not include computed values)

• Traffic Analysis Zone ID – ZONE [integer]

• Total Population - POP [integer]

• Number of Households – TOTALHH [integer]

• Median Household Income (2002 Dollars) – AVINC [integer]

• Average Number of Workers/Household – AVWHH [real (X.XXX)]

• Enrollment of Schools in Zone – SENROLL [integer]

• EMPLOYMENT CATEGORIES

o Agricultural Employment – AEMP [integer]

o Mining Employment – BEMP [integer]

o Construction Employment – CEMP [integer]

o Manufacturing Employment – DEMP [integer]

o Transportation/Public Utilities Employment – EEMP [integer]

o Wholesale Trade Employment – FEMP [integer]

o Retail Trade Employment – GEMP [integer]

o Finance, Insurance and Real Estate Employment – HEMP [integer]

o Service Employment – IEMP [integer]

o Government Employment – JEMP [integer]

o University Employment – UEMP [integer]

o School Employment – SEMP [integer]

o Medical/Health Services Employment – XEMP [integer]

• Work Trip Parking Costs (2002 Dollars-Daily) - WPKCOST [real (X.XX)]

• Other Trip parking Cost (2002 Dollars-Monthly) - WPKCOST [real (X.XX)]


Appendix B Model Feedback Process

ANC/TP41158.DOC/050460017 B-1 FINAL DRAFT

APPENDIX B

Model Feedback Process

Feedback Using the Method of Successive Averages (Source: TransCAD V4.7 Travel Demand Forecasting)

In [direct feedback], congested travel times taken from the trip assignment results are directly fed back into the highway skim procedure. This method is commonly called the Direct Feedback Method. While Direct Feedback is relatively easy to understand and implement, many modelers who have implemented this method have reported a requirement of many feedback iterations before convergence occurs, or worse, the failure to reach convergence.

Theoretically, there is no guarantee of convergence with this method (TMIP: Incorporating Feedback in Travel Forecasting, March, 1996).

Because of this problem, alternatives to the Direct Feedback Method have been developed. One method that has had success is the Method of Successive Averages (MSA). In the MSA method, output volumes from trip assignment from previous iterations are weighted together to produce the current iteration’s link volumes. Adjusted congested times are then calculated based on the normal volume-delay relationship. This adjusted congested time is then fed back to the skimming procedures.

The adjusted volume is calculated based on the following equation:


Appendix C Commercial Vehicle Travel Model


APPENDIX C

Commercial Vehicle Travel Model

This section documents the estimation of base year commercial vehicle (or truck) models to support the update of the Anchorage Transportation demand Model. Truck models were developed using local socioeconomic (employment and population) data; local truck classification data recently collected by both the Municipality of Anchorage (MOA) and Alaska Department of Transportation (ADOT); and techniques outlined in the Quick Response Freight Manual (QFRM). The QFRM was developed by Cambridge Systematics for the Federal Highway Administration (FHWA) as a sketch planning tool for Metropolitan Planning Organizations (MPOs) and state departments of Transportation (State DOTs) to implement cost-effective truck activity models.

The Anchorage truck models represent base year daily (24-hour) conditions of single unit and combination commercial vehicle traffic. The truck vehicle categories correspond to those specified in the QFRM and also correspond to the standard vehicle classifications as specified by FHWA. These include:

• Single unit trucks – FHWA Classes 5 to 7; and • Combination trucks – FHWA Classes 8 to 13.

The following sections present the data used to support truck model development, the model development and application procedures used to both estimate and apply truck models for Anchorage, and the outputs generated from the truck modeling process. The model development and application section presents the methods used to develop truck trip generation, distribution, and assignment models, as well as the procedures used to validate the models.

1.1 Data Inputs The data inputs used to develop truck activity models for the Anchorage metropolitan area included employment data, transportation networks, and truck classification data.

Socioeconomic Dataset Socioeconomic data, primarily employment data, were obtained from the MOA to represent base year conditions by transportation analysis zone (TAZ) contained in the transportation network. Data were aggregated to be consistent with QRFM techniques. Employment data were aggregated into four categories: 1) Agriculture, Mining, and Construction; 2) Manufacturing, Transportation, Communications, Utilities, and Wholesale Trade; 3) Retail Trade; and 4) Office and Services. Household data were also used to represent household activity by TAZ.

COMMERCIAL VEHICLE TRAVEL MODEL

ANC/TP41158.DOC/050460017 C-2 FINAL DRAFT

Transportation Network The updated highway network from the Anchorage Transportation Demand Model was refined to incorporate roadway restrictions of heavy-duty truck movements on specific segments of Northern Lights Boulevard and E Street. This revised network was used to obtain zone-to-zone travel times and impedances to apply truck trip distribution (gravity) models.

Vehicle Classifications Truck classification data obtained from the MOA and ADOT were incorporated into the highway networks and were used for truck model validation. Classification data were obtained from the Highway Performance Monitoring System (HPMS) dataset. HPMS data was supplemented by classification data from the MOA and ADOT. Vehicle counts and classification data were also obtained from the Port of Anchorage and the Anchorage Landfill to specifically model special truck generators.

1.2 Model Development and Application Results This section presents the model development, application, and validation procedures used to estimate truck models for the Anchorage metropolitan area.

Truck Trip Generation Models The QFRM procedures were used to perform truck trip generation. The data processing and modeling steps for truck trip generation are presented below.

Internal TAZs Truck trip generation rates from the QFRM, as shown in Table 1.1, were applied to employment and households in each internal TAZ. Employment by category and household were taken from two Excel files developed by the MOA: EMPLOY.XLS and POPTAZ.XLS. According to a note in the employment file, military employment was not included. However, there did seem to be some employment allocated to the military base zone, which is, perhaps, civilian employment on the base. Also note that there were a number of zones with neither employment nor households. Accordingly, these internal zones did not generate any truck trips.


Table 1.1 Trip Generation Rates Applied to Internal Zones

Commercial Vehicle Trip Destinations (or Origins) per Day

QFRM Rates Adjusted Rates

Generator Single Unit (6+ Tires) Combination

Single Unit (6+ Tires) Combination

Employment:

Agriculture, Mining and Construction

0.289 0.174 0.358 0.230

Manufacturing, Transportation, Communications, Utilities and Wholesale Trade

0.242 0.104 0.300 0.137

Retail Trade 0.253 0.065 0.314 0.086

Office and Services 0.068 0.009 0.084 0.012

Households 0.099 0.038 0.123 0.050

Source: Table 4.1, page 4-4, Quick Response Freight Manual (Cambridge Systematics, 1996).

Trip generation rates were increased by 24 and 32 percent for single unit and combination vehicles, respectively, to bring the total screenline volumes closer to the validation data.

External TAZs As recommended by the QFRM, vehicle classification count data was used to estimate truck productions and attractions at the two external stations. Truck counts on the links closest to the external stations in the validation database were used for this purpose, ensuring a good fit of trips generated by the model to the observed data.

Special Generators Based on information provided by the MOA, the Port of Anchorage and the Anchorage Landfill were treated as special generators based on high levels of truck activity compared to the average transportation analysis zone within the Anchorage metropolitan area.

Port of Anchorage (TAZ 13) – For the Port TAZ, the number of total truck trips generated using the QFRM technique, including light duty, single unit, and combination vehicles, was comparable to the number of combination vehicle trips cited in the Traffic Flow Study conducted for the Port of Anchorage in 1996 (see Table 1.2 below). Since the Port’s Traffic Flow Study figures are corroborated by information gathered during the recently-conducted carrier interviews, the combination vehicle trip productions and attractions were increased, resulting in the following trips generated by vehicle category:



Table 1.2 Comparison of Combination Vehicle Traffic at the Port

Port Tenant From

Port Study From

Interviews QFRM Trip Generation

MAPCO 55 30 to 40

Tesoro 48

Texaco 55

SeaLand 202 100 to 200

TOTE 213 164

TOTAL 573 Combination: 39 Total Commercial Vehicle: 571

Note: Port study and interviews represent Tuesday, a high volume day for Port traffic.

Landfill (TAZ 648) – For the landfill, information provided by the MOA, including the total number of trips by vehicle category for 1997 and monthly volumes, was used to generate the following productions and attractions:

• Single unit – 75 trips (these represent mostly garbage trucks); and • Combination – 41 trips (these represent tractor-trailer combinations).

Trips were scaled by the total volume for the month of September to represent an average annual condition.

Total Trips The total trips generated by vehicle category for the Anchorage metropolitan area are shown in Table 1.3.

TransCAD Files The truck trip generation data is stored in the geographic layer TRIPGEN.DBD. Productions and attractions are stored in the fields P_SINGLE, A_SINGLE, P_COMBI, and A_COMBI. The scaled up versions of the productions and attractions are stored in fields with the prefixes P2 and A2. Productions and attractions were balanced and written to the binary data tables BALANCED.BIN and BALANCD_2.BIN.

1.2.2 Trip Distribution Models The QFRM procedures were also used to perform truck trip distribution as presented below.


Table 1.3 Summary of Households, Employment, and Commercial Vehicle Trips

Variable Count

(TAZs with value) Sum

RETAIL 554 24,188

FIRESVCS 554 73,317

MANUFWTCU 554 22,214

RESOURCES 554 11,819

TOTALEMP 554 131,538

SFHH 580 40,366

MFHH 580 47,782

POP 580 232,853

TOTHH 580 88,148

Single Unit Productions 557 36,388

Single Unit Attractions 557 36,359

Combination Productions 557 14,073

Combination Attractions 557 14,080

Note: Trips represent productions and attractions scaled up from QFRM.

Gravity Models The trip distribution models were estimated using gravity models with an exponential specification and with coefficients by vehicle category as recommended by the QFRM. The models take the following form:

)*( ijtcij eF −=

where:

F is the friction factor;

t is the travel time between Zones i and j; and

c is a coefficient.

The coefficients recommended by the QFRM are 0.1 for single unit vehicles and 0.03 for combination vehicles. These coefficients were used when distributing internal trips. However, a modified coefficient was applied when distributing trips from the external stations, as explained under the validation and calibration section.

Transportation Networks and Travel Times The highway network used to generate zone-to-zone travel times was basically the same as for passenger travel modeling with one exception: several links along Northern Lights Boulevard



and on E Street north of 3rd were disabled for combination vehicles. This adjustment was taken to simulate the more circuitous path that heavy-duty vehicles are likely to take when traveling to and from the airport or port. Separate travel time matrices for single unit and combination vehicles were applied in conjunction with the coefficients described above to produce the truck friction factor matrices for each vehicle class.

TransCAD Files The network file used in producing travel times was SS_STRTS.NET. A selection set (NO_TRKS) in the map (TRUCKS.MAP) was used to disable links, when necessary, using the network update feature. The friction factors are contained in a matrix file called TRKFFACT.MTX with matrices SINGLE and COMBI. The final truck trip table is stored in TRUCKS_3.MTX.

1.2.3 Truck Trip Assignment Models The truck trip tables were assigned using an all-or-nothing assignment method contained in TransCAD and the SS_STRTS.NET network file. Again, specific links on Northern Lights Boulevard and E Street were disabled when assigning combination vehicle traffic. Link-level outputs from the final model run were stored in two binary data files: SINGLE_3.BIN and COMBI_3.BIN. Total VMT without centroids was 225,334 and 113,513 for single unit and combination traffic, respectively.

1.2.4 Truck Model Validation and Calibration This section presents the process used to validate truck models for the Anchorage metropolitan area. As stated previously under truck trip generation, trip generation rates were increased by 24 and 32 percent for single unit and combination trucks to better match observed data.

Validation Database The database of vehicle classification counts was developed from several different sources, the primary source being the HPMS database. This source was expanded and modified in the following manner:

• The percent single unit and combination traffic in the HPMS data were combined with the 1994 AADT and split into two-way traffic to come up with truck counts by direction on each link that had HPMS classification data;

• Where new classification count data were available (those locations counted in September 1998), the new truck percents were scaled by the 1994 AADT on the link to come up with validation counts; and

• In some cases, where 1994 data were not available, AADT from adjacent links or AADT from the next most recent year were applied to come up with the validation counts.

These steps produced the most consistent and complete validation database possible with the available data.


Validation Procedures Because validation data were not available for many links, it was not possible to successfully use the built-in TransCAD screenline analysis procedures. Rather, the screenlines were drawn on a map as graphic representations and then a selection set of links with validation data that crossed the screenlines was developed for comparison of counts to assigned flows. The screenline locations are shown in Figure 1.1 and include:

1. Glenn Highway near northernmost external station;

2. Glenn Highway approaching Anchorage proper;

3. Port of Anchorage (Ocean Dock Road);

4. North-south between International Airport and Raspberry;

5. North-south between O’Malley and Huffman;

6. Seward Highway near the southernmost external station;

7. East-west airport approaches;

8. East-west, west of Seward Highway;

9. East-west, east of Boniface; and

10. East-west, west of Birch, Abbott, and Rabbit Creek.

Summary of Calibration Three separate model runs were completed before the model was judged to perform satisfactorily. The results and adjustments made are described below.

• Baseline 1 – The initial run revealed a problem with trip distribution in that almost all the trips generated at the external stations were remaining intra-TAZ trips. This occurred because the external zones are connected to the network with extraordinarily long connectors (almost 40 miles for Zone 901) and trips were simply not being distributed out of the zone with the QFRM gravity models. Some experimentation with the exponential function was made to select a coefficient that would result in a “flatter” curve, resulting in lower friction factors over the 30- to 40-mile distance range. This coefficient, 0.015, was applied only when distributing trips to or from the external stations in subsequent model runs.

• The other adjustment made at this stage was that the trip generation from the Port of Anchorage TAZ was revised to match the link counts on Ocean Dock Road. These were the only problems diagnosed at this stage and other aspects of the model performance, such as overall screenline performance, were not examined at this time.

• Baseline 2 – An analysis of links crossing the screenlines revealed that the total assigned volumes were approximately 60 percent of the link counts by approximately 24 and 32 percent for single unit and combination vehicles, respectively (see attached Table 1.4). In addition, the link volumes crossing Screenline 1 were quite low. Two adjustments were made to the trip generation models for the next model run:



The HPMS-based counts, rather than the previously-used ADOT classification count were used for trip generation at TAZ 900; and

The QFRM trip generation rates were increased by 24 and 32 percent to bring up the total link volumes.

• Baseline 3 – After making the above-described adjustments, another model run and screenline analysis were made. As shown in the attached Table 1.5, the total assigned volume fit was much closer to the observed volumes. The combination vehicle volume was 4 percent higher compared to observed data, while the single-unit traffic was low by 12 percent. The combination vehicle traffic was higher than expected because of the change in traffic generated at the Port TAZ between Baseline 2 and Baseline 3. In Baseline 2, the Port trip generation had been matched to the counts on Ocean Dock Road. In Baseline 3, the trip generation from the Port was inadvertently left at the higher levels developed through the special generator analyses. Since the level of traffic generated at the Port is subject to judgement, these results were left in place for the time being.

TransCAD Files

A selection set of the links used in the screenline analysis has been saved in TRUCKS.MAP.

1.3 Outputs Daily truck trips, trip tables by single and combination units, and a combined truck network assignment were validated for the truck model. These data will be used as inputs into the time-of-day and trip assignment elements of the base year Anchorage Transportation Demand Model. The procedures presented above will also form the foundation for developing future forecasts of truck travel in the Anchorage metropolitan region.


Table 1.4 Screenline Validation Results – Baseline 2

Count Data Link Flows Link Flow/Link Count Ratio

Single Unit Combination Single Unit Combination Single Unit Combination Link ID FEANME AB_H_S BA_H_S AB_H_C BA_H_C AB_FLOW BA_FLOW AB_FLOW BA_FLOW AB BA Total AB BA Total

1 15402 GLENN 813 813 542 542 432 353 221 127 0.53 0.43 0.48 0.41 0.24 0.32 2 15809 NEWGLENN 621 621 99 99 1,203 1,277 615 704 1.94 2.05 2.00 6.21 7.11 6.66 3 3894 OCEANDOCK 28 86 20 217 102 101 383 380 3.63 1.18 1.78 19.14 1.75 3.22 4 15725 BRAYTON 151 151 60 60 - 65 - 31 - 0.43 0.22 - 0.51 0.25 4 2163 C 991 991 142 142 570 700 233 299 0.58 0.71 0.64 1.65 2.11 1.88 4 16181 MINNESOTA 1,120 1,120 92 92 834 592 325 236 0.74 0.53 0.64 3.54 2.57 3.05 4 15726 NEWSEWARD 2,012 2,012 1,006 1,006 1,788 1,662 838 765 0.89 0.83 0.86 0.83 0.76 0.80

4,273 4,273 1,300 1,300 3,192 3,019 1,397 1,330 0.75 0.71 0.73 1.07 1.02 1.05 5 5857 BIRCH 210 210 9 9 32 36 13 15 0.15 0.17 0.16 1.40 1.65 1.52 5 5067 LAKEOTIS 108 108 15 15 76 59 30 22 0.71 0.54 0.62 1.91 1.43 1.67 5 4397 C 84 84 28 28 89 88 34 33 1.06 1.04 1.05 1.21 1.19 1.20 5 239 SOUTHPORT 22 22 - - 9 10 4 4 0.43 0.47 0.45 - - -

424 424 53 53 207 193 80 74 0.49 0.46 0.47 1.52 1.41 1.47 6 3904 NEWSEWARD 126 155 65 58 126 155 65 58 1.00 1.00 1.00 1.00 1.00 1.00 7 15894 DIMOND 119 119 22 22 - - - - - - - - - - 7 760 RASPBERRY 20 20 7 7 11 11 5 5 0.54 0.54 0.54 0.70 0.69 0.69 7 16205 INTERNATIONAL

AIRPORT 462 462 115 115 204 211 94 95 0.44 0.46 0.45 0.81 0.83 0.82

7 1578 OLDINTERNATIONALAIRPORT

192 192 38 38 154 143 65 63 0.80 0.75 0.77 1.70 1.63 1.67

793 793 183 183 369 365 164 163 0.46 0.46 0.46 0.90 0.89 0.89 8 11233 WHITNEY 374 374 110 110 152 105 74 69 0.41 0.28 0.34 0.67 0.63 0.65 8 8572 16TH 80 80 34 34 87 47 22 9 1.08 0.58 0.83 0.63 0.27 0.45 8 8266 NORTHERNLIGHTS 232 232 116 116 - 533 - 164 - 2.30 1.15 - 1.41 0.71 8 8140 TUDOR 3,208 3,208 2,085 2,085 786 1,015 335 440 0.25 0.32 0.28 0.16 0.21 0.19 8 6861 76TH 94 94 16 16 22 88 8 43 0.23 0.94 0.59 0.51 2.72 1.61 8 4581 100TH 27 27 5 5 4 4 1 1 0.15 0.15 0.15 0.31 0.29 0.30

4,015 4,015 2,366 2,366 1,052 1,792 440 726 0.26 0.45 0.35 0.19 0.31 0.25 9 11592 COMMERCIAL 462 462 132 132 35 33 12 12 0.08 0.07 0.07 0.09 0.09 0.09 9 9572 PENLAND 295 295 147 147 169 168 49 49 0.57 0.57 0.57 0.33 0.33 0.33 9 9494 DEBARR 593 720 22 - 739 651 282 248 1.25 0.90 1.06 12.82 - 12.82 9 15987 TUDOR 1,662 1,662 1,662 1,662 1,383 1,431 649 698 0.83 0.86 0.85 0.39 0.42 0.41

3,012 3,139 1,964 1,942 2,325 2,283 992 1,006 0.77 0.73 0.75 0.51 0.52 0.51 10 5946 O’MALLEY 933 933 41 41 163 169 64 67 0.17 0.18 0.18 1.59 1.66 1.62 10 5692 HUFFMAN 66 66 8 8 23 18 10 7 0.35 0.28 0.32 1.17 0.88 1.03

998 998 49 49 186 187 74 74 0.19 0.19 0.19 1.52 1.53 1.52 15,105 15,319 6,639 6,807 9,193 9,726 4,431 4,642 0.61 0.63 0.62 0.67 0.68 0.67



Table 1.5 Screenline Validation Results – Baseline 3

Count Data Link Flows Link Flow/Link Count Ratio

Single Unit Combination Single Unit Combination Single Unit Combination

SC Link ID FEANME AB_H_S BA_H_S AB_H_C BA_H_C AB_FLOW BA_FLOW AB_FLOW BA_FLOW AB BA Total AB BA Total

1 15402 GLENN 813 813 542 542 803 803 520 520 0.99 0.99 0.99 0.96 0.96 0.96 2 15809 NEWGLENN 621 621 99 99 1,775 1,773 1,100 1,101 2.86 2.85 2.86 11.11 11.12 11.12 3 3894 OCEANDOCK 28 86 20 217 102 101 384 385 3.63 1.18 1.78 19.22 1.77 3.25 4 2163 C 991 991 142 142 567 569 257 252 0.57 0.57 0.57 1.81 1.78 1.80 4 15725 BRAYTON 151 151 60 60 – – – – – – – – – – 4 15726 NEWSEWARD 2,012 2,012 1,006 1,006 2,231 2,214 1,084 1,074 1.11 1.10 1.10 1.08 1.07 1.07 4 16181 MINNESOTA 1,120 1,120 92 92 1,143 1,192 482 503 1.02 1.06 1.04 5.24 5.46 5.35 5 239 SOUTHPORT 22 22 – – 10 13 4 5 0.47 0.58 0.52 – – – 5 4397 C 84 84 28 28 131 128 55 52 1.56 1.52 1.54 1.95 1.85 1.90 5 5067 LAKEOTIS 108 108 15 15 44 37 15 12 0.41 0.34 0.38 1.00 0.78 0.89 5 5857 BIRCH 210 210 9 9 41 32 17 13 0.19 0.15 0.17 1.89 1.37 1.63 6 3904 NEWSEWARD 126 155 65 58 126 155 65 58 1.00 1.00 1.00 1.00 1.00 1.00 7 760 RASPBERRY 20 20 7 7 14 14 6 6 0.67 0.67 0.67 0.91 0.91 0.91 7 1578 OLDINTERNATIONAL

AIRPORT 192 192 38 38 190 177 86 83 0.99 0.92 0.96 2.23 2.15 2.19

7 15894 DIMOND 119 119 22 22 0 0 0 0 0.00 0.00 0.00 0.00 0.00 0.00 7 16205 INTERNATIONALAIR

PORT 462 462 115 115 241 264 123 126 0.52 0.57 0.55 1.07 1.09 1.08

8 4581 100TH 27 27 5 5 15 21 6 8 0.55 0.78 0.66 1.26 1.72 1.49 8 6861 76TH 94 94 16 16 70 123 36 63 0.75 1.31 1.03 2.33 3.99 3.16 8 8140 TUDOR 3,208 3,208 2,085 2,085 1,357 1,580 660 759 0.42 0.49 0.46 0.32 0.36 0.34 8 8266 NORTHERNLIGHTS 232 232 116 116 – 1,454 – 621 – 6.26 3.13 – 5.35 2.68 8 8572 16TH 80 80 34 34 58 35 15 7 0.72 0.43 0.57 0.43 0.20 0.32 8 11233 WHITNEY 374 374 110 110 369 320 255 172 0.99 0.86 0.92 2.32 1.56 1.94 9 9494 DEBARR 593 720 22 – 691 607 280 219 1.17 0.84 0.99 12.74 – 22.71 9 9572 PENLAND 295 295 147 147 229 200 72 60 0.78 0.68 0.73 0.49 0.41 0.45 9 11592 COMMERCIAL 462 462 132 132 168 57 72 22 0.36 0.12 0.24 0.55 0.16 0.36 9 15987 TUDOR 1,662 1,662 1,662 1,662 1,954 2,080 1,064 1,123 1.18 1.25 1.21 0.64 0.68 0.66 10 5692 HUFFMAN 66 66 8 8 33 41 14 18 0.50 0.62 0.56 1.64 2.21 1.93 10 5946 O’MALLEY 933 933 41 41 148 137 61 55 0.16 0.15 0.15 1.50 1.36 1.43 15,105 15,319 6,639 6,807 12,509 14,128 6,733 7,317 0.83 0.92 0.88 1.01 1.07 1.04


Appendix D Special Generators

ANC/TP41158.DOC/050460017 F-25 FINAL DRAFT

APPENDIX D

Special Generators TAZ PRATE PID ARATE AID HBWP HBSP HBSCHP HBOP NHBWP NHBNWP SINGLEP COMBIP HBWA HBSA HBSCHA HBOA NHBWA NHBNWA SINGLEA COMBIA LOCID

1 0.0492 Base Emp 0.0492 Base Emp 0.000 0.000 0.000 0.000 0.000 0.000 0.700 0.300 0.000 0.000 0.000 0.000 0.000 0.000 0.700 0.300 Elmendorf-Post Rd

2 1.4250 Base Pop 2.1700 Base Emp 0.300 0.330 0.183 0.183 0.565 0.264 0.086 0.084 Ft Richardson

2 0.0240 Base Emp 0.0240 Base Emp 0.000 0.000 0.000 0.000 0.000 0.000 0.700 0.300 0.000 0.000 0.000 0.000 0.000 0.000 0.700 0.300 Ft Richardson - Truc

13 1.5420 Emp 1.5420 Emp 0.203 0.797 0.203 0.797 Ship Creek (Trucks)

13 114.94 Freight Tons (M) 114.94 Frieght Tons (M) 0.000 0.000 0.000 0.000 0.000 0.000 0.150 0.850 0.000 0.000 0.000 0.000 0.000 0.000 0.150 0.850 Ship Creek - Port

142 4.3950 Beds 16.535 Beds 0.500 0.500 0.499 0.218 0.142 0.141 Columbia Hospital

142 0.0960 Emp 0.0960 Emp 0.000 0.000 0.000 0.000 0.000 0.000 0.875 0.125 0.000 0.000 0.000 0.000 0.000 0.000 0.875 0.125 Columbia Hos - Truck

217 0.0770 Students 0.9570 Students 0.608 0.392 0.240 0.000 0.565 0.053 0.068 UAA

217 0.1060 Emp 0.1060 Emp 0.000 0.000 0.000 0.000 0.000 0.000 0.875 0.125 0.000 0.000 0.000 0.000 0.000 0.000 0.875 0.125 UAA-Trucks

249 0.1150 Students 1.4360 Students 0.608 0.392 0.240 0.000 0.565 0.053 0.068 APU

249 0.1060 Emp 0.1060 Emp 0.000 0.000 0.000 0.000 0.000 0.000 0.875 0.125 0.000 0.000 0.000 0.000 0.000 0.000 0.875 0.125 APU-Trucks

253 0.0770 Students 0.9570 Students 0.608 0.392 0.240 0.000 0.565 0.053 0.068 UAA

253 0.1060 Emp 0.1060 Emp 0.000 0.000 0.000 0.000 0.000 0.000 0.875 0.125 0.000 0.000 0.000 0.000 0.000 0.000 0.875 0.125 UAA-Trucks

267 0.2270 Out Patients 0.9120 OutPatients 0.603 0.397 0.507 0.185 0.084 0.108 Providence Medical

267 0.0960 Emp 0.0960 Emp 0.000 0.000 0.000 0.000 0.000 0.000 0.875 0.125 0.000 0.000 0.000 0.000 0.000 0.000 0.875 0.125 Provid Hosp-Trucks

272 22.73 Emp 55.54 Emp 1.000 0.036 0.553 0.411 Library

288 None 2.8600 Emp 0.414 0.408 0.049 0.049 Alaska Native Hospit

288 0.0960 Emp 0.0960 Emp 0.000 0.000 0.000 0.000 0.000 0.000 0.875 0.125 0.000 0.000 0.000 0.000 0.000 0.000 0.875 0.125 Al Native Hosp - Trk

328 0.0020 Enplanements 0.0180 Enplanements 0.500 0.500 0.231 0.454 0.086 0.086 Airport

328 0.0065 Landings 0.0065 Landings 0.000 0.000 0.000 0.000 0.000 0.000 0.808 0.192 0.000 0.000 0.000 0.000 0.000 0.000 0.808 0.192 Airport-Trucks

343 None 1.3600 Unit 0.375 0.313 0.156 0.156 Air National Guard

648 0.0590 Region Pop(100s) 0.0590 Region Pop(100s) 0.650 0.350 0.650 0.350 Eagle River Site

900 38.65 Ext Pop(100s) 18.825 Ext Pop(100s) 0.253 0.100 0.000 0.447 0.060 0.100 0.026 0.014 0.060 0.180 0.684 0.052 0.028 North External

901 2.4050 Region Emp(100s) 4.0350 Region Pop(100s) 0.175 0.080 0.312 0.117 0.200 0.069 0.046 0.120 0.721 0.120 0.023 0.016 South External

910 2.4300 Base Pop 3.2890 Base Emp 0.300 0.330 0.183 0.183 0.565 0.264 0.086 0.084 Elmendorf-Gov't Hill

911 2.4300 Base Pop 3.2890 Base Emp 0.300 0.330 0.183 0.183 0.565 0.264 0.086 0.084 Elemendorf-Boniface


Appendix E Hotel/Motel Visitor Model

ANC/TP41158.DOC/050460017 E-1 FINAL DRAFT

APPENDIX E

Hotel/Motel Visitor Model ZONE ROOMS OCC NAME ADDRESS

27 59 0.60 Ramada Limited Hotel of A 207 MULDOON RD

30 100 0.60 Comfort Inn 111 W SHIP CREEK AVE

43 20 0.60 Anchorage Uptown Suites 234 E SECOND AVE

45 31 0.60 Anchorage Grand Hotel 505 W SECOND AVE

53 600 0.80 Hilton Anchorage Hotel 500 WEST THIRD AVE

53 26 0.60 Historic Anchorage Hotel 330 E ST

69 90 0.60 Ramada Inn Anchorage Dowt 115 E THIRD AVE

70 39 0.60 Merrill Field Inn 420 SITKA ST

71 41 0.60 Econo Lodge 642 E FIFTH AVE

79 547 0.80 The Hotel Captain Cook FOURTH AVENUE AT K STREET

79 14 0.60 Copper Whale Inn 440 L ST

81 251 0.75 Holiday Inn Downtown 239 W FOUTH AVE

81 43 0.60 Anchor Arms Hotel 433 EAGLE ST

85 60 0.60 Rodeway Inn 1124 E 5TH AVE

86 130 0.60 Days Inn Downtown 321 E FIFTH AVE

89 198 0.75 Westmark Anchorage Hotel 720 WEST FIFTH AVE

89 84 0.60 Inlet Inn 539 H ST

91 38 0.60 The Voyager Hotel 501 K ST

102 392 0.80 Anchorage Marriott Downtown 820 WEST 7TH AVE

106 148 0.75 Marriott Residence Inn 1025 E 35TH AVE.

107 375 0.80 Sheraton Anchorage Hotel 401 E SIXTH AVE

113 111 0.75 Hawthorn Suites 1110 W EIGHTH AVE

116 111 0.75 Clarion Suites 325 W EIGHTH AVE

116 95 0.60 Aspen Hotel 108 E 8TH AVE.

153 180 0.60 Inlet Tower Hotel & Suite 1200 L ST

155 20 0.60 Anchorage Suit Lodge 441 E 15TH AVE

257 102 0.75 Springhill Suites 3401 A ST

268 83 0.60 Best Western Golden Lion 1000 E 36TH AVE

279 40 0.60 Best Inns & Suites 4110 SPENARD RD

279 38 0.60 Spenard Hotel Motel Inn 3960 SPENARD RD

279 34 0.60 Chelsea Inn Hotel 3836 SPENARD RD

298 101 0.75 Hampton Inn 4301 CREDIT UNION DR

300 86 0.60 Puffin Inn 4400 SPENARD RD

303 217 0.75 Best Western Barratt Inn 4616 SPENARD RD

303 128 0.60 Holiday Inn Express 4411 SPENARD RD

303 102 0.60 Best Value Inn Executive S 4360 SPENARD RD

310 248 0.80 Millennium Alaskan Anchor 4800 SPENARD ROAD

316 125 0.75 Hilton Garden Inn 100 W TUDOR RD

316 122 0.75 Homewood Suites by Hilton 140 W TUDOR RD

316 106 0.75 Fairfield Inn & Suites 5060 A STREET

319 141 0.75 Coast International Inn 3333 W INT'L AIRPORT RD

326 154 0.75 Courtyard by Marriott 4901 SPENARD RD

326 79 0.60 Microtel Inn & Suites 5205 NORTHWOOD DR

335 24 0.60 Arctic Inn Motel 842 W INT'L AIRPORT RD

394 109 0.60 Dimond Center Hotel 700 E DIMOND BLVD


Appendix F Screenlines


APPENDIX F

Screenlines (101) North of Tudor – Minnesota to O’Malley

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLMinnesotaDr_SB 22236 20636 42872 24900 24900 49800 21808 17536 39345ArcticBlvd_SB 11080 11907 22987 15000 15000 30000 4278 3544 7823CSt_NB 14749 15643 30392 24900 24900 49800 9434 13157 22591DenaliSt_NB 4100 4100 8200 15000 15000 30000 5324 2748 8071OldSewardHwy_SB 6135 6135 12270 16600 16600 33200 2748 4423 7171NewSewardHwy_NB 26812 0 26812 58500 0 58500 27744 0 27744NewSewardHwy_SB 26812 0 26812 58500 0 58500 29334 0 29334MacInnesSt_SB 748 748 1496 5000 5000 10000 31 24 54LakeOtisPkwy_SB 12795 12795 25590 23000 23000 46000 14426 11983 26409BragawSt_SB 3820 3504 7324 15000 15000 30000 3476 3866 7341BonifacePkwy_SB 7566 7173 14739 16600 16600 33200 8993 8716 17709BaxterRd_SB 2522 2522 5044 5000 5000 10000 624 556 1180MuldoonRd_SB 12000 12000 24000 23000 23000 46000 8818 8712 17530PattersonSt_SB 1522 1522 3044 5000 5000 10000 800 744 1544Screenline Totals 152897 98685 251582 306000 189000 495000 137836 76010 213846

05000

100001500020000250003000035000400004500050000

Minn

esot

aDr_

SBArc

ticBlvd

_SB

CSt_NB

Denali

St_NB

OldSew

ardH

wy_SB

NewSew

ardH

wy_NB

NewSew

ardH

wy_SB

Mac

Inne

sSt_

SBLa

keOtis

Pkwy_

SBBra

gawSt_

SBBon

iface

Pkwy_

SBBax

terR

d_SB

Muld

oonR

d_SB

Patte

rson

St_SB

Counts

Model

SCREENLINES


(201) North of Dimond Avenue – Minnesota to Birch

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLMinnesotaDr_SB 17849 0 17849 39000 0 39000 16896 0 16896MinnesotaDr_NB 17800 0 17800 39000 0 39000 19434 0 19434CSt_NB 5800 5800 11600 20750 20750 41500 3636 3677 731388thAve_EB 2500 2685 5185 6500 6500 13000 1645 1771 3416LakeOtisPkwy_SB 5905 5905 11810 23000 23000 46000 8941 7190 16132KingSt_SB 750 750 1500 6500 6500 13000 1109 1095 2204NewSewardHwy_SB 27816 0 27816 39000 0 39000 27150 0 27150NewSewardHwy_NB 28089 0 28089 39000 0 39000 25503 0 25503OldSewardHwy_SB 8801 8801 17602 23000 23000 46000 7544 10593 18137HomerDr_SB 1780 0 1780 18000 0 18000 3022 0 3022BraytonDr_SB 0 3660 3660 0 18000 18000 0 1718 1718DimondBlvd_EB 2500 2500 5000 5500 5500 11000 3123 2345 5467AbbottLoopRd_SB 1581 1581 3162 5000 5000 10000 1178 933 2110Screenline Totals 121171 31682 152853 264250 108250 372500 119182 29320 148502

0

5000

10000

15000

20000

25000

30000

Minn

esot

aDr_

SBM

innes

otaD

r_NB

CSt_NB

88th

Ave_E

BLa

keOtis

Pkwy_

SB

KingSt_

SB

NewSew

ardH

wy_SB

NewSew

ardH

wy_NB

OldSew

ardH

wy_SB

Homer

Dr_SB

Brayto

nDr_

SBDim

ondB

lvd_E

BAbb

ottL

oopR

d_SB

Counts

Model


(301) South of O’Malley – C Street to Hillside

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLCSt_SB 4367 4404 8771 15000 15000 30000 4317 4522 8839OldSewardHwy_SB 11302 11302 22604 7500 7500 15000 8868 9323 18191BraytonDr_SB 0 5400 5400 0 18000 18000 0 1220 1220NewSewardHwy_SB 15155 0 15155 39000 0 39000 16564 0 16564NewSewardHwy_NB 13500 0 13500 39000 0 39000 16560 0 16560ElmoreRd_SB 626 626 1252 7400 7400 14800 621 662 1284BirchRd_SB 1146 1146 2292 5000 5000 10000 719 827 1546HillsideDr_SB 2236 2236 4472 7400 7400 14800 2509 2568 5078Screenline Totals 48332 25114 73446 120300 60300 180600 50158 19123 69282

0

5000

10000

15000

20000

25000

CSt_SB

OldSew

ardH

wy_SB

Brayto

nDr_

SBNew

Sewar

dHwy_

SBNew

Sewar

dHwy_

NB

Elmor

eRd_

SB

BirchR

d_SB

Hillside

Dr_SB

Counts

Model

SCREENLINES


(401) South of Glenn Hwy – Ingra to Muldoon

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLGambellSt_SB 18800 0 18800 33200 0 33200 22687 0 22687IngraSt_SB 0 19478 19478 0 33200 33200 0 18155 18155KarlukSt_SB 1200 1200 2400 5000 5000 10000 1552 1023 2576AirportHeightsDr_SB 6451 5737 12188 15000 15000 30000 10773 8929 19702TurpinSt_SB 3530 3530 7060 5000 5000 10000 1131 1330 2460BonifacePkwy_SB 9776 9554 19330 16600 16600 33200 8130 8426 16556BragawSt_SB 9965 9965 19930 12400 12400 24800 8517 8106 16624McCarreySt_NB 2382 1200 3582 6500 6500 13000 997 1036 2033OklahomaDr_SB 800 800 1600 5000 5000 10000 615 287 902MuldoonRd_SB 13943 15310 29253 18400 18400 36800 7288 8025 15314Screenline Totals 66847 66774 133621 117100 117100 234200 61691 55317 117008

0

5000

10000

15000

20000

25000

30000

35000

Gambe

llSt_

SB

Ingr

aSt_

SB

Karluk

St_SB

Airpor

tHeig

htsD

r_SB

Turpin

St_SB

Bonifa

cePkw

y_SB

Braga

wSt_SB

McC

arre

ySt_

NBOkla

hom

aDr_

SBM

uldoo

nRd_

SB

Counts

Model


(501) West of Muldoon – Tudor to Glenn Hwy

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOL36thAve_EB 676 676 1352 5000 5000 10000 1072 1042 2113NorthernLightsBlvd_EB 7478 7234 14712 16600 16600 33200 5367 5176 10544DebarrRd_EB 7175 7175 14350 23000 23000 46000 5459 6293 11753MuldoonRd_SB 12000 12000 24000 23000 23000 46000 8818 8712 17530NewGlennHwy_EB 24085 0 24085 58500 0 58500 28093 0 28093NewGlennHwy_WB 24085 0 24085 58500 0 58500 27881 0 278816thAve_EB 778 778 1556 5000 5000 10000 782 745 1528BoundaryDr_EB 300 300 600 5000 5000 10000 217 435 652Screenline Totals 76577 28163 104740 194600 77600 272200 77689 22404 100093

0

5000

10000

15000

20000

25000

30000

36th

Ave_E

BNor

ther

nLigh

tsBlvd

_EB

Debar

rRd_

EB

Muld

oonR

d_SB

NewGlen

nHwy_

EB

NewGlen

nHwy_

WB

6thA

ve_E

B

Bound

aryD

r_EB

Counts

Model

SCREENLINES


(601) West of Boniface – Tudor to Davis

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLDebarrRd_EB 12700 12700 25400 23000 23000 46000 11161 12269 23430TudorRd_EB 21201 20825 42026 23000 23000 46000 21061 21895 42955NorthernLightsBlvd_EB 12410 12410 24820 16600 16600 33200 10844 11232 22076NewGlennHwy_WB 21615 0 21615 39000 0 39000 26579 0 26579DavisSt_EB 4504 4504 9008 11250 11250 22500 2756 2945 5701NewGlennHwy_EB 21615 0 21615 39000 0 39000 25787 0 25787Screenline Totals 94045 50439 144484 151850 73850 225700 98188 48341 146529

05000

100001500020000250003000035000400004500050000

Debar

rRd_

EB

Tudor

Rd_EB

North

ernL

ightsB

lvd_E

B

NewGlen

nHwy_

WB

DavisS

t_EB

NewGlen

nHwy_

EB

Counts

Model


(602) West of Birch – Rabbit Creek to Abbott

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLRabbitCreekRd_EB 3626 3626 7252 7400 7400 14800 4410 4471 8881DeArmounRd_EB 1010 1010 2020 7400 7400 14800 535 552 1088HuffmanRd_EB 1030 1030 2060 7400 7400 14800 976 932 1908OMalleyRd_EB 3600 3600 7200 7400 7400 14800 3580 3670 7250AbbottRd_EB 3239 3239 6478 7400 7400 14800 3341 3600 6941Screenline Totals 12505 12505 25010 37000 37000 74000 12843 13225 26068

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

RabbitCreekRd_EB DeArmounRd_EB HuffmanRd_EB OMalleyRd_EB AbbottRd_EB

Counts

Model

SCREENLINES


(701) East of Lake Otis – Tudor to Commercial

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLTudorRd_EB 24810 24810 49620 23000 23000 46000 22419 23144 45563ProvidenceDr_EB 5981 5269 11250 15000 15000 30000 4821 4316 9137NorthernLightsBlvd_EB 18000 18000 36000 16600 16600 33200 19270 20617 3988720thAve_EB 1300 1300 2600 5000 5000 10000 1042 1174 221616thAve_EB 200 200 400 5000 5000 10000 5 16 21DebarrRd_EB 17500 17500 35000 16600 16600 33200 16786 19038 35823MerrillFieldDr_EB 200 200 400 5500 5500 11000 181 193 374CommercialDr_EB 6679 6679 13358 15000 15000 30000 5390 5229 106185thAve_EB 21954 23281 45235 16600 16600 33200 23190 22170 45361Screenline Totals 96624 97239 193863 118300 118300 236600 93104 95896 189001

0

10000

20000

30000

40000

50000

60000

Tudor

Rd_EB

Provid

ence

Dr_EB

North

ernL

ightsB

lvd_E

B

20th

Ave_E

B

16th

Ave_E

B

Debar

rRd_

EBM

errill

FieldD

r_EB

Comm

ercia

lDr_

EB

5thA

ve_E

B

Counts

Model


(702) East of Lake Otis – DeArmoun to Dowling

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLDeArmounRd_EB 2130 2106 4236 7400 7400 14800 2723 2872 5594AbbottRd_EB 6390 6390 12780 7400 7400 14800 6604 6144 12749HuffmanRd_EB 3605 3605 7210 7400 7400 14800 4048 4056 8104OMalleyRd_EB 6825 6825 13650 7400 7400 14800 5740 5985 1172580thAve_EB 2805 2805 5610 7400 7400 14800 3432 3796 722872ndAve_EB 1466 1467 2933 7400 7400 14800 2031 2173 420468thAve_EB 3317 3317 6634 7400 7400 14800 2797 2937 5734DowlingRd_EB 1453 1453 2906 7500 7500 15000 1582 1784 336784thAve_WB 750 750 1500 7400 7400 14800 1486 1610 3096Screenline Totals 28741 28718 57459 66700 66700 133400 30443 31358 61801

02000400060008000

10000120001400016000

DeArm

ounR

d_EB

Abbot

tRd_

EB

Huffm

anRd_

EB

OMall

eyRd_

EB

80th

Ave_E

B

72nd

Ave_E

B

68th

Ave_E

B

Dowlin

gRd_

EB

84th

Ave_W

B

Counts

Model

SCREENLINES


(801) East of New Seward Hwy – Rabbit Creek to 3rd

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOL15thAve_EB 11677 11677 23354 20750 20750 41500 15023 13381 28404BensonBlvd_EB 20500 0 20500 33200 0 33200 19981 0 19981NorthernLightsBlvd_EB 0 20500 20500 0 33200 33200 0 22377 223775thAve_EB 0 24000 24000 0 24900 24900 0 22688 226884thAve_EB 9600 0 9600 22500 0 22500 7781 0 77813rdAve_WB 0 8162 8162 0 22500 22500 0 6282 62829thAve_EB 1380 1380 2760 5000 5000 10000 962 1307 22696thAve_WB 20000 0 20000 24900 0 24900 22887 0 22887LoreRd_EB 500 500 1000 5000 5000 10000 86 1142 1228DimondBlvd_EB 8800 8800 17600 16600 16600 33200 12173 10859 2303268thAve_EB 400 400 800 5000 5000 10000 105 337 442OMalleyRd_EB 7974 8182 16156 8300 8300 16600 8294 8580 16873AcademyDr_EB 1500 1500 3000 7500 7500 15000 104 127 230OldSewardHwy_SB 4000 4000 8000 12450 12450 24900 5078 5117 10195DeArmounRd_EB 3926 3926 7852 7400 7400 14800 2505 2846 5352HuffmanRd_EB 7715 7715 15430 7500 7500 15000 6043 5815 11857DowlingRd_EB 17500 17500 35000 20750 20750 41500 12760 12607 25367TudorRd_EB 22021 22318 44339 23000 23000 46000 21244 22137 4338136thAve_EB 8500 8500 17000 15000 15000 30000 4021 4535 8556Screenline Totals 145993 149060 295053 234850 234850 469700 139045 140137 279182

05000

100001500020000250003000035000400004500050000

15th

Ave_E

B

Benso

nBlvd

_EB

North

ernL

ightsB

lvd_E

B5t

hAve

_EB

4thA

ve_E

B3r

dAve

_WB

9thA

ve_E

B6t

hAve

_WB

Lore

Rd_EB

Dimon

dBlvd

_EB

68th

Ave_E

BOM

alley

Rd_EB

Acade

myD

r_EB

OldSew

ardH

wy_SB

DeArm

ounR

d_EB

Huffm

anRd_

EBDow

lingR

d_EB

Tudor

Rd_EB

36th

Ave_E

B

Counts

Model


(901) South of International Airport Road – Minnesota to Lake Otis

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLMinnesotaDr_NB 24579 0 24579 39000 0 39000 22821 0 22821MinnesotaDr_SB 25317 0 25317 39000 0 39000 23222 0 23222ArcticBlvd_SB 7643 7791 15434 15000 15000 30000 7110 7675 14785CSt_NB 9385 9855 19240 20750 20750 41500 7402 7974 15376NewSewardHwy_NB 30446 0 30446 39000 0 39000 28293 0 28293NewSewardHwy_SB 30495 0 30495 39000 0 39000 31683 0 31683OldSewardHwy_SB 12987 12987 25974 16600 16600 33200 7354 14689 22043HomerDr_SB 2010 0 2010 18000 0 18000 4569 0 4569BraytonDr_SB 0 5000 5000 0 18000 18000 0 3113 3113LakeOtisPkwy_SB 17000 17000 34000 23000 23000 46000 17015 14734 31749Screenline Totals 159862 52633 212495 249350 93350 342700 149468 48185 197652

05000

10000150002000025000300003500040000

Minn

esot

aDr_

NBM

innes

otaD

r_SB

Arctic

Blvd_S

B

CSt_NB

NewSew

ardH

wy_NB

NewSew

ardH

wy_SB

OldSew

ardH

wy_SB

Homer

Dr_SB

Brayto

nDr_

SBLa

keOtis

Pkwy_

SB

Counts

Model

SCREENLINES


(1001) South of Dimond – Minnesota to Birch

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLFrontageRd_SB 530 0 530 18000 0 18000 0 0 0MinnesotaDr_SB 13480 0 13480 39000 0 39000 13016 0 13016FrontageRd_NB 3100 0 3100 18000 0 18000 841 0 841MinnesotaDr_NB 13480 0 13480 39000 0 39000 7682 0 7682CSt_NB 100 100 200 18400 18400 36800 248 237 485NewSewardHwy_SB 19057 0 19057 39000 0 39000 17685 0 17685NewSewardHwy_NB 18918 0 18918 39000 0 39000 19891 0 19891HomerDr_SB 3380 0 3380 18000 0 18000 178 0 178BraytonDr_SB 0 5000 5000 0 18000 18000 0 2876 2876OldSewardHwy_SB 9221 9221 18442 16600 16600 33200 7581 7267 14848IndependenceDr_SB 2211 2211 4422 5000 5000 10000 2912 1726 4638LakeOtisPkwy_SB 7337 7857 15194 16600 16600 33200 7371 6372 13742BirchRd_SB 1329 1329 2658 7400 7400 14800 916 1100 2015Screenline Totals 92143 25718 117861 274000 82000 356000 78320 19577 97897

0

5000

10000

15000

20000

25000

Front

ageR

d_SB

Minn

esot

aDr_

SBFro

ntag

eRd_

NBM

innes

otaD

r_NB

CSt_NB

NewSew

ardH

wy_SB

NewSew

ardH

wy_NB

Homer

Dr_SB

Brayto

nDr_

SBOldS

ewar

dHwy_

SB

Inde

pend

ence

Dr_SB

Lake

OtisPkw

y_SB

BirchR

d_SB

Counts

Model


(2001) North of Eagle River – Glenn Highway

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLGlennHwy_SB 12278 12321 24599 39000 39000 78000 12501 11943 24443Screenline Totals 12278 12321 24599 39000 39000 78000 12501 11943 24443

24350

24400

24450

24500

24550

24600

24650

GlennHwy_SB

Counts

Model

SCREENLINES


(2002) North of Eagle River Road Access – Glenn to Birchwood

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLNewGlennHwy_NB 17490 0 17490 39000 0 39000 16841 0 16841NewGlennHwy_SB 17490 0 17490 39000 0 39000 17181 0 17181OldGlennHwy_NB 1245 1245 2490 6200 6200 12400 789 802 1591Screenline Totals 36225 1245 37470 84200 6200 90400 34811 802 35613

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

NewGlennHwy_NB NewGlennHwy_SB OldGlennHwy_NB

Counts

Model


(2003) South of Hiland – Glenn Highway

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLNewGlennHwy_NB 24892 0 24892 58500 0 58500 27237 0 27237NewGlennHwy_SB 23332 0 23332 58500 0 58500 29464 0 29464Screenline Totals 48224 0 48224 117000 0 117000 56701 0 56701

0

5000

10000

15000

20000

25000

30000

35000

NewGlennHwy_NB NewGlennHwy_SB

Counts

Model

SCREENLINES


(2005) North of 3rd – C Street to Port Access

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLOceanDockRd_NB 1299 1299 2598 6500 6500 13000 818 764 1582PortAccessRd_NB 0 8910 8910 0 16600 16600 0 8834 8834PortAccessRd_NB 7563 0 7563 16600 0 16600 9393 0 9393Screenline Totals 8862 10209 19071 23100 23100 46200 10210 9599 19809

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

OceanDockRd_NB PortAccessRd_NB PortAccessRd_NB

Counts

Model


(2006) West of Gambell – 3rd to 16th

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOL16thAve_EB 1630 1630 3260 6500 6500 13000 1197 2124 332115thAve_EB 6204 7861 14065 15000 15000 30000 5013 8487 135004thAve_EB 9432 0 9432 22500 0 22500 10492 0 104923rdAve_EB 0 5693 5693 0 15000 15000 0 6712 67126thAve_EB 13242 0 13242 24900 0 24900 13646 0 136469thAve_EB 4048 3448 7496 15000 15000 30000 2183 1821 40045thAve_EB 0 19578 19578 0 24900 24900 0 11200 11200Screenline Totals 34556 38210 72766 83900 76400 160300 32532 30343 62875

0

5000

10000

15000

20000

25000

16thAve_EB 15thAve_EB 4thAve_EB 3rdAve_EB 6thAve_EB 9thAve_EB 5thAve_EB

Counts

Model

SCREENLINES


(2007) South of 9th – L to Medfra

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLASt_SB 0 12873 12873 0 24900 24900 0 14792 14792GambellSt_SB 23233 0 23233 33200 0 33200 25017 0 25017LSt_SB 16339 0 16339 24900 0 24900 13703 0 13703ISt_SB 0 16270 16270 0 24900 24900 0 21326 21326ESt_SB 1793 1793 3586 5000 5000 10000 383 597 980CSt_SB 19339 0 19339 24900 0 24900 13172 0 13172IngraSt_SB 0 19478 19478 0 33200 33200 0 19840 19840MedfraSt_SB 748 748 1496 5000 5000 10000 249 271 520Screenline Totals 61452 51162 112614 93000 93000 186000 52523 56826 109349

0

5000

10000

15000

20000

25000

30000

ASt_SB GambellSt_SB LSt_SB ISt_SB ESt_SB CSt_SB IngraSt_SB MedfraSt_SB

Counts

Model


(2008) East of Ingra – 3rd to 15th

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOL15thAve_EB 11320 11320 22640 20750 20750 41500 14437 12360 267985thAve_EB 21109 21966 43075 16600 16600 33200 22189 22392 445813rdAve_EB 5181 5181 10362 15000 15000 30000 5629 4400 10029Screenline Totals 37610 38467 76077 52350 52350 104700 42255 39153 81407

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

50000

15thAve_EB 5thAve_EB 3rdAve_EB

Counts

Model

SCREENLINES


(2010) North of Fireweed/ Northern Lights – Minnesota to Muldoon

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLASt_SB 0 16296 16296 0 24900 24900 0 19589 19589MinnesotaDr_SB 0 15600 15600 0 16600 16600 0 16673 16673MinnesotaDr_SB 17436 0 17436 24900 0 24900 15150 0 15150SpenardRd_SB 0 4142 4142 0 12000 12000 0 2456 2456CSt_SB 16810 0 16810 24900 0 24900 17688 0 17688ArcticBlvd_SB 3600 3600 7200 5000 5000 10000 1399 1617 3016NewSewardHwy_SB 26200 27500 53700 33200 33200 66400 28029 29026 57055LakeOtisPkwy_SB 10000 10000 20000 12450 12450 24900 9199 4498 13697BragawSt_SB 9297 9154 18451 18750 18750 37500 10298 8599 18897BonifacePkwy_SB 10303 9912 20215 16600 16600 33200 12808 12771 25579BaxterRd_SB 3300 3300 6600 5000 5000 10000 2262 2156 4417MuldoonRd_SB 16294 16958 33252 23000 23000 46000 11265 11725 22990Screenline Totals 113240 116462 229702 163800 167500 331300 108099 109109 217208

0

10000

20000

30000

40000

50000

60000

ASt_SB

Minn

esot

aDr_

SBM

innes

otaD

r_SB

Spena

rdRd_

SB

CSt_SB

Arctic

Blvd_S

BNew

Sewar

dHwy_

SBLa

keOtis

Pkwy_

SBBra

gawSt_

SBBon

iface

Pkwy_

SBBax

terR

d_SB

Muld

oonR

d_SB

Counts

Model


(2013) East of Northwood – Northern Lights to International Airport

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLIntlAirportRd_WB 18074 18074 36148 30000 30000 60000 19560 18938 38498SpenardRd_SB 9710 9710 19420 15000 15000 30000 6988 11255 18244McRaeRd_WB 1382 1382 2764 5000 5000 10000 1747 1819 3566NorthernLightsBlvd_WB 12491 12549 25040 16600 16600 33200 9786 10796 20582Screenline Totals 41657 41715 83372 66600 66600 133200 38081 42808 80889

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

IntlAirportRd_WB SpenardRd_SB McRaeRd_WB NorthernLightsBlvd_WB

Counts

Model

SCREENLINES


(2014) North of International Airport – Spenard to Lake Otis

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLMinnesotaDr_SB 23716 0 23716 24900 0 24900 24632 0 24632CSt_NB 12977 12977 25954 24900 24900 49800 9303 12393 21696MinnesotaDr_NB 21567 0 21567 58500 0 58500 18188 0 18188SpenardRd_SB 9405 9405 18810 7500 7500 15000 8534 11553 20087NorthwoodSt_SB 1065 1066 2131 5000 5000 10000 756 536 1292ArcticBlvd_SB 7355 6772 14127 15000 15000 30000 5207 8691 13898NewSewardHwy_NB 30446 0 30446 39000 0 39000 28293 0 28293HomerDr_SB 1011 0 1011 18000 0 18000 1388 0 1388

0 0 1260 1260 0 18000 18000 0 3113 3113NewSewardHwy_SB 30495 0 30495 39000 0 39000 31683 0 31683OldSewardHwy_SB 11500 11500 23000 16600 16600 33200 8772 12636 21408LakeOtisPkwy_SB 17000 17000 34000 23000 23000 46000 17015 14734 31749Screenline Totals 166537 59980 226517 271400 110000 381400 153771 63655 217426

05000

10000150002000025000300003500040000

Minn

esot

aDr_

SB

CSt_NB

Minn

esot

aDr_

NBSpe

nard

Rd_SB

North

woodS

t_SB

Arctic

Blvd_S

BNew

Sewar

dHwy_

NBHom

erDr_

SB 0New

Sewar

dHwy_

SBOldS

ewar

dHwy_

SBLa

keOtis

Pkwy_

SB

Counts

Model


(2016) West of Minnesota – Raspberry to Klatt

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLDimondBlvd_WB 14076 15004 29080 24900 24900 49800 12655 17010 29665100thAve_WB 3915 3915 7830 15000 15000 30000 5606 2161 7767KlattRd_WB 3015 3015 6030 13000 13000 26000 2699 2917 5616StrawberryRd_EB 900 900 1800 0 24000 24000 0 0 0RaspberryRd_WB 11850 12014 23864 12400 12400 24800 8343 5415 13757Screenline Totals 33756 34848 68604 65300 89300 154600 29303 27503 56806

0

5000

10000

15000

20000

25000

30000

35000

DimondBlvd_WB 100thAve_WB KlattRd_WB StrawberryRd_EB RaspberryRd_WB

Counts

Model

SCREENLINES


(2020) South End of Study Area – New Seward Highway

ABLKNME ABCOUNT BACOUNT TOTCOUNT ABCAP BACAP TOTCAP ABVOL BAVOL TOTVOLNewSewardHwy_SB 4480 4480 8960 7400 7400 14800 4687 4067 8754Screenline Totals 4480 4480 8960 7400 7400 14800 4687 4067 8754

8650

8700

8750

8800

8850

8900

8950

9000

NewSewardHwy_SB

Counts

Model

APPENDIX B

TBG121212233731SCO B-1

Appendix B: Economic Forecast Disaggregation

APPENDIX B

TBG121212233731SCO B-2


Anchorage and Matanuska-Susitna Borough 2035 Land Use Allocation and Forecast

TBG072710184349ANC Page 1

Memo Federal Project No: NH-000S(588)

AKSAS Project No. 58544

To: Jim Childers, PE, ADOT&PF Project Manager

From: CH2M HILL Project Team Project: H2H

CC: John McPherson, AICP, Consultant Project Manager Project File

Subject: Anchorage and Matanuska-Susitna Borough 2035 Land Use Allocation and Forecast

Date: July 29, 2010 Job No: 80510

1 INTRODUCTION

The purpose of this technical memorandum is to document the methodology used to allocate and forecast Anchorage and Matanuska-Susitna Borough (MSB) land use for use as input into the transportation demand model. Land use variables discussed in this memorandum are population, households, and employment.

1.1 Project Background

The Seward Highway to Glenn Highway Connection project, referred to as the “H2H” project in this document, is being undertaken by the Federal Highway Administration (FHWA), the Alaska Department of Transportation & Public Facilities (ADOT&PF), and the Federal Transit Administration (FTA) in cooperation with the Municipality of Anchorage (MOA) and the U.S. Army Corps of Engineers (USACE). The purpose of the project is to reduce congestion by improving mobility and access for people and goods that use the arterial connection between the Seward and Glenn Highways.

1.2 Transportation Demand Model

The transportation demand model (the model) used for the H2H project is a regional model that incorporates the MSB. The model is a computer program designed to simulate future travel by comparing the demand for transportation (the need to travel generally expressed in terms of travel generation and attraction) to the supply (the available transportation network). The model documentation is being completed concurrently with this technical memorandum. The base year for the model is 2007, and 2035 is the forecast year.



1.3 Land Use Variables

The geographic location of population and employment plays an important role in determining travel demand. As a result, one of the most critical inputs into the model is the geographic distribution of land use and socioeconomic variables such population, households, employment, average income, workers per household, children per household, and household size. This technical memorandum focuses specifically on the allocation and forecast for population, households, and employment. The travel demand model breaks down employment into 13 employment categories, as follows: Agriculture, forestry, and fisheries

Mining

Construction

Manufacturing

Transportation, communications, and utilities

Wholesale trade

Retail trade

Government

School employment

Finance, insurance, and real estate

University

Services, except health services

Health services

The model uses land use variables based on transportation analysis zones (TAZs) that summarize land use variables into geographic areas that are closely related to, but more refined than, census tracts. TAZs are used by the model for the actual calculation of travel generation and attraction. Figure 1-1 shows an overview of TAZs and regional areas used in the model.

2 BASE YEAR (2007) LAND USE ALLOCATION

The first step in the land use allocation was to gather available data from existing models. The Anchorage Metropolitan Area Transportation Solutions (AMATS) Long Range Transportation Plan update was prepared in 2007 and serves as the base year allocation for AMATS area TAZs. The model used for the Knik Arm Bridge and Toll Authority (KABATA) environmental impact statement (EIS), published in December 2007, used zone information for 2005 for the MSB.



Figure 1-1 Regional Transportation Demand Model Boundaries

It should be noted that the AMATS Traffic Analysis Zone (TAZ) structure was modified along the H2H alternative project corridors in order to provide a more refined travel projection for the study area. This required distributing the 2007 socioeconomic data in the original TAZs between the new split zones. Also of note is the methodology used to develop the 2007 MSB socioeconomic data. Since the MSB only had socioeconomic data available for the years 2005



and 2015, the 2007 base year socioeconomic data was developed through an interpolation of the MSB TAZ data for the years 2005 and 2015. Table 2-1 shows the base year control totals for population, households, and employment for the MOA and the MSB. The base year employment by category is discussed later in this technical memorandum and is shown in Table 2-2.

Table 2-1. 2007 Regional Land Use Allocation for Population, Households, and Total Employment

Category MOA MSB Regional Total

Population 280,546 72,707 353,250

Households 102,503 27,598 130,101

Total Employment 163,704 28,988 192,692

Table 2-2. 2007 Regional Employment Allocation by Category


Agriculture, forestry, and fisheries

239 60 299

Mining 2,787 31 2,818

Construction Total 7,831 3,569 11,400

Manufacturing 2,705 232 2,937


9,083 850 9,933

Wholesale trade 4,899 1,278 6,177

Retail trade 21,916 4,489 26,405

FIRE 7,224 4,527 11,751


47,655 6,859 54,514

Government 16,743 3,280 20,023

School employment 5,500 900 6,400

University 745 35 780

Health services 9,788 2,208 11,996

3 Regional Land Use Allocation and Forecast Control Totals

The second step in the land use allocation and forecast was to allocate population, households, and employment to regional control totals for the forecast year. This section discusses available land use forecasts and validates regional control totals developed for the H2H project by the Institute for Socioeconomic Research (ISER) of the University of Alaska Anchorage (UAA). The primary source of population, household, and employment forecast control totals come from ISER, which developed a project specific forecast for 2035 for the MOA and the MSB.



3.1 Institute of Social and Economic Research Regional Forecasts

ISER has been conducting population, household, and employment forecasts for the State of Alaska for more than 35 years, and is widely accepted and used in Alaska for long-range planning. AMATS, the metropolitan planning organization responsible for development of the Long-Range Transportation Plan for Anchorage, has used ISER forecasts as the basis of its transportation demand model projections since the early 1990s. ISER forecasts use an economic scenario developed by ISER in consultation with other Alaska researchers in the private and public sectors. Each economic scenario assumes various levels of future industry activity within the state, national variables, state fiscal policy variables, and other exogenous factors that are expected to influence the future pattern of economic and demographic trends. Each of the ISER forecasts is broken down into low, base, and high projections. Typically, the base projections are used in transportation modeling, because they are assumed the most likely scenario to occur.

3.1.1 Prior ISER Regional Forecasts

ISER updates its forecasts every 3 to 5 years; the forecasts for 1998, 2003, and 2005 are used for comparison purposes in this technical memorandum. The 1998 ISER forecast is from the January 1998 ISER report prepared for the Alaska Department of Transportation titled “Population, Employment, and Income Projections for Alaska Census Areas.” Table 3-1 shows population and employment (wage and salary) forecasts for 2000 to 2025 from the 1998 ISER forecast.

Table 3-1. 1998 ISER Population and Employment Forecast

Year

Population Wage and Salary Employment

MOA MSB Region MOA MSB Region

2000 265.80 55.87 321.67 124.74 11.12 135.86

2005 279.43 62.26 341.69 127.73 11.93 139.66

2010 303.29 70.94 374.23 137.64 13.54 151.18

2015 334.46 81.61 416.07 150.52 15.63 166.15

2020 365.73 92.92 458.65 164.00 17.94 181.94

2025 399.32 105.70 505.02 178.94 20.63 199.57

The 2003 ISER forecast is from the June 2003 ISER report titled “Economic Projections for Transportation Planning in Southcentral Alaska 2000-2035.” Table 3-2 shows population, households, and employment (wage and salary) projects for 2000 to 2035 from the 2003 ISER forecast.



Table 3-2. 2003 ISER Population, Household, and Employment Forecast

Year

Population Households Wage and Salary Employment

MOA MSB Region MOA MSB Region MOA MSB Region

2000 259.90 59.40 319.30 94.80 20.50 115.30 131.70 12.10 143.80

2005 275.70 72.50 348.20 101.10 25.20 126.30 138.60 14.50 153.10

2010 285.00 81.20 366.20 104.80 28.30 133.10 141.80 15.80 157.60

2015 304.80 92.60 397.40 110.80 32.00 142.80 149.00 17.70 166.70

2020 332.20 108.20 440.40 122.30 37.80 160.10 160.40 20.60 181.00

2025 361.20 126.60 487.80 134.90 44.90 179.80 173.00 24.20 197.20

2030 372.10 141.40 513.50 140.10 50.50 190.60 178.00 27.10 205.10

2035 379.20 155.60 534.80 143.20 55.80 199.00 182.70 30.20 212.90

The 2005 ISER forecast is from the September 2005 ISER report titled “Memorandum on the Economic and Demographic Impacts of the Knik Arm Bridge.” Table 3-3 shows population and employment (wage and salary) projects for 2005 to 2030 as forecast in the 2005 ISER forecast.

Table 3-3. 2005 ISER Population and Employment Forecast

Year

Population Wage and Salary Employment

MOA MSB Region MOA MSB Region

2005 285.72 72.70 358.42 148.38 16.46 164.84 2010 297.29 96.04 393.33 154.42 23.77 178.19 2015 321.11 124.56 445.67 161.21 31.40 192.61 2020 322.42 144.43 466.85 161.63 37.73 199.36 2025 334.38 173.51 507.89 168.19 46.90 215.09 2030 345.45 203.76 549.21 175.09 56.72 231.81

3.1.2 2009 ISER Regional Forecast for the H2H Project

The ISER H2H was extended to forecast expected regional population, households, and employment for 2035. ISER produced a final forecast report titled “Economic and Demographic Projections for Alaska and Greater Anchorage” for HDR Alaska, Inc., in December 2009 (referred to in this memorandum as “the 2009 ISER forecast”). Table 3-4 shows the population, household, and employment regional control totals from the 2009 ISER forecast. The zone structure in the travel demand model does not encompass the entire region that ISER forecasts, resulting in adjustments to the 2009 ISER forecast prior to using the new 2035 ISER population, household, and employment forecast in the transportation demand model.



Table 3-4. 2009 ISER Population, Household, and Employment Forecast

Year

Population Households Wage and Salary Employment

MOA MSB Region MOA MSB Region MOA MSB Region

2000 260.30 59.30 319.60 130.90 12.40 143.30

2005 277.90 74.00 351.90 101.80 25.70 127.50 143.40 16.80 160.20

2010 289.20 80.30 369.50 109.20 28.80 138.00 147.60 18.80 166.40

2015 288.80 96.40 385.20 110.30 34.60 144.90 151.40 22.80 174.20

2020 314.50 117.20 431.70 120.60 42.70 163.30 161.50 29.40 190.90

2025 333.70 153.60 487.30 128.20 56.10 184.30 172.90 33.80 206.70

2030 343.10 169.00 512.10 132.40 62.00 194.40 176.10 43.60 219.70

2035 351.30 170.80 522.10 136.60 63.10 199.70 177.60 46.40 224.00

3.1.2.1 Population

The 2009 ISER population for 2030 shows an overall regional (MOA and MSB) population reduction of 550,500 to 512,100 from the previous 2005 ISER forecast. The majority of this reduction is located in the MSB. As stated in the 2009 ISER report, the reasons for the reduction in forecast population growth are as follows:

A delay in the assumed construction schedule for the Knik Arm Bridge (in the later forecast) results in a decrease in the overall growth in the MSB, particularly in the Knik Arm area closest to the new bridge.

Adverse impact on population caused by recession beginning in early in 2009, presuming that both the AMATS and MSB areas will see slowed population growth; however, the slowed growth will be less severe in Anchorage.

3.1.2.2 Household

The reduction in the number of households forecast for the region follows the same trends as the population forecast.

3.1.2.3 Total Employment

The reduction in the regional employment total is not as severe as the reduction in the population and households. As with population, the majority of the change occurs in the employment forecast for the MSB. As stated in the 2009 ISER report, the reasons for the change in forecast employment distribution between the AMATS and the MSB areas are as follows: A larger share of the basic job growth, mostly military, is located in the AMATS area

(primarily Anchorage) rather than in the MSB when compared to 2005 ISER forecast.

The Knik Arm Bridge construction occurs later in the 2009 ISER forecast than in the 2005 ISER forecast, and thus delays the movement of basic jobs from AMATS to the MSB.



The number of basic jobs that shift from AMATS to the MSB each year has been reduced based on recent historical experience. Therefore, the rate of growth in the number of commuters who live in the MSB but work in AMATS is also delayed.

Adverse impact on population due to recession beginning in early in 2009, presuming that both the AMATS and MSB areas will experience drops in employment in 2009 and 2010.

4 ADJUSTMENTS TO 2009 ISER REGIONAL FORECAST

Step 3 in the process of allocating and forecasting population, households, and employment was adjusting the 2009 ISER forecast to account for the modeled area related to the forecast area. These adjustments were required prior to using the new 2035 ISER population, household, and employment forecast in the transportation demand model.

4.1 Population and Household Adjustments

The first adjustment is for the population and household for both AMATS and the MSB areas, and is required to reflect the percentage of the total population and households that reside within the model area. The AMATS area adjustment is a subtraction of the population and households that are expected to live in the Turnagain Arm/Girdwood area, which is outside the model area. The MSB area adjustment is reduction of population and households by 6.28 percent (or a multiplier of 93.72 percent of the total forecast) ,which, according to the 2000 U.S. Census, is the percentage of the MSB population and households residing in the model area. The assumptions made are the following: The AMATS area population in the Turnagain Arm/Girdwood area is expected to grow at an

annual growth rate of 5.6 percent, resulting in a 2035 forecast for the Turnagain Arm/Girdwood population of 5,569 and a household estimate of 2,542.

The MSB area ratio of population and households within the modeled area compared to the total MSB population and households would remain the same as in the 2000 Census, 93.72 percent.

Subtracting the Turnagain Arm/Girdwood area population from the population control total for the AMATS region resulted in a population forecast of 345,731 (the 2009 ISER forecast for 2035 of 351,300 minus the assumed Turnagain Arm/Girdwood area growth of 5,569). The population control total for the MSB region would, therefore, be 160,073 (the 2009 ISER forecast for 2035 of 170,800 multiplied by 93.72 percent). Subtracting minus the Turnagain Arm/Girdwood area population from the household control total for the AMATS region resulted in a household forecast of 134,058 (the 2009 ISER forecast for 2035 of 135,600 minus the assumed Turnagain Arm/Girdwood area growth of 2,542). The household control total for the MSB region would, therefore, be 59,165 (the 2009 ISER forecast for 2035 of 63,172 multiplied by 93.72 percent). Table 4-1 summarizes the adjusted population and household control totals for the modeled area using the 2009 ISER forecast for the Anchorage, Eagle River, and MSB areas.



Table 4-1. 2035 Regional Population and Household Control Totals

MOA MSB Region

2035 Population 345,738 159,052 504,790

2035 Households 134,229 59,165 193,394

4.2 Employment Adjustments

The second adjustment was for employment for both the AMATS and the MSB areas. No data are currently available for use in determining how much of total MSB employment lies within the modeled area. The two data sources used (the AMATS model, and the MSB model) differ in their treatment of self-employment, resulting in a required adjustment to the 2009 ISER forecast. The assumptions made are the following: An insignificant amount of the total MSB employment is located outside the model area.

In calculating the 2035 MOA self-employment total, the ratio of self-employment to wage-and-salary employment would not change from 2007 to 2035.

No growth in employment is located in Turnagain Arm/Girdwood area.

Two major projects involving the expansion of the Alyeska Ski Resort to the Winner Creek area and the development of a golf course are on hold, understanding that a reassessment of employment growth in the Turnagain Arm should be conducted if one or both of these projects is completed.

Total employment excludes uniformed military personnel. Uniformed military employment (estimated to be 14,343 in 2035) is a special generator in Anchorage and, therefore, does not need to be allocated. MSB uniformed military (estimated to be 631 in 2035) is an insignificant part of the total employment and, therefore, was not included in total employment.

Total employment control totals by category and total employment, after adjustments discussed in this section for the MOA and MSB regions, are shown in Table 4.2. The development of these totals is discussed in Sections 4.2.2 and 5.3.

Table 4-2. 2035 Regional Employment Control Totals


Agriculture, forestry, and fisheries 215 680 895

Mining 2,512 463 2975

Construction total 8,827 5,785 14612

Manufacturing 2,434 2,810 6244

Transportation, communications, and utilities 10,418 2,856 13274

Wholesale trade 6,404 1,670 8074

Retail trade 28,627 12,159 40786


TBG072710184349ANC Page 10

Table 4-2. 2035 Regional Employment Control Totals


Agriculture, forestry, and fisheries 215 680 895

Mining 2,512 463 2975

FIRE 9,443 6,577 16020

Services, except health services 62,298 12,355 74653

Government 22,043 6,484 28527

School employment 7,095 4,314 11409

University 964 138 1102

Health services 12,779 6,470 19249

Total employment* 208,203 63,714 271,917

*Note: Special generators are included in Total Employment

4.2.1 Self -employment Adjustments

The AMATS model includes wage-and-salary and self-employed employment, and the MSB model includes only wage-and-salary employment. The America Community Survey (2005 – 2007) states that the total number of self-employed in the MOA was 10,538, representing 7.12 percent of the Alaska Department of Labor (ADOL) estimate of 2007 wage-and-salary employment of 148,033. Self-employment in the MSB is a substantial share (40 percent) of the total employment (as noted in the 2009 ISER forecast report). Thus, MSB self-employment needed to be adjusted to avoid underestimation of total employment in the MSB. The total employment control total (wage-and-salary and self-employed) for the AMATS region is 189,879 (calculated by multiplying the 2009 ISER wage-and-salary forecast for 2035 of 177,260 by 7.12 percent and adding it to the 2009 ISER wage-and-salary forecast for 2035 of 177,260). The total employment control total (wage-and-salary and self-employed) for the MSB region is 63,223 (calculated by adding the 2009 ISER self-employed MSB estimate of 16,400, the 2009 ISER wage-and-salary estimate of 46,400, and the MSB unidentified employment estimate of 423).

4.2.2 Employment Category Adjustments

The employment in the travel demand model is described in Section 1; however, ISER does not use the same employment categories as the travel model uses. ISER employment is broken down into basic, government, and support sectors, and forecasts employment growth rates by these three sectors for the MOA and the MSB are shown in Table 4-3. To apply the ISER sector growth rates shown in Table 4-1 to the base year (2007) categorized employment data used by the travel demand model, an equivalency table for ISER employment sectors and travel demand model employment categories was developed (see Table 4-4).


TBG072710184349ANC Page 11

Table 4-3. 2009 ISER Forecast Employment Growth by Sector (2008 to 2035)

Employment Sector MOA MSB

Basic employment -6.7% 718.9%

Government 22.1% 51.5%

Support 21.1% 74.5%

Table 4-4. ISER Employment Sectors and Travel Demand Model Categories Equivalency

ISER Employment Sector Travel Demand Model Employment Categories

Basic Agriculture, forestry & fisheries Mining Construction Manufacturing Transportation, communications & utilities

Government: local, state, and federal Government School Employment University

Support Wholesale trade Retail Trade Finance, insurance, and real estate Services, except health services Health services

Special consideration was necessary prior to finalizing the employment category control totals. First, growth rates of several of the AMATS model employment categories were recognized as being distinct, requiring a separate growth rate analysis. These included school employment, which was assumed to grow in proportion to the growth in school age population, and university employment was assumed to grow at the same rate as student growth.

4.2.3 Special Generator Employment Adjustments

Adjustments to the employment category totals in Anchorage were necessary in order to avoid double-counting special generator employment. Table 4-5 shows the estimated 2035 employment accounted for by the eight special generators located in the region.

Table 4-5. 2035 Special Generator Employment by Employment Category

Special Generator Employment Category Total Employment

Elmendorf Government 17,634

Fort Richardson Government 4750

Port of Anchorage Transportation, communications, and utilities

762

Ted Stevens Anchorage Transportation, communications, and 4,475


TBG072710184349ANC Page 12

Table 4-5. 2035 Special Generator Employment by Employment Category

Special Generator Employment Category Total Employment International Airport utilities

Government Wholesale trade Retail trade Services, except health services

1,122 51

369 486

Anchorage Library Government 89

Alaska Native Hospital Health services 1,633

Columbia Regional Hospital Heath services 1,129

Providence Hospital Health services 3,051

Alaska Pacific University University 349

Matanuska-Susitna Borough Regional Hospital

Health services 953

University of Alaska Anchorage University 2,333

5 SUBAREA POPULATION, HOUSEHOLD, AND EMPLOYMENT ALLOCATION

The fourth step in the 2009 ISER population and employment forecast allocation of population, household, and employment between the MSB and the MOA was to allocate the regional forecast into subareas. This was an intermediate step between the ISER regional forecast and the final TAZ allocation. The revised MSB Comprehensive Plan, which was published by the MSB Planning and Land Use Department in 2005, does not contain this type of guidance. As a result, the methodology used to allocate population, households, and employment to the MSB skips this fourth step and disaggregates the ISER regional forecast directly to the TAZ level (Step 5). Step 4 was required within the AMATS region to ensure that the final allocation to the TAZ level correlated with the land use policies contained in the adopted comprehensive land use plans. The guiding comprehensive plans are the Anchorage Bowl Comprehensive Plan (last updated in 2002 and referred to as “the Anchorage Comprehensive Plan” in this technical memorandum) and the Chugiak-Eagle Comprehensive Plan (published by the MOA, last updated in 2006, and referred to as the “CER Comprehensive Plan” in this technical memorandum). These policies provide specific guidance regarding the share of growth by defined subareas of the community. Figure 5-1 shows the MOA subareas: Chugiak-Eagle River, Northwest Anchorage, Northeast Anchorage, Southwest Anchorage, Central Anchorage, and South Anchorage. Table 5-1 is a summary of the forecast population, households, and total employment for the MOA subareas; these numbers are discussed in detail in Sections 5.1, 5.2, and 5.3.


TBG072710184349ANC Page 13

Figure 5-1. MOA Subareas

Table 5-1. 2035 Population, Household, and Employment Control Total by MOA Subarea

Category

Chugiak-Eagle River

Northwest Anchorage

Northeast Anchorage

Central Anchorage

Southeast Anchorage

Southwest Anchorage MOA

Population 65,021 57,730 85,697 53,866 34,086 49,338 345,738

Households 22,346 27,156 33,000 21,610 11,948 18,169 134,229

Employment 10,164 81,129 42,464 33,377 3,128 19,071 208,203*

*Note: Military employment of 18870 is included in a separate subarea.

5.1 Population Subarea Allocation

The Anchorage Bowl and Chugiak-Eagle River have separate comprehensive land use plans. Both plans project population to grow at a much faster rate in Chugiak-Eagle River than in the Anchorage Bowl because of its larger supply of vacant land. The CER Comprehensive Plan projects population growth rate based on a percentage of the overall MOA growth rate. Thus, the first step in the subarea disaggregation process was to estimate the share of the total MOA growth that belongs to Chugiak-Eagle River. The Anchorage Bowl Comprehensive plan does not specifically allocate population by subarea, but does allocate number of households by subarea. The number of households in each TAZ can be used to estimate Anchorage Bowl subarea population by using the average household size.


TBG072710184349ANC Page 14

5.1.1 Chugiak-Eagle River

The CER Comprehensive Plan states that the population growth in Chugiak-Eagle River is expected to contain an estimated 15 percent of the total MOA population by the year 2025. Since the Plan only goes out to the year 2025, it is necessary to extrapolate the growth rate another 10 years in order to arrive at an estimate for 2035. According to the Plan, the percentage of the total MOA population living in Chugiak-Eagle River is projected to increase by 1 percent between 2020 and 2025. If this growth rate is applied to each 5-year period (2025 to 2030 and 2030 to 2035), the total percentage of the MOA population living in Chugiak-Eagle River would be approximately 17 percent. Given that the total 2035 ISER population forecast for the MOA is 351,300, the 2035 Chugiak-Eagle River population would, therefore, be 59,721. However, refinement of population allocation within the MOA utilizing existing (2007) population allocation and known future development, the control total for Chugiak-Eagle River population in 2035 is 65,021, included in Table 5-1.

5.1.2 Anchorage

Population was not specifically allocated by subarea but was calculated for each TAZ by multiplying the average household size (received from the 2000 census) by the number of households in each respective TAZ. The resulting population by TAZ was then adjusted to the overall Anchorage Bowl control totals. Although not discussed in this section, the summary of population control total by Anchorage bowl subarea for 2035 is shown in Table 5-1.

5.2 Households

The CER Comprehensive Plan does not explicitly provide a household forecast, but it does provide a means of calculating the future average household size. On the other hand, the Anchorage Comprehensive Plan provides specific guidance regarding the allocation of future household growth to five subareas of the Anchorage Bowl, as shown in Figure 5-1.


According to the CER Comprehensive Plan, the average household size in Chugiak-Eagle River decreased from 3.08 persons in 1990 to 2.9 persons in 2000 and is expected to continue decreasing in the future. The 2009 ISER population and household forecast for the entire MOA shows an overall drop in household size from 2.74 persons in 2007 to 2.57 persons in 2035. This represents a 6 percent decrease in household size or 0.2 percent per year. Applying a decrease of 0.2 percent per year to the current 2.87 persons per household size for Chugiak-Eagle River, the 2035 household size would be expected to be 2.67. Dividing the estimated 2035 CER population estimate of 59,721 by 2.67 yields a total 2035 household estimate of 22,346, as shown in Table 5-1.

5.2.2 Anchorage

The growth allocation table on page 59 of the Anchorage Bowl Comprehensive Plan was intended to supplement the Land Use Policy Map (page 50 of the Plan) and to provide additional land use guidance regarding where future residential growth will take place. In order to facilitate


TBG072710184349ANC Page 15

comparison between the model and Plan allocation of households within the Anchorage Bowl, it was decided to develop subarea control totals using the same subarea geographic boundaries. In 2004, the MOA Planning Department prepared a buildable lands inventory and capacity analysis (MOA, 2007) to evaluate the ability of each of the subareas to accommodate the growth allocated by the Comprehensive Plan. The Planning Department began by updating the vacant land inventory and re-estimating the re-developable land and then proceeded to estimate the total residential build-out capacity of each subarea, based upon current assumptions about land use density. This preliminary analysis found that the northwest subareas would have a major deficit of between 5,000 and 7,000 households when compared to the Comprehensive Plan allocation. On the other hand, the northeast subarea would have a minor surplus of between (300 and 2,500 housing units). In order to correct these inconsistencies, the Planning Department has recommended reducing the overall residential holding capacity in the Northeast subareas by approximately 1,000 housing units and increasing the residential capacity in mixed-use commercial areas near Downtown and Midtown in the Northwest Subarea by 3,500 housing units. The resulting subarea capacities were further reduced by subtracting the housing development that has occurred since the data used in the MOA buildable lands inventory and capacity report was collected (2004) and 2007 (the model’s base year). The percent of total Anchorage Bowl capacity by subarea was then multiplied by the Anchorage Bowl household growth control total of 22,517 to calculate the 2035 housing growth allocation by subarea, as summarized in Table 5-2. The 2035 Anchorage Bowl household growth control total of 22,517 was calculated by subtracting the Chugiak-Eagle River household control total (22,346 from Section 5.2.1) and the Turnagain Arm household growth control total (2,542 from Section 4.1) from the overall MOA household control total (136,600 from the 2009 ISER forecast) and subtracting this total from the 2007 Anchorage Bowl household figure of 134,229.

Table 5-2. 2035 Anchorage Bowl Household Allocation by Subarea and Capacity

MOA Buildout Capacity

(2004)

Household Permits

(2004 – 2006)

MOA Buildout Capacity

(2007)

2007 Capacity as Percent of Anchorage

Bowl

Household Allocation by

Subarea (2035)

Northwest 8594 426 8168 31.95% 7113

Northeast 6405 881 5524 21.61% 4810

Central 5481 914 4567 17.86% 3977

Southwest 4463 766 3697 14.46% 3220

Southeast 4334 433 3901 15.26% 3397

A comparison of the percentage subarea allocation using the above methodology to the Comprehensive Plan housing allocation ranges by subarea is shown in Table 5-3. The new subarea allocation percentages are similar to the original subarea allocation estimates provided in the Comprehensive Plan. The new northwest subarea capacity is slightly higher than that originally projected in the Comprehensive Plan, while the southwest and central subarea


TBG072710184349ANC Page 16

capacities are lower than originally projected. This is due to a lag in new housing starts (see “2004-2006 Household Permits” column in Table 5-2) in the northwest subarea compared to the other subareas. As a result, capacity decreased at a more rapid rate for the southwest and central subareas than for the northwest, in turn decreasing these areas’ percentage of the total remaining Anchorage Bowl capacity. While the percentage allocation by subarea is slightly different from the original Comprehensive Plan estimate, it made sense to use the new subarea allocation estimates that take into account the land use changes that have occurred since those estimates were made in 1998.

Table 5-3. Comparison of Comprehensive Plan Subarea Allocation

Subarea 2007 Capacity as Percent of

Anchorage Bowl Comprehensive Plan Subarea Allocation

Northwest 31.95% 26.67% (7,000 - 9000 households)

Northeast 21.61% 20.0% (5,000 - 7,000 households)

Central 17.86% 20.0% (5,000 - 7,000 households)

Southwest 14.46% 16.7% (4,000 - 6,000 households)

Southeast 15.26% 16.7% (4,000 - 6,000 households)

5.3 Employment

The CER Comprehensive Plan provides guidance about future employment, based on a percent of population. The Anchorage 2020, Anchorage Bowl Comprehensive Plan, published by the MOA in February 2001, did not identify subarea employment allocations in the same manner as it allocated households. As a result, the methodology used to allocate employment to the Anchorage Bowl subareas skips this intermediate subarea allocation step and disaggregates the Anchorage Bowl employment forecast directly to the TAZ level (Step 5). Table 5-4 summarizes employment by category for AMATS region subareas and is discussed in Sections 5.3.1 and 5.3.2.


The CER Comprehensive Plan estimates that employment will equal 15 percent of the Chugiak-Eagle River population in 2025, increasing from 13 percent of the population in 2004. For the purposes of this allocation study, it was assumed that employment as a percent of population will continue to increase and will be 16 percent of the population by 2035. Since the 2035 population of Chugiak-Eagle River is estimated at 65,021, the total employment would be 11,054 in that same year. However, further refinement of employment based on existing allocation and known future development, a control total for the Chugiak-Eagle River total employment is 10,164, as shown in Table 5-1.


TBG072710184349ANC Page 17

Table 5-4. Employment allocation by Subarea for Anchorage, Chugiak-Eagle River

Employment Category Chugiak-

Eagle River Northwest Anchorage

Northeast Anchorage

Central Anchorage

Southeast Anchorage

Southwest Anchorage MOA

Agriculture, forestry, and fisheries

1 51 2 106 0 55 239

Mining 1 1092 734 554 1 130 2,787

Construction total 395 2111 577 4185 480 498 7,831

Manufacturing 47 882 664 655 19 167 2,705


209 5117 1543 1617 94 1690 9,083

Wholesale trade 49 1945 317 3781 25 286 4,899

Retail trade 1,409 10143 6935 7862 12 1198 21,916

FIRE 146 6330 1047 1263 76 569 7,224


1,338 33291 8475 10268 1455 5366 47,655

Government 556 14089 3860 1717 137 1228 16,743

School employment 272 1116 1915 608 764 994 5,500

University 51 348 565 0 0 0 745

Health services 297 3729 7345 751 66 391 9,788

Total employment* 10,164 81,129 42,464 33,377 3,128 19,071 208,203

*Note: Military employment of 18870 is included in a separate subarea.


TBG072710184349ANC Page 18

The AMATS Transportation Demand Model requires that the total employment be broken down into 13 separate employment categories, each of which has a different trip attraction rate. As a result, it is not sufficient to allocate only total employment between Chugiak-Eagle River and the Anchorage Bowl. It is also necessary to disaggregate total Chugiak-Eagle River employment into the individual employment categories, particularly because the Chugiak-Eagle River employment composition is so different from that of the Anchorage Bowl. Control totals for 2035 categorical employment in Chugiak-Eagle River are shown in Table 5-4.

5.3.2 Anchorage

Employment by employment category for the Anchorage Bowl was derived by simply subtracting the Chugiak-Eagle River employment by employment category as shown in Table 5-4 from the total MOA employment by employment category (Table 5-1).

6.1. Municipality of Anchorage TAZ Allocation

The fifth and final step involved disaggregation of the subarea totals into the TAZs. This section documents and summarizes these steps and the resulting forecasts and allocations.

6.1.1 Households

The starting point for the MOA (including Chugiak-Eagle River) allocation is the year 2007. Obviously, development did not stop between 2007 and 2009 (the date that this update was completed). As a result, it was necessary to fill in this 2-year gap with information about recent development activity. In the past, AMATS has relied on MOA permit data to provide the number and locations of new housing units. Each new permit must then be geocoded in order to identify the TAZ to which it should be assigned. Permit data from June 2006 to December 2008 were download and geocoded to the correct TAZ, and the total was added to the existing 2007 database to produce an updated 2009 estimated number of households by TAZ. A total of 959 housing permits were issued during that time period, with 28 issued outside the model area in Turnagain Arm communities. (It is assumed that all of the household permits issued during that timeframe were built.) The next step in allocating the subarea household control totals to the TAZ level was to allocate the planned short-term housing development (pipeline development). While it is sometimes difficult to forecast how many and where housing development will occur in the long term, it makes sense to take into account the existence of large subdivisions that are either actively being developed or were recently approved by the MOA Platting Board. Recently, new housing starts have slowed considerably within the MOA. As a result, only four developments were included in the pipeline development assumptions used in this housing allocation effort (see Table 6-1).


TBG072710184349ANC Page 19

Table 6-1. Pipeline Housing Development Assumptions

Name Number of Units TAZ

Sand Lake Gravel Pits 850 397

Abbott Loop Church 64 441

Weidner Midtown Apts. 330 299

Lazy Mtn. Town Square 220 632

Special consideration was also necessary with respect to the allocation of new housing units in the Downtown and Midtown districts. The majority of these districts are zoned commercial. Moreover, little if any new housing development has occurred in these districts during the past 20 years. As a result, it is not possible to utilize a trend analysis of past housing development to predict future housing development in these areas. Nevertheless, the Anchorage Bowl Comprehensive Plan states (page 51) that “higher density mixed-use development that includes residential uses would also be encouraged within the employment center core (Midtown and Downtown).” According to the April 2007 MOA Planning Department report, Draft Land Capacity of the Anchorage Bowl to Accommodate Projected Growth, the number of housing units that could be built in the commercial areas of Downtown is estimated to be around 75 per year. Multiplying this figure by 26 years (2009 to 2035) resulted in an estimate of 1,950 new Downtown housing units. In lieu of a more detailed housing market analysis, this figure was adopted as the housing growth control total for Downtown. In order to allocate the Downtown housing control total to the TAZ level, additional assumptions needed to be made regarding the location and density of these future housing developments. The newly adopted Anchorage Downtown Comprehensive Plan (2007) provides direction as to where and at what density new housing might be built in Downtown Anchorage. The Plan divides the Downtown into several districts, including a core district and five periphery districts surrounding the core. Using the densities provided by the Downtown Comprehensive Plan, a potential capacity for each downtown TAZ was calculated based on the total acreage of the TAZ times the final dwelling units per acre (DU/A) shown in Table 6-2. The calculated TAZ housing capacity as a proportion of the total downtown capacity was then multiplied by the downtown housing control total (1950) to arrive at the estimate housing growth for each TAZ.

Table 6-2. Downtown Subdistrict Densities

Range DU/A

Midpoint DU/A

Final DU/A*

Barrow Street 15 to 30 22.5 22.5

Pioneer Slope 15 to 30 22.5 22.5

Park Strip North 20 to 50 35 35

East Avenues 20 to 60 40 20

Legal/Office 20 to 60 40 20


TBG072710184349ANC Page 20

The Midtown Employment District does not have the advantage of having an adopted plan. A Public Hearing Draft of the Midtown District Plan has been prepared but is awaiting additional work prior to being submitted to the MOA Planning and Zoning Commission for review. The public hearing draft does, however, contain a housing market study, which states that the Midtown District should attempt to retain its current 5 percent share of the MOA housing market. Using this figure as a guide, the Midtown housing stock should grow by 1,578 by 2035 (31,562 times 0.05). The 2007 Draft Land Capacity of the Anchorage Bowl to Accommodate Projected Growth report also provides some direction. That report estimated that the number of housing units that could be built in the commercial areas of Midtown is approximately 25 units per year ,or a total of 650 over 26 years. The draft Midtown Plan identifies Denali Center as one of the prime areas for future mixed-use residential development. The 650 new housing units to be allocated to commercial areas of Midtown were, therefore, allocated to TAZs 291 and 271 within the Denali Subdistrict. An additional 1,232 new housing units were allocated to the existing residentially zoned areas of Midtown (primarily in the western portion of the district), for a total of 1,882 new housing units in Midtown. The Weidner Midtown Apartment proposal had previously been added to TAZ 299 as one of the pipeline developments. Thus, the total Midtown housing allocation was estimated at 2,272, substantially above the Midtown housing market projection of 1,578. The final step in the MOA housing allocation procedure was to allocate the future housing units remaining after subtracting the permits, pipeline development, and downtown and midtown allocations from the subarea control totals. These housing units were allocated to TAZs based on the development potential for each TAZ remaining after taking into account the permitted and pipeline development calculated in the previous steps. The new development potential, represented as a percentage of the total subarea housing development, was then multiplied by the total remaining housing allocation for that subdistrict.

6.1.2 Population

Population by TAZ was derived from the household estimate. It was calculated by multiplying the number of households by the average household size (AHHS). The resulting population estimate was then adjusted to the subarea population control totals shown in Table 5-1.

6.1.3 Employment

Similar to the household database, the 2007 employment database needed to be updated to take into account the new commercial permits issued between 2007 and 2009. It is a little more complicated, however, to derive employment estimates from permit files than to derive household numbers. The MOA permit database contains information on the general type of employment and square footage of developments but does not contain information on number of employees. As a result, it was necessary to convert the square footage of development (provided by the permit data) into an estimate of the number of employees by employment category. The conversion estimates (see Table 6-3) primarily relied on the Institute of Transportation Engineers (ITE) as the source for the number of employees per square feet. Each new commercial permit, along with the estimated number of employees was geocoded in order to identify the TAZ to which it should be assigned.


TBG072710184349ANC Page 21

Table 6-3. Employment/Building Size Conversion Ratios

Employment Type Number of Employees per 1000 Square Feet

Big Box Store 1.53

Medical Offices 4.8

Industrial Uses 0.15

Warehouse 0.3

Retail 1.8

Banks 2.1

Office 3.4

Restaurant 7.5

An attempt was made to ascertain the likely near-term, commercial-related pipeline development, geocode it, and then convert the estimated square footage into number of employees by employment category. Once again, given the current slow growth environment in Anchorage, the amount of estimated near-term commercial development is fairly limited. The employment by TAZ was calculated using the same square footage to employee conversion factors shown in Table 6-3. Table 6-4 lists the pipeline employment by TAZ within the MOA.

Table 6-4. Pipeline Employment by TAZ

Name of Development TAZ Retail

Employment FIRE

Services, Except for

Health Services

Wal-Mart 649 137

Sam’s Club 649 137

Target (retail satellite lots) 488 78

Augustine Office Bldg. 89 27 1071

Takatna Commons 5 693

Commercial Development: Southwest Corner of Minnesota Drive and C Street Development

470 32 476 45

Commercial Development: Northwest Corner of C Street and International Airport Road*

533 306 308

Commercial Development: Former Church Site Near SE Corner of Lake Otis Blvd. and Abbott Road

441 136


TBG072710184349ANC Page 22

Table 6-4. Pipeline Employment by TAZ

Name of Development TAZ Retail

Employment FIRE

Services, Except for

Health Services

Wal-Mart 649 137

Sam’s Club 649 137

Weidner Midtown Mixed Use Development

299 34

Total 1104 1887 489

* Two new proposed hotels, located in TAZ 533, were added to the hotel special generator file but were not added to the employment numbers in order to avoid double counting trips.

As a general rule, employment was allocated to individual TAZs based on the availability of vacant land and re-developable land. The exception was Downtown Anchorage, where estimating the availability of developable land is complicated by the fact that new development is largely dependent on the redevelopment of surface parking lots. Given this difficulty in estimating future commercial development, it was decided to assume that the Downtown area would maintain its existing share of the MOA total employment (10.3 percent in 2007). Using this figure as the basis, the future employment in Downtown was calculated to grow by 3,824 from 2007 to 2035. The final step in the MOA employment allocation procedure was to allocate the employment remaining after subtracting the permits, pipeline development, and downtown allocations from the Anchorage Bowl and Chugiak-Eagle River control totals. Employment (by employment category) was allocated to individual TAZs, based on the development potential for each TAZ. The development potential of each TAZ, represented as a percentage of the total Anchorage Bowl and Chugiak-Eagle River employment potential, was then multiplied by the total remaining employment (for each employment category) for both the Anchorage Bowl and Chugiak-Eagle River.

4.2 Matanuska-Susitna Borough TAZ Allocation

The allocation of land uses within the MSB presents unique problems. While the MOA has adopted zoning district regulations to guide development, the MSB has no such regulations. As a result, it is much more difficult to predict where and at what density development will occur in the MSB. As a means of compensating for the lack of zoning guidance, the MSB, through its contractor HDR Alaska, developed a land use allocation spreadsheet for the modeled area, using an expert panel of public and private development experts. This modified Dephi Technique provided growth rates for both residential and commercial development by TAZ out to the year 2030; these rates were then applied to base year (2000) household and employment figures to arrive at the future year estimates.


TBG072710184349ANC Page 23

4.2.1 Households

Rather than recalculate the MSB growth factors and household estimates by TAZ, it was decided to use the original 2030 household estimates and simply adjust the figures to reflect the new 2009 ISER projections. The 2030 HDR allocation of households within the MSB was based on a control total conforming to the 2005 ISER forecast. This forecast was more than 20 percent higher than the most recent MSB household forecast released by ISER in December 2009 for the comparable year (2030). Thus, for this land use allocation update, the total estimated 2030 households by TAZ was uniformly reduced to match the ISER 2035 MSB household control total of 59,165.

4.2.2 Population

Population per TAZ was calculated by multiplying the number of TAZ households by each TAZ’s household size (persons per household).

4.2.3 Employment

The calculation of employment by TAZ for the MSB was a two-step process. The first step involved extending the growth rates for retail and non-retail from 2030 to 2035 and recalculating the employment totals for each TAZ. (The original HDR spreadsheet included only two employment categories, retail and non-retail.) These figures were adjusted to match the 2035 ISER control totals. The second step involved converting the non-retail employment to the 13 employment categories used by the AMATS transportation demand model. This was accomplished by multiplying the percentage share of non-retail and retail employment for each TAZ by the control total for each employment category (see Table 5-4). The allocation of school-related employment by TAZ required a separate analysis since it is driven by the school location. In order to calculate these estimates, it was first necessary to obtain the location of the MSB schools by TAZ. School staff, estimated to be 2,062 in 2008 was then grown in proportion to the projected 2035 school age population to arrive at a school employment control total of 4,310 in 2035. Since no information is known regarding the location of future schools in the MSB, the additional school-related employment was proportionately added to the existing 2008 school-related employment. It was also necessary to manually allocate university employment to the TAZ in which the MSB branch of UAA is located (TAZ 713). Hotels were not included as part of the original MSB transportation demand model. Research indicated that there were a total of 16 hotels within the MSB modeled area in 2010, with 410 rooms. Each of these hotels was geocoded to the proper TAZ and added to the Anchorage hotel database. No assumptions were made regarding future location of hotels in the MSB.


TBG072710184349ANC Page 24

7 RESULTS

The methodology described in the previous sections provided estimates of households, population, employment by employment category, and total employment growth between 2007 and 2035 for each TAZ within the regional model boundaries. These growth estimates were added to the existing 2007 figures to provide 2035 estimated totals for each TAZ.

7.1 Household Growth

7.1.1 Anchorage Bowl Household Growth

Figure 7-1 shows the calculated household growth for the Anchorage Bowl between 2007 and 2035. Future household growth appears to be distributed throughout the area, with growth hotspots occurring around the sand land gravel pits in the southwest and in the area south of Rabbit Creek Road in the southeast. Figure 7-2 shows, in more detail, the projected household growth in the downtown and midtown districts. As illustrated, a substantial amount of new housing is projected to be built in the Downtown core, as well as in the commercially zoned areas of Midtown. Because little housing has been built in these areas within the last few decades, a substantial policy change (as encouraged by the 2020 Anchorage Bowl Comprehensive Plan) will be required to accomplish this outcome. 7.1.2 Chugiak-Eagle River Household Growth Figure 7-3 shows the forecast household growth in Chugiak-Eagle River between 2007 and 2035. The Chugiak-Eagle River has a substantial amount of vacant land zoned for residential use. The areas of high growth generally reflect the location of the suitable vacant residential land. 7.1.3 Matanuska-Susitna Borough Household Allocation The MSB household allocation used 2000 as the base year. As a result, it was deemed more appropriate to present the results in terms of the total estimated 2035 households by TAZ (see Figure 7.4). As the figure shows, a substantial amount of new housing development is expected to occur in the Port McKenzie area as a result of the construction of the Knik Arm Bridge. This area currently has a very low population base. If the bridge were not built, the estimated households in this area would be substantially reduced. 7.2 Employment Growth 7.2.1 Anchorage Bowl Employment Growth Figure 7-5 shows the areas of projected employment growth between 2007 and 2035 within the Anchorage Bowl. In general, future employment growth correspond very closely to existing employment patterns with midtown and the UMED districts expected to receive the largest share of the future Anchorage employment. Downtown is expected to maintain its existing share of total employment (Figure 7-6 is a more detailed map of downtown and midtown employment growth). However, given the lack of recent office development in the Downtown core, a more active public-private partnership approach may be needed in order to achieve this outcome.


TBG072710184349ANC Page 25

7.2.2 Chugiak-Eagle Employment Growth

Employment growth is expected to be substantially less in Chugiak-Eagle River than in the Anchorage Bowl. With respect to the future location of this employment growth, the 2006 CER Comprehensive Plan “Encourages and reinforces the continued growth of employment in the business districts of Eagle River and Peters Creek (see objective on page 29 of the Plan). Figure 7-7, which shows the allocation of future employment growth used in the model, is in line with this objective.

7.2.3 Matanuska-Susitna Borough Employment Allocation

Figure 7-8 shows future employment continuing to be concentrated in the existing employment centers of Wasilla and Palmer. However, the construction of the Knik Arm Bridge will have a substantial impact on future employment in the Port McKenzie area particularly around the Port, which currently has a relatively low employment base.


TBG072710184349ANC Page 26

Figure 7-1. Anchorage Bowl Household Growth (2007 – 2035)


TBG072710184349ANC Page 27

Figure 7-2. Downtown and Midtown Household Growth (2007 – 2035)


TBG072710184349ANC Page 28

Figure 7-3. Downtown and Midtown Household Growth (2007 – 2035)


TBG072710184349ANC Page 29

Figure 7-4. 2035 MSB Total Households


TBG072710184349ANC Page 30

Figure 7-5 Anchorage Bowl Total Employment Growth (2007 – 2035)


TBG072710184349ANC Page 31

Figure 7-6. Downtown and Midtown Total Employment Growth (2007 – 2035)


TBG072710184349ANC Page 32

Figure 7-7. Chugiak-Eagle River Total Employment Growth 2007 – 2035)


TBG072710184349ANC Page 33

Figure 7-8. 2035 MSB Total Households


TBG072710184349ANC Page 34

6 References

Goldsmith, Scott; University of Alaska Anchorage, Institute of Social and Economic Research (ISER). Economic and Demographic Projections for Alaska and Greater Anchorage, 2010–2035. Prepared for HDR Alaska, Inc., in association with Northern Economics, Inc. December 2009. Goldsmith, Scott; University of Alaska Anchorage, Institute of Social and Economic Research (ISER). Memorandum on Economic and Demographic Impacts of a Knik Arm Bridge. Prepared for Northern Economics, Inc., as input to Knik Arm Bridge and Toll Authority Environmental Impact Statement. September 2005. HDR Alaska, Inc.. 2008. Midtown District Plan (public hearing draft). Prepared for the Municipality of Anchorage. Matanuska-Susitna Borough, Planning and Land Use Department. Matanuska-Susitna Borough Comprehensive Development Plan, 2005 Update. 2005. Municipality of Anchorage, Anchorage Metropolitan Area Transportation Solutions (AMATS). Anchorage Bowl 2025 Long-Range Transportation Plan with 2027 Revisions. April 2007. Municipality of Anchorage, Planning Department (MOA). Anchorage 2020, Anchorage Bowl Comprehensive Plan. February 2001. Municipality of Anchorage, Planning Department (MOA). 2007. Anchorage Downtown Comprehensive Plan. Available online at http://www.muni.org/Planning/CBD_CompPlanPHD_Mar07.cfm. December. Municipality of Anchorage, Planning Department (MOA). Chugiak-Eagle River Comprehensive Plan Update. 2006. Municipality of Anchorage Planning Department (MOA). Draft Land Capacity of the Anchorage Bowl to Accommodate Projected Growth. April 20, 2007.

APPENDIX C

TBG121212233731SCO C-1

Appendix C: Model Inventory

APPENDIX C

TBG121212233731SCO C-2


APPENDIX C

TBG121212233731SCO C-3

Networks and Paths Build Highway Network and Paths Build Transit Network and Paths

Trip Generation Land Use and Density Inputs Household Disaggregation Trip Generation Special Generators and Trucks

Trip Distribution Preprocess Estimation

Time of Day/Growth Factoring/Mode Choice Time of Day Growth Factoring Mode Choice

Network Loading Preprocess Highway Assignment Transit Assignment

Outputs/Displays Tables Merge Highway Assignment Merge Highway Turns Merge Truck Assignments Matrix Marginals District Tables Merge Transit On/Offs Trip Length Frequencies

Displays Facility Class + Capacities Map V/C + Volume Map (Directional) V/C + Volume Map (Link Total) Model Vs. Ground Count (Directional) Model Vs. Ground Count (Link Total) Link Travel Times

Evaluation Inputs O-D Travel Times Screenline Volume-Count Tables Facility Class Volume-Count Tables Household/Employment Accessibility Accessibility Indices

APPENDIX C

TBG121212233731SCO C-4

Out of Direction VMT Import/Export Import Export Export Network to Arc Shape

Utilities/Processors Utilities Merge Assigned Travel Times Update Intersection Names Update MSA Travel Times (Debug) Test Model Convergence (Debug) Test TC File Assembly (Debug)

Transit Route Management Build TransCAD Routes Unbuild TansCAD Routes

Database Management Clean/Repair Network DB Global Schema Updates

Count Utilities Import Traffic Count Data Map Active Counts to Network

Air Quality Build Emissions Grid Report AM Loads by Purpose Report PM Loads by Purpose Report Off Peak Loads by Purpose Merge Trip Purpose Loads Report Gridded Link Data Report Gridded Zone Data

Spreadsheets Mode Choice Summary Screenline Summary Functional Classification Summary Highway LOS Summary Highway VMT/VHT Summary Employment Accessibility Air Quality Analysis Noise Impact Analysis Daily Delay Travel Time Intersection LOS Summary Transit Assignment Report

OD Estimation Re-estimate AM Hwy Triptable

APPENDIX C

TBG121212233731SCO C-5

Re-estimate PM Hwy Triptable Re-estimate OP Highway Triptable Allocate Re-estimated AM Triptables Allocate Re-estimated PM Triptables Allocate Re-estimated OP Triptables

Subarea Analysis Build AM Subarea Triptables Build PM Subarea Triptables Build OP Subarea Triptables Generate AM VISSIM File Generate PM VISSIM File Generate OP VISSIM File

Batch Processing Distribution Feedback Initial Distribution Feedback Iteration Generate MOBILE Inputs

Projects Manager

Scenario Manager

POV Navigator

View/Edit Function Notes

APPENDIX C

TBG121212233731SCO C-6

FIGURE C-1: AMATS MODEL FLOWCHART

Define/UpdateRoad Network

Elements

Update Speeds/Capacities

Highway Network Development

Calculate AM Peak Period Paths

CalculatePM Peak Period Paths

Calculate Off Peak Period Paths

Highway Paths

Define/Update Transit Network

Elements

Update Modal Running Times

Transit Network Development

Calculate AM Peak Period Paths

Calculate PM Peak Period Paths

Calculate Off Peak Period Paths

Transit Paths

Estimate Household Size Groups by TAZ

Estimate Household Income Groups by

TAZ

Estimate Household Worker Groups by

TAZ

Estimate Auto Availability Groups

by TAZ

Household Disaggregation

Generate Home Based Work Trips

Generate Home Based Shop Trips

Generate Home Based School Trips

Generate Home Based Other Trips

Generate Non-home Based Work Trips

Trip Productions and Attractions

Hotel Visitor Trip Generation

Special Trip Rates and Purpose

Allocation

Truck Trip Generation

Special Generators

Calculate AM Peak Period Intrazonal

Times

Calculate PM Peak Period Intrazonal

Times

Calculate Off Peak Period Intrazonal

Times

Highway Intrazonals

Calculate AM Peak Period Intrazonal

Times

Calculate PM Peak Period Intrazonal

Times

Calculate Off Peak Period Intrazonal

Times

Transit Intrazonals

Home Based Shop Trip Distribution

Home Based Other Trip Distribution

Non-home Based Work Trip

Distribution

Non-home Based Nonwork Trip Distribution

Home Based School Growth Distirbution

Trip Distribution

Home Based Work Trip Distribution

Truck Trip Distribution (QRS)

Generate Non-home Based Nonwork

Trips

Factor AM Peak Period Trips by

Purpose

Factor PM Peak Period Trips by

Purpose

Factor Off Peak Period Trips by

Purpose

Estimate Time of Day Triptables

Home Based Work Mode Choice

Home Based Shop Mode Choice

Home Based School Mode Choice

Home Based Other Mode Choice

Non-home Based Work Mode Choice

Mode Choice Models (Logit Based)

Non-home Based Nonwork Mode

Choice

Assign AM Peak Period

HighwayTrips

Assign PM Peak Period

HighwayTrips

Assign Off Peak Period

HighwayTrips

Highway Trip Assignment(Equilibrium)

Assign AM Peak Period TransitTrips

Assign PM Peak Period TransitTrips

Assign Off Peak Period TransitTrips

Transit Trip Assignment

Estimate Street Intersection

Densities for TAZs

Calculate Mixed Use Land Use Index

Calculate 1 Mile Buffer Land Use

Densities

Calculate 1/2 Mile Buffer Land Use

Densities

Estimate TAZ Connectivity Index

Land Use/Density Variables

Growth Factor AM Peak Period Trips

Growth Factor PM Peak Period Trips

Growth Factor Off Peak Period Trips

Growth Factor Time of Day Triptables

APPENDIX D

TBG121212233731SCO D-1

Appendix D: Variable Dictionary AMATS Model Socioeconomic Database

APPENDIX D

TBG121212233731SCO D-2


APPENDIX D

TBG121212233731SCO D-3

(Note: This table does include computed values.)

TABLE D-1 Socioeconomic Database Dictionary

SE Data Field Attribute Type Description

ZONE Integer Traffic analysis zone ID

POP Integer Total zone population

TOTALHH Integer Total number of households

AVHHS Real Number Average household size. Measured at the TAZ level as zonal population divided by zonal households.

AVINC Integer Median household income (2002 dollars)

AVAO** Real Number Average auto ownership per household

AVWHH Real Number Average number of workers per household

AVCHH Real Number Average number of children per household. Measured at the TAZ level as zonal number of children divided by zonal households.

SENROLL Integer School enrollment in zone

PSENROLL Integer Private school enrollment in zone

AEMP Integer Agricultural employment

BEMP Integer Mining employment

CEMP Integer Construction employment

DEMP Integer Manufacturing employment

EEMP Integer Transportation & Public Utilities employment

FEMP Integer Wholesale Trade employment

GEMP Integer Retail Trade employment

HEMP Integer Finance, Insurance, & Real Estate employment

IEMP Integer Service employment

JEMP Integer Government employment

SEMP Integer School employment

UEMP Integer University employment

XEMP Integer Health Services employment

TOTEMP Integer Total zone employment (including special generators not included in categorical employment attributes)

HHS1** Real Number Percent of households in zone with 1 person

HHS2** Real Number Percent of households in zone with 2 people

HHS3** Real Number Percent of households in zone with 3 people

HHS4** Real Number Percent of households in zone with 4 or more people

APPENDIX D

TBG121212233731SCO D-4



INC1** Real Number Percent of households in zone with low income (<40k/year in 2002 dollars)

INC2** Real Number Percent of households in zone with medium income (40k-70k/year in 2002 dollars)

INC3** Real Number Percent of households in zone with high income (>70k/year in 2002 dollars)

AO0** Real Number Percent of households in zone with 0 automobiles

AO1** Real Number Percent of households in zone with 1 automobile

AO2** Real Number Percent of households in zone with 2 automobiles

AO3** Real Number Percent of households in zone with 3 or more automobiles

WHH0** Real Number Percent of households in zone with 0 workers

WHH1** Real Number Percent of households in zone with 1 worker

WHH2** Real Number Percent of households in zone with 2 workers

WHH3** Real Number Percent of households in zone with 3 or more workers

INTDENS* Real Number Estimated street intersection density (per square mile)

WALKDENS* Real Number Estimated walk density by zone

MUIDX* Real Number Mixed Use Index is a ratio of households and employment per zone.

AUTOSPP** Real Number Average auto ownership per person. Measured at the TAZ level as zonal auto ownership divided by zonal population.

SRZFLAG Integer Super Retail Zone flag (1-SRZ) – Examples are Costco and WalMart

PKCOST Real Number Trip parking cost (2002 dollars – Monthly) – Not for work trips

WPKCOST Real Number Work trip parking cost (2002 dollars – Daily)

DUMCBD Real Number Central Business District Flag (1=CBD)

RTRATIO* Real Number Ratio of zone retail employment to total employment

DPOVTT Real Number Value of 0 or 3. All MOA is value of 3, and all Mat-Su is a value of 0.

PAOVTT Real Number Value of 0 or 5. All MOA is value of 5, and all Mat-Su is a value of 0.

LNEMP1* Real Number Natural log of the total employment within 1 mile of the TAZ.

LNEMPH* Real Number Natural log of the total employment within 1/2 mile of the TAZ.

LNNREH* Real Number Natural log of the non-retail employment within ½ mile of the TAZ.

TEMP1* Real Number Total employment within 1 mile of the TAZ.

TEMPH* Real Number Total employment within 1/2 mile of the TAZ.

APPENDIX D

TBG121212233731SCO D-5



NREMPH* Real Number Total non-retail employment within ½ mile of the TAZ.

RERATIOH* Real Number The ratio of retail employment to total employment within ½ mile of the TAZ.

NRETDENS* Real Number The density of non-retail employment within 1 mile of the TAZ.

RATIOPEF Real Number The ratio of the pedestrian equivalency factor.

RETDENS* Real Number The density of retail employment within 1 mile of the TAZ.

POPDENS* Real Number Population density within 1 mile of the TAZ.

*Note: Calculated during trip generation, all fields are related to land use and density inputs. **Note: Calculated during trip generation, all fields are related to household disaggregation.

APPENDIX D

TBG121212233731SCO D-2


APPENDIX E

TBG121212233731SCO E-1

Appendix E: Special Generators

APPENDIX E

TBG121212233731SCO E-2


APPENDIX E

TBG121212233731SCO E-3

Table E-1: Home Based Special Generator Rates ZONE PRATE PID ARATE AID HBWP HBSP HBSCHP HBOP HBWA HBSA HBSCHA HBOA LOCID

1 0.0492 Base Employment 0.0492 Base Employment 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Elmendorf Gate - Post Rd 2 1.4250 Base Pop 2.1700 Base Employment 0.300 0.000 0.000 0.330 0.565 0.000 0.000 0.264 Fort Richardson 2 0.0240 Base Employment 0.0240 Base Employment 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Fort Richardson – Trucks

13 1.5420 Employment 1.5420 Employment 0.000 0.000 0.000 0.000 0.293 0.000 0.000 0.000 Ship Creek 13 114.9400 Freight Tons (M) 114.9400 Freight Tons (M) 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Ship Creek - Port Trucks 94 1.0000 Unit 1.0000 Unit 0.000 0.000 0.000 0.000 0.040 0.000 0.000 0.000 Convention Center Parking Lot 99 1.0000 Unit 1.0000 Unit 0.000 0.000 0.000 0.000 0.040 0.000 0.000 0.000 Convention Center Parking Lot

102 1.0000 Unit 1.0000 Unit 0.000 0.000 0.000 0.000 0.499 0.000 0.000 0.000 Convention Center Parking Lot 142 4.3950 Beds 16.5350 Beds 0.000 0.000 0.000 0.000 0.499 0.000 0.000 0.218 Columbia Hospital 142 0.0960 Employment 0.0960 Employment 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Columbia Hospital - Trucks 217 0.0770 Students 0.9570 Students 0.000 0.000 0.000 0.000 0.240 0.000 0.000 0.565 UAA 217 0.1060 Employment 0.1060 Employment 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 UAA - Trucks 249 0.1150 Students 1.4360 Students 0.000 0.000 0.000 0.000 0.240 0.000 0.000 0.565 APU 249 0.1060 Employment 0.1060 Employment 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 APU - Trucks 253 0.0770 Students 0.9570 Students 0.000 0.000 0.000 0.000 0.240 0.000 0.000 0.565 UAA 253 0.1060 Employment 0.1060 Employment 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 UAA - Trucks 267 0.2270 Out Patients 0.9120 Out Patients 0.000 0.000 0.000 0.000 0.507 0.000 0.000 0.185 Providence Medical 267 0.0960 Employment 0.0960 Employment 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Providence Hospital - Trucks 272 22.7300 Employment 55.5400 Employment 0.000 0.000 0.000 0.000 0.036 0.000 0.000 0.553 Library 274 1.0000 Employment 1.0000 Employment 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Providence Hospital, UAA 274 1.0000 Employment 1.0000 Employment 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Providence Hospital, UAA 288 0.0000 None 2.8600 Employment 0.000 0.000 0.000 0.000 0.414 0.000 0.000 0.408 Alaska Native Hospital 288 0.0960 Employment 0.0960 Employment 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Alaska Native Hospital – Trucks 328 0.0020 Enplanements 0.0180 Enplanements 0.000 0.000 0.000 0.000 0.231 0.000 0.000 0.454 Airport 328 0.0065 Landings 0.0065 Landings 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Airport-Trucks 343 0.0000 None 1.3600 Unit 0.000 0.000 0.000 0.000 0.375 0.000 0.000 0.313 Air National Guard 532 1.0000 Unit 1.0000 Unit 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 UAA Sports Arena 648 0.0590 Region Pop(100s) 0.0590 Region Pop(100s) 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Eagle River Site 900 1.0000 Unit 1.0000 Unit 0.253 0.100 0.000 0.447 0.060 0.000 0.000 0.180 North External

901 2.4050 Region Employment (100s) 4.0350 Region Pop(100s) 0.175 0.080 0.000 0.312 0.120 0.000 0.000 0.721 South External

910 2.4300 Base Pop 3.2890 Base Employment 0.300 0.000 0.000 0.330 0.565 0.000 0.000 0.264 Elmendorf-Gov't Hill 911 2.4300 Base Pop 3.2890 Base Employment 0.300 0.000 0.000 0.330 0.565 0.000 0.000 0.264 Elmendorf-Boniface 912 1.0000 Unit 1.0000 Unit 0.235 0.079 0.000 0.355 0.151 0.000 0.000 0.000 Knik Arm Extension

APPENDIX E

TBG121212233731SCO E-4

Table E-2: Non-Home Based Special Generator Rates ZONE PRATE PID ARATE AID NHBWP NHBNWP NHBWA NHBNWA LOCID

1 0.0492 Base Employment 0.0492 Base Employment 0.000 0.000 0.000 0.000 Elmendorf Gate - Post Rd 2 1.4250 Base Pop 2.1700 Base Employment 0.183 0.183 0.086 0.084 Fort Richardson 2 0.0240 Base Employment 0.0240 Base Employment 0.000 0.000 0.000 0.000 Fort Richardson – Trucks

13 1.5420 Employment 1.5420 Employment 0.222 0.000 0.346 0.134 Ship Creek 13 114.9400 Freight Tons (M) 114.9400 Freight Tons (M) 0.000 0.000 0.000 0.000 Ship Creek - Port Trucks 94 1.0000 Unit 1.0000 Unit 0.000 1.000 0.550 0.000 Convention Center Parking Lot 99 1.0000 Unit 1.0000 Unit 0.000 1.000 0.550 0.000 Convention Center Parking Lot

102 1.0000 Unit 1.0000 Unit 0.000 1.000 0.550 0.000 Convention Center Parking Lot 142 4.3950 Beds 16.5350 Beds 0.500 0.500 0.142 0.141 Columbia Hospital 142 0.0960 Employment 0.0960 Employment 0.000 0.000 0.000 0.000 Columbia Hospital - Trucks 217 0.0770 Students 0.9570 Students 0.608 0.392 0.053 0.068 UAA 217 0.1060 Employment 0.1060 Employment 0.000 0.000 0.000 0.000 UAA - Trucks 249 0.1150 Students 1.4360 Students 0.608 0.392 0.053 0.068 APU 249 0.1060 Employment 0.1060 Employment 0.000 0.000 0.000 0.000 APU - Trucks 253 0.0770 Students 0.9570 Students 0.608 0.392 0.053 0.068 UAA 253 0.1060 Employment 0.1060 Employment 0.000 0.000 0.000 0.000 UAA - Trucks 267 0.2270 Out Patients 0.9120 Out Patients 0.603 0.397 0.084 0.108 Providence Medical 267 0.0960 Employment 0.0960 Employment 0.000 0.000 0.000 0.000 Providence Hospital - Trucks 272 22.7300 Employment 55.5400 Employment 0.000 1.000 0.000 0.411 Library 274 1.0000 Employment 1.0000 Employment 0.000 0.000 0.000 0.000 Providence Hospital, UAA 274 1.0000 Employment 1.0000 Employment 0.000 0.000 0.000 0.000 Providence Hospital, UAA 288 0.0000 None 2.8600 Employment 0.000 0.000 0.049 0.049 Alaska Native Hospital 288 0.0960 Employment 0.0960 Employment 0.000 0.000 0.000 0.000 Alaska Native Hospital – Trucks 328 0.0020 Enplanements 0.0180 Enplanements 0.500 0.500 0.086 0.086 Airport 328 0.0065 Landings 0.0065 Landings 0.000 0.000 0.000 0.000 Airport-Trucks 343 0.0000 None 1.3600 Unit 0.000 0.000 0.156 0.156 Air National Guard 532 1.0000 Unit 1.0000 Unit 0.000 0.000 0.000 0.000 UAA Sports Arena 648 0.0590 Region Pop(100s) 0.0590 Region Pop(100s) 0.000 0.000 0.000 0.000 Eagle River Site 900 1.0000 Unit 1.0000 Unit 0.060 0.100 0.000 0.684 North External

901 2.4050 Region Employment (100s) 4.0350 Region Pop(100s) 0.117 0.200 0.000 0.120 South External

910 2.4300 Base Pop 3.2890 Base Employment 0.183 0.183 0.086 0.084 Elmendorf-Gov't Hill 911 2.4300 Base Pop 3.2890 Base Employment 0.183 0.183 0.086 0.084 Elmendorf-Boniface 912 1.0000 Base Employment 0.0492 Base Employment 0.101 0.168 0.238 0.000 Knik Arm Extension

APPENDIX E

TBG121212233731SCO E-5

Table E-3: Truck Special Generator Rates

ZONE PRATE PID ARATE AID SINGLEP COMBIP SINGLEA COMBIA LOCID 1 0.0492 Base Employment 0.0492 Base Employment 0.700 0.300 0.700 0.300 Elmendorf Gate - Post Rd 2 1.4250 Base Pop 2.1700 Base Employment 0.000 0.000 0.000 0.000 Fort Richardson 2 0.0240 Base Employment 0.0240 Base Employment 0.700 0.300 0.700 0.300 Fort Richardson – Trucks

13 1.5420 Employment 1.5420 Employment 0.000 0.000 0.000 0.000 Ship Creek 13 114.9400 Freight Tons (M) 114.9400 Freight Tons (M) 0.320 0.680 0.320 0.680 Ship Creek - Port Trucks 94 1.0000 Unit 1.0000 Unit 0.000 0.000 0.410 0.000 Convention Center Parking Lot 99 1.0000 Unit 1.0000 Unit 0.000 0.000 0.410 0.000 Convention Center Parking Lot

102 1.0000 Unit 1.0000 Unit 0.000 0.000 0.410 0.000 Convention Center Parking Lot 142 4.3950 Beds 16.5350 Beds 0.000 0.000 0.000 0.000 Columbia Hospital 142 0.0960 Employment 0.0960 Employment 0.875 0.125 0.875 0.125 Columbia Hospital - Trucks 217 0.0770 Students 0.9570 Students 0.000 0.000 0.000 0.000 UAA 217 0.1060 Employment 0.1060 Employment 0.875 0.125 0.875 0.125 UAA - Trucks 249 0.1150 Students 1.4360 Students 0.000 0.000 0.000 0.000 APU 249 0.1060 Employment 0.1060 Employment 0.875 0.125 0.875 0.125 APU - Trucks 253 0.0770 Students 0.9570 Students 0.000 0.000 0.000 0.000 UAA 253 0.1060 Employment 0.1060 Employment 0.875 0.125 0.875 0.125 UAA - Trucks 267 0.2270 Out Patients 0.9120 Out Patients 0.000 0.000 0.000 0.000 Providence Medical 267 0.0960 Employment 0.0960 Employment 0.875 0.125 0.875 0.125 Providence Hospital - Trucks 272 22.7300 Employment 55.5400 Employment 0.000 0.000 0.000 0.000 Library 274 1.0000 Employment 1.0000 Employment 0.875 0.125 0.875 0.125 Providence Hospital, UAA 274 1.0000 Employment 1.0000 Employment 0.875 0.125 0.875 0.125 Providence Hospital, UAA 288 0.0000 None 2.8600 Employment 0.000 0.000 0.000 0.000 Alaska Native Hospital 288 0.0960 Employment 0.0960 Employment 0.875 0.125 0.875 0.125 Alaska Native Hospital – Trucks 328 0.0020 Enplanements 0.0180 Enplanements 0.000 0.000 0.000 0.000 Airport 328 0.0065 Landings 0.0065 Landings 0.808 0.192 0.808 0.192 Airport-Trucks 343 0.0000 None 1.3600 Unit 0.000 0.000 0.000 0.000 Air National Guard 532 1.0000 Unit 1.0000 Unit 0.000 0.000 0.000 0.000 UAA Sports Arena 648 0.0590 Region Pop(100s) 0.0590 Region Pop(100s) 0.650 0.350 0.650 0.350 Eagle River Site 900 1.0000 Unit 1.0000 Unit 0.026 0.014 0.052 0.028 North External

901 2.4050 Region Employment (100s) 4.0350 Region Pop(100s) 0.069 0.046 0.023 0.016 South External

910 2.4300 Base Pop 3.2890 Base Employment 0.000 0.000 0.000 0.000 Elmendorf-Gov't Hill 911 2.4300 Base Pop 3.2890 Base Employment 0.000 0.000 0.000 0.000 Elmendorf-Boniface 912 1.0000 Base Employment 0.0492 Base Employment 0.043 0.023 0.512 0.064 Knik Arm Extension

APPENDIX E

TBG121212233731SCO E-2


APPENDIX F

TBG121212233731SCO F-1

Appendix F: Hotel/Motel Visitor Model

APPENDIX F

TBG121212233731SCO F-2


APPENDIX F

TBG121212233731SCO F-3

Table F-1: Hotel/Motel Visitor Model

Zone Number of

Rooms Average

Occupancy Rate Hotel/Motel Name Street Address

10 32 0.50 Sourdough Visitors Lodge 801 ERICKSON ST

27 59 0.60 Ramada Limited Hotel of Anchorage 207 MULDOON RD

30 100 0.60 Comfort Inn 111 W SHIP CREEK AVE

43 20 0.60 Anchorage Uptown Suites 234 E SECOND AVE

45 31 0.60 Anchorage Grand Hotel 505 W SECOND AVE

53 26 0.60 Historic Anchorage Hotel 330 E ST

53 600 0.80 Hilton Anchorage Hotel 500 WEST THIRD AVE

66 94 0.50 Mush Inn Hotel 333 CONCRETE ST

69 90 0.60 Ramada Inn Anchorage Downtown 115 E THIRD AVE

70 39 0.60 Merrill Field Inn 420 SITKA ST

71 41 0.60 Econo Lodge 642 E FIFTH AVE

79 14 0.60 Copper Whale Inn 440 L ST

79 547 0.80 The Hotel Captain Cook FOURTH AVENUE AT K STREET

81 43 0.60 Anchor Arms Hotel 433 EAGLE ST

81 251 0.75 Holiday Inn Downtown 239 W FOUTH AVE

85 55 0.70 Red Roof 1104 E. 5th AVE

85 60 0.60 Rodeway Inn 1124 E 5TH AVE

86 130 0.60 Days Inn Downtown 321 E FIFTH AVE

89 84 0.60 Inlet Inn 539 H ST

89 198 0.75 Westmark Anchorage Hotel 720 WEST FIFTH AVE

91 38 0.60 The Voyager Hotel 501 K ST

95 14 0.60 Caribou Inn 501 L ST

102 392 0.80 Anchorage Marriott Downtown 820 WEST 7TH AVE

106 40 0.75 Ingra House 641 INGRA ST

107 375 0.80 Sheraton Anchorage Hotel 401 E SIXTH AVE

APPENDIX F

TBG121212233731SCO F-4


Zone Number of

Rooms Average


113 111 0.75 Hawthorn Suites 1110 W EIGHTH AVE

116 90 0.70 Extended Stay Deluxe Downtown 108 E. 8th AVE

116 111 0.75 Clarion Suites 325 W EIGHTH AVE

120 28 0.60 Dukes 8th Ave. Hotel 630 W. 8th AVE

153 180 0.60 Inlet Tower Hotel & Suites 1200 L ST

155 20 0.60 Anchorage Suite Lodge 441 E 15TH AVE

155 63 0.60 Black Angus Inn 1430 GAMBELL ST

236 169 0.75 Embassy Suites 600 E BENSON

245 148 0.75 Marriott Residence Inn 1025 E 35TH AVE.

257 102 0.75 Springhill Suites 3401 A ST

258 138 0.60 Extended Stay Deluxe Midtown 700 E. 34th AVE

262 85 0.65 Super 8 Motel 3501 MINNESOTA DR

267 43 0.80 Providence House PROVIDENCE HOSPITAL CAMPUS

268 83 0.60 Best Western Golden Lion 1000 E 36TH AVE

279 34 0.60 Chelsea Inn Hotel 3836 SPENARD RD

279 38 0.60 Spenard Hotel Motel Inn 3960 SPENARD RD

279 40 0.60 Best Inns & Suites 4110 SPENARD RD

283 54 0.50 Royal Suite Lodge 3811 MINNESOTA DR

298 101 0.75 Hampton Inn 4301 CREDIT UNION DR

300 44 0.60 Lakeshore Motor Inn 3009 LAKESHORE DR

300 86 0.60 Puffin Inn 4400 SPENARD RD

303 33 0.60 Eagle Nest Best Suites 4110 SPENARD RD

303 102 0.60 Best Value Inn Executive Suites 4360 SPENARD RD

303 128 0.60 Holiday Inn Express 4411 SPENARD RD

303 217 0.75 Best Western Barratt Inn 4616 SPENARD RD

310 54 0.50 Long House Alaskan Hotel 4335 WISCONSIN

APPENDIX F

TBG121212233731SCO F-5


Zone Number of

Rooms Average


310 248 0.80 Millennium Alaskan Anchorage 4800 SPENARD ROAD

316 83 0.75 Motel 6 5000 A St

316 106 0.75 Fairfield Inn & Suites 5060 A STREET

316 122 0.75 Homewood Suites by Hilton 140 W TUDOR RD

316 125 0.75 Hilton Garden Inn 100 W TUDOR RD

316 161 0.75 Holiday Inn 109 W INTERNATIONAL RD

318 50 0.70 Parkwood Inn 4455 JUNEAU ST

319 141 0.75 Coast International Inn 3333 W INT'L AIRPORT RD

326 4 0.65 Lake Hood Inn 4702 SPENARD RD

326 79 0.60 Microtel Inn & Suites 5205 NORTHWOOD DR

326 154 0.75 Courtyard by Marriott 4901 SPENARD RD

335 24 0.60 Arctic Inn Motel 842 W INT'L AIRPORT RD

394 109 0.60 Dimond Center Hotel 700 E DIMOND BLVD

621 60 0.60 Microtel Inn Eagle River 13049 OLD GLENN HWY

632 32 0.50 Eagle River Motel 111111 OLD EAGLE RIVER RD

686 30 0.75 Alaska Choice Inn Motel 1930 Glenn Hwy

705 28 0.60 Pioneer Motel 124 Arctic Ave. (MSB)

705 43 0.60 Colony Inn 325 E. Elmwood (MSB)

705 28 0.60 Gold Miners Hotel 918 S. Colony Way (MSB)

712 21 0.60 Peak Inn 775 West Evergreen Ave

717 20 0.70 Tannenhof Llama Inn 2200 N. Laurel Dr.

720 14 0.70 Motherlode Lodge 14N Palmer Fishhook Rd.

730 6 0.60 Alaska Birch Cottages Cotswald Circle (MSB)

736 11 0.70 Agate Inn 4725 E. Begich Circle

739 79 0.60 Grand View Inn & Suites 2900 E. Parks Hwy

APPENDIX F

TBG121212233731SCO F-6


Zone Number of

Rooms Average


750 15 0.80 Alaska's Select Inn Motel 3451 E. Palmdale Dr.

754 12 0.80 Sullys Sourdough Inn Mi. 515 Parks Highway

767 54 0.60 Best Western Lake Lucille 1300 W. Lake Lucille Dr.

767 10 0.60 A&W Windbreak Hotel 2201 E Parks Hwy. (MSB)

780 6 0.70 The Silver Fox Inn 50 W. Parks Hwy

808 12 0.75 Valley Hotel Inc. 606 S. Alaska St.

832 43 0.60 Alaska View Motel 2650 E. Parks Hwy

APPENDIX G

TBG121212233731SCO G-1

Appendix G: Screenlines

APPENDIX G

TBG121212233731SCO G-2


APPENDIX G

TBG121212233731SCO G-3

Average Daily Traffic Screenline Evaluation

FIGURE G-1: COMPARISON OF ALL AVAILABLE MEASURED AND MODELED 2007 WEEKDAY TRAFFIC VOLUMES

AM Peak Period Traffic Screenline Evaluation

FIGURE G-2: COMPARISON OF ALL AVAILABLE MEASURED AND MODELED 2007 AM PEAK PERIOD TRAFFIC VOLUMES

R² = 0.9233

0

5000

10000

15000

20000

25000

30000

35000

0 5000 10000 15000 20000 25000 30000 35000

20

07

Co

un

t

Model Volume

R² = 0.9104

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000

20

07

Co

un

t

Model Volume

APPENDIX G

TBG121212233731SCO G-4

PM Peak Period Traffic Screenline Evaluation

FIGURE G-3: COMPARISON OF ALL AVAILABLE MEASURED AND MODELED 2007 PM PEAK PERIOD TRAFFIC VOLUMES

R² = 0.8143

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000

20

07

Co

un

t

Model Volume

APPENDIX G

TBG121212233731SCO G-5

Daily Screenline Evaluation

FIGURE G-4: SCREENLINE 101: NORTH OF TUDOR ROAD – MINNESOTA DR. TO MULDOON RD.

SouthboundSouthbound

VolumeSouthbound

CapacitySouthbound

Volume w/ CountsSouthbound 2007

Count Northbound Northbound VolumeNorthbound

CapacityNorthbound Volume

w/ CountsNorthbound 2007

COUNTMinnesotaDr 24373 24900 24373 17505 MinnesotaDr 22313 24900 22313 17301HarrisonSt 394 5000 HarrisonSt 386 5000ArcticBlvd 3734 20750 ArcticBlvd 4388 20750CSt 13724 24900 13724 14962 CSt 12051 24900 12051 15298BSt 906 5500 BSt 1529 5500DenaliSt 4697 15000 DenaliSt 6700 15000OldSewardHwy 5476 16600 5476 6067 OldSewardHwy 6744 16600 6744 6099NewSewardHwy 25996 58500 25996 23898

NewSewardHwy 26748 58500 26748 23805LakeOtisPkwy 16431 23000 LakeOtisPkwy 13413 23000PiperSt 3696 5000 PiperSt 3749 5000BragawSt 3810 15000 BragawSt 3745 15000AKNativeMedicalCenterEast 1066 449955 AKNativeMedicalCenterEast 1075 449955BonifacePkwy 10959 16600 10959 7093 BonifacePkwy 10136 16600 10136 7066BaxterRd 436 5000 BaxterRd 517 5000TudorRd 6955 23000 TudorRd 6919 23000PattersonSt 575 5000 PattersonSt 570 5000Totals 123228 713705 80528 69525 120983 713705 77992 69569

MTP 2007 Base - N of Tudor Rd - Minnesota to Muldoon - Southbound-Daily

0

5000

10000

15000

20000

25000

30000

Southbound Volume Southbound 2007 Count

MTP 2007 Base - N of Tudor Rd - Minnesota to Muldoon - Northbound-Daily

0

5000

10000

15000

20000

25000

30000

Northbound Volume Northbound 2007 COUNT

APPENDIX G

TBG121212233731SCO G-6

FIGURE G-5: SCREENLINE 201: NORTH OF DIMOND BOULEVARD – MINNESOTA DR. TO BIRCH RD.


VolumeSouthbound

CapacitySouthbound





COUNTMinnesotaDr 21405 39000 21405 17354

MinnesotaDr 21888 39000 21888 17295ArcticBlvd 3143 12450 ArcticBlvd 3095 12450CSt 8445 16600 8445 7258 CSt 8393 16600 8393 731488thAve 4062 6500 88thAve 3971 6500LakeOtisPkwy 9961 23000 9961 9525 LakeOtisPkwy 9134 23000 9134 9542KingSt 1638 6500 KingSt 1490 6500NewSewardHwy 28378 39000 28378 22083

NewSewardHwy 20874 39000 20874 22723OldSewardHwy 9488 23000 9488 8752 OldSewardHwy 11619 23000 11619 8665HomerDr 877 18000

BraytonDr 7965 18000 7965 9388DimondBlvd 2430 5500 DimondBlvd 3108 5500HartzellRd 2866 6500 HartzellRd 3534 6500AbbottLoopRd 3607 13800 AbbottLoopRd 3382 13800Totals 96300 209850 77677 64972 98453 209850 79873 74927

MTP 2007 Base - N of Dimond Av - Minnesota to Birch - Southbound-Daily

0

5000

10000

15000

20000

25000

30000


MTP 2007 Base - N of Dimond Av - Minnesota to Birch - Northbound-Daily

0

5000

10000

15000

20000

25000

30000

35000

40000

45000


APPENDIX G

TBG121212233731SCO G-7

FIGURE G-6: SCREENLINE 301: SOUTH OF O’MALLEY ROAD – C ST. TO HILLSIDE DR.


VolumeSouthbound

CapacitySouthbound





COUNTCStOMalleyRdEBOnramp 1817 12000

OMalleyRdEBOfframpCSt 1533 12000CSt 2443 23000 CSt 3196 23000

BraytonDr 1123 18000 1123 986OldSewardHwy 14334 15000 OldSewardHwy 12385 15000NewSewardHwy 15119 39000 15119 12926

NewSewardHwy 16352 39000 16352 13027LakeOtisPkwy 6287 20750 LakeOtisPkwy 5327 20750ElmoreRd 971 7400 971 846 ElmoreRd 920 7400 920 844BirchRd 975 5000 BirchRd 981 5000HillsideDr 2791 7400 HillsideDr 2757 7400Totals 44453 129550 16090 13772 44858 147550 18395 14857

MTP 2007 Base - S of O'Malley - C St to Hillside - Southbound-Daily

0

2000

4000

6000

8000

10000

12000

14000

16000


MTP 2007 Base - S of O'Malley - C St to Hillside - Northbound-Daily

0

2000

4000

6000

8000

10000

12000

14000

16000

18000


APPENDIX G

TBG121212233731SCO G-8

FIGURE G-7: SCREENLINE 401: SOUTH OF GLENN HIGHWAY – INGRA ST. TO MULDOON RD.


VolumeSouthbound

CapacitySouthbound





COUNTCSt 8192 24900 8192 13111

ASt 12769 24900 12769 12036CordovaSt 1076 6500 CordovaSt 1884 6500GambellSt 27094 33200

IngraSt 24886 33200 24886 197137thAve 105 5000 7thAve 177 5000KarlukSt 1048 5000 KarlukSt 473 5000MedfraSt 621 5000 MedfraSt 373 5000AirportHeightsDr 10853 15000 10853 6344 AirportHeightsDr 10818 15000 10818 6259Totals 48989 94600 19045 19455 51380 94600 48473 38008

MTP 2007 Base - S of Glenn Hwy - Ingra to Muldoon - Southbound-Daily

0

2000

4000

6000

8000

10000

12000

14000


MTP 2007 Base - S of Glenn Hwy - Ingra to Muldoon - Northbound-Daily

0

5000

10000

15000

20000

25000

30000


APPENDIX G

TBG121212233731SCO G-9

FIGURE G-8: SCREENLINE 501: WEST OF MULDOON ROAD – TUDOR RD. TO GLENN HWY.

Eastbound Eastbound Volume Eastbound CapacityEastbound Volume

w/ CountsEastbound 2007

Count Westbound Westbound VolumeWestbound

CapacityWestbound Volume

w/ CountsWestbound 2007

COUNT

36thAve 1211 5000 36thAve 1294 5000NorthernLightsBlvd 6371 16600 NorthernLightsBlvd 7172 16600DebarrRd 6005 23000 DebarrRd 7470 23000TudorRd 6919 23000 TudorRd 6955 2300020thAve 746 5000 20thAve 853 5000GlennHwy 25003 58500

GlennHwy 24473 585006thAve 913 5000 6thAve 895 5000BoundaryDr 214 5000 BoundaryDr 627 5000Totals 47382 141100 0 0 49739 141100 0 0

MTP 2007 Base - W of Muldoon - Tudor Rd to Glenn Hwy - Eastbound-Daily

0

0

0

0

0

1

1

1

1

1

1

Eastbound Volume Eastbound 2007 Count

MTP 2007 Base - W of Muldoon - Tudor Rd to Glenn Hwy - Westbound-Daily

0

0

0

0

0

1

1

1

1

1

1

Westbound Volume Westbound 2007 COUNT

APPENDIX G

TBG121212233731SCO G-10

FIGURE G-9: SCREENLINE 601: WEST OF BONIFACE PARKWAY – TUDOR RD. TO DAVIS RD.






COUNT

DebarrRd 11234 23000 DebarrRd 12572 23000TudorRd 19559 23000 19559 18966 TudorRd 20094 23000 20094 19040NorthernLightsBlvd 12228 24900 NorthernLightsBlvd 13183 16600

GlennHwy 28290 58500MountainViewDr 2143 7500 MountainViewDr 2537 7500GlennHwy 29853 58500 29853 21659LionheartDr 1381 5000 LionheartDr 1486 5000Totals 76398 141900 49412 40625 78162 133600 20094 19040

MTP 2007 Base - W of Boniface - Tudor Rd to Davis - Eastbound-Daily

0

5000

10000

15000

20000

25000

30000

35000


MTP 2007 Base - W of Boniface - Tudor Rd to Davis - Westbound-Daily

0

5000

10000

15000

20000

25000


APPENDIX G

TBG121212233731SCO G-11

FIGURE G-10: SCREENLINE 602: WEST OF BIRCH ROAD – RABBIT CREEK RD. TO ABBOTT RD.






COUNTRabbitCreekRd 3886 7400 3886 3913 RabbitCreekRd 4143 7400 4143 3913DeArmounRd 1348 7400 DeArmounRd 1398 7400HuffmanRd 821 7400 821 672 HuffmanRd 794 7400 794 670OMalleyRd 4311 7400 4311 4418 OMalleyRd 4641 7400 4641 4416AbbottRd 3493 7400 3493 3142 AbbottRd 3029 7400 3029 3144Totals 13859 37000 12511 12145 14005 37000 12607 12143

MTP 2007 Base - W of Birch - Rabbit Creek to Abbott - Eastbound-Daily

0

500

1000

1500

2000

2500

3000

3500

4000

4500

5000


MTP 2007 Base - W of Birch - Rabbit Creek to Abbott - Westbound-Daily

0

500

1000

1500

2000

2500

3000

3500

4000

4500

5000


APPENDIX G

TBG121212233731SCO G-12

FIGURE G-11: SCREENLINE 701: EAST OF LAKE OTIS PARKWAY – TUDOR RD TO COMMERCIAL RD.






COUNTTudorRd 21863 23000 TudorRd 21835 23000ProvidenceDr 9291 15000 ProvidenceDr 9190 15000NorthernLightsBlvd 20953 16600 NorthernLightsBlvd 23784 1660020thAve 1659 5000 20thAve 1619 5000DebarrRd 16636 16600 DebarrRd 16584 16600MerrillFieldDr 263 5500 MerrillFieldDr 283 5500CommercialDr 5931 15000 CommercialDr 4879 150005thAve 27828 23000 27828 20892 5thAve 27697 23000 27697 20601BluffDr 53 7500 BluffDr 48 7500Totals 104477 127200 27828 20892 105919 127200 27697 20601

MTP 2007 Base - E of Lake Otis - Tudor Rd to Commercial Rd - Eastbound-Daily

0

5000

10000

15000

20000

25000

30000


MTP 2007 Base - E of Lake Otis - Tudor Rd to Commercial Rd - Westbound-Daily

0

5000

10000

15000

20000

25000

30000

Westbound Volume

APPENDIX G

TBG121212233731SCO G-13

FIGURE G-12: SCREENLINE 702: EAST OF LAKE OTIS PARKWAY – DE ARMOUN RD TO DOWLING RD.






COUNTDeArmounRd 3160 7400 DeArmounRd 3539 7400AbbottRd 5998 7400 5998 5395 AbbottRd 6711 7400 6711 5392HuffmanRd 4041 7400 HuffmanRd 3975 7400OMalleyRd 6795 7400 6795 6273 OMalleyRd 6925 7400 6925 624680thAve 4403 7400 80thAve 3988 740072ndAve 1747 7400 1747 1261 72ndAve 1838 7400 1838 125168thAve 3703 7400 68thAve 3775 7400DowlingRd 2305 7500 DowlingRd 2144 750084thAve 2145 7400 84thAve 1982 7400Totals 34297 66700 14540 12929 34877 66700 15474 12889

MTP 2007 Base - E of Lake Otis - De Armoun to Dowling - Eastbound-Daily

0

1000

2000

3000

4000

5000

6000

7000

8000


MTP 2007 Base - E of Lake Otis - De Armoun to Dowling - Westbound-Daily

0

1000

2000

3000

4000

5000

6000

7000

8000


APPENDIX G

TBG121212233731SCO G-14

FIGURE G-13: SCREENLINE 801: EAST OF NEW SEWARD HIGHWAY – RABBIT CREEK RD TO 3RD AVE.

Westbound Westbound VolumeWestbound



Count Eastbound Eastbound Volume Eastbound CapacityEastbound Volume


COUNT13thAve 208 5000 13thAve 23 5000

BensonBlvd 28166 33200NorthernLightsBlvd 27563 3320015thAve 9411 23000 15thAve 11459 23000FireweedLn 931 15000 FireweedLn 1115 150005thAve 27169 24900

4thAve 6289 225003rdAve 4658 225009thAve 1711 5000 1711 1321 9thAve 926 5000 926 13277thAve 65 5000 7thAve 28 5000

6thAve 28233 2490011thAve 43 5000 11thAve 731 5000Totals 71759 138600 1711 1321 76970 138600 926 1327

MTP 2007 Base - E of New Seward - Rabbit Creek to 3rd - Westbound-Daily

0

200

400

600

800

1000

1200

1400

1600

1800

Westbound Volume Westbound 2007 Count

MTP 2007 Base - E of New Seward - Rabbit Creek to 3rd - Eastbound-Daily

0

200

400

600

800

1000

1200

1400

Eastbound Volume Eastbound 2007 COUNT

APPENDIX G

TBG121212233731SCO G-15

FIGURE G-14: SCREENLINE 901: SOUTH OF INTERNATIONAL AIRPORT ROAD – MINNESOTA DR. TO LAKE OTIS PKWY.

Northbound Northbound VolumeNorthbound



Count SouthboundSouthbound

VolumeSouthbound

CapacitySouthbound


COUNT

MinnesotaDr 26354 39000 26354 22753MinnesotaDr 26382 39000 26382 22848

ArcticBlvd 6058 15000 ArcticBlvd 6192 15000CSt 14722 16600 14722 10158 CSt 14703 24900 14703 10138NewSewardHwy 27784 39000 27784 27965

NewSewardHwy 30364 39000 30364 27678OldSewardHwy 14139 16600 14139 10501 OldSewardHwy 9887 16600 9887 10608

HomerDr 3810 18000 3810 2513BraytonDr 4994 18000 4994 1341LakeOtisPkwy 15192 23000 LakeOtisPkwy 17638 23000AbbottLoop 4316 16600 4316 4909 AbbottLoop 4722 16600 4722 4883Totals 113559 183800 92309 77627 113698 192100 89868 78668

MTP 2007 Base - S of Int'l Airport - Minnesota to Lake Otis - Northbound-Daily

0

5000

10000

15000

20000

25000

30000

Northbound Volume Northbound 2007 Count

MTP 2007 Base - S of Int'l Airport - Minnesota to Lake Otis - Southbound-Daily

0

5000

10000

15000

20000

25000

30000

35000

Southbound Volume Southbound 2007 COUNT

APPENDIX G

TBG121212233731SCO G-16

FIGURE G-15: SCREENLINE 1001: SOUTH OF 92ND AVENUE/ABBOTT ROAD – MINNESOTA DR. TO BIRCH RD.





VolumeSouthbound

CapacitySouthbound



FrontageRd 2251 18000MinnesotaDr 11130 39000 11130 11238CSt 2947 16600 2947 3367 CSt 2564 16600 2564 3376

NewSewardHwy 17937 39000 17937 18226NewSewardHwy 20798 39000 20798 18095KingSt 1452 6500 KingSt 1440 6500BraytonDr 2806 18000 2806 2335OldSewardHwy 8672 16600 8672 8631 OldSewardHwy 10315 16600 10315 8699IndependenceDr 1937 5000 IndependenceDr 3926 5000LakeOtisPkwy 8679 16600 LakeOtisPkwy 8841 16600BirchRd 1221 7400 BirchRd 1463 7400Totals 61893 182700 46353 43666 60880 146700 45210 41515

MTP 2007 Base - S of 92nd/Abbott - Minnesota to Birch - Northbound-Daily

0

5000

10000

15000

20000

25000


MTP 2007 Base - S of 92nd/Abbott - Minnesota to Birch - Southbound-Daily

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000


APPENDIX G

TBG121212233731SCO G-17

FIGURE G-16: SCREENLINE 2001: NORTH OF EAGLE RIVER – GLENN HWY.





VolumeSouthbound

CapacitySouthbound


COUNTGlennHwy 13076 39000 13076 12862 GlennHwy 14706 39000 14706 12777Totals 13076 39000 13076 12862 14706 39000 14706 12777

MTP 2007 Base - N of Eagle River - Glenn Hwy - Northbound-Daily

12750

12800

12850

12900

12950

13000

13050

13100


MTP 2007 Base - N of Eagle River - Glenn Hwy - Southbound-Daily

11500

12000

12500

13000

13500

14000

14500

15000


APPENDIX G

TBG121212233731SCO G-18

FIGURE G-17: SCREENLINE 2002: NORTH OF EAGLE RIVER ROAD ACCESS – GLENN HWY.





VolumeSouthbound

CapacitySouthbound


COUNTGlennHwy 17776 39000

GlennHwy 19622 39000OldGlennHwy 2078 6200 2078 2127 OldGlennHwy 2162 6200 2162 2129Totals 19854 45200 2078 2127 21784 45200 2162 2129

MTP 2007 Base - N of Eagle River Rd Access - Glenn to Birchwood - Northbound-Daily

0

500

1000

1500

2000

2500


MTP 2007 Base - N of Eagle River Rd Access - Glenn to Birchwood - Southbound-Daily

0

500

1000

1500

2000

2500


APPENDIX G

TBG121212233731SCO G-19

FIGURE G-18: SCREENLINE 2003: SOUTH OF HILAND ROAD – GLENN HIGHWAY





VolumeSouthbound

CapacitySouthbound


COUNTGlennHwy 25431 58500 25431 23431Totals 25431 58500 25431 23431 27574 58500 27574 23309

MTP 2007 Base - S of Hiland - Glenn Hwy - Northbound-Daily

22000

22500

23000

23500

24000

24500

25000

25500

26000


MTP 2007 Base - S of Hiland - Glenn Hwy - Southbound-Daily

0

0

0

0

0

1

1

1

1

1

1


APPENDIX G

TBG121212233731SCO G-20

FIGURE G-19: SCREENLINE 2005: NORTH OF 3RD AVENUE – C ST. TO PORT ACCESS RD.





VolumeSouthbound

CapacitySouthbound


COUNTOceanDockRd 583 6500 OceanDockRd 556 6500

LoopRoad 7245 16600LoopRoad 7084 16600Totals 7667 23100 0 0 7801 23100 0 0

MTP 2007 Base - N of 3rd St - C St to Port Access - Northbound-Daily

0

0

0

0

0

1

1

1

1

1

1


MTP 2007 Base - N of 3rd St - C St to Port Access - Southbound-Daily

0

0

0

0

0

1

1

1

1

1

1


APPENDIX G

TBG121212233731SCO G-21

FIGURE G-20: SCREENLINE 2006: WEST OF GAMBELL STREET – 3RD AVE. TO 16TH AVE.






COUNT16thAve 281 6500 16thAve 435 650015thAve 2758 15000 2758 6322 15thAve 4518 15000 4518 641413thAve 197 5000 13thAve 191 50009thAve 5734 15000 5734 3074 9thAve 4697 15000 4697 309811thAve 276 5000 11thAve 328 50006thAve 13219 24900

5thAve 13280 249003rdAve 4020 15000

4thAve 8983 22500 8983 11537Totals 31448 93900 17475 20933 27469 86400 9215 9512

MTP 2007 Base - W of Gambell -3rd St to 16th - Eastbound-Daily

0

2000

4000

6000

8000

10000

12000

14000


MTP 2007 Base - W of Gambell -3rd St to 16th - Westbound-Daily

0

1000

2000

3000

4000

5000

6000

7000


APPENDIX G

TBG121212233731SCO G-22

FIGURE G-21: SCREENLINE 2007: SOUTH OF 9TH AVENUE – L ST. TO MEDFRA ST.





VolumeSouthbound

CapacitySouthbound


COUNTASt 19718 24900 19718 12106

LSt 12418 24900ISt 11921 24900ESt 1084 5000 1084 1479 ESt 999 5000 999 1554

CSt 12641 24900CordovaSt 821 5000 CordovaSt 732 5000

GambellSt 27576 33200IngraSt 23894 33200KarlukSt 623 5000 KarlukSt 960 5000MedfraSt 747 5000 747 503 MedfraSt 642 5000 642 510Totals 58808 103000 21549 14088 55968 103000 1641 2064

MTP 2007 Base - S of 9th - L St to Medfra - Northbound-Daily

0

5000

10000

15000

20000

25000


MTP 2007 Base - S of 9th - L St to Medfra - Southbound-Daily

0

200

400

600

800

1000

1200

1400

1600

1800


APPENDIX G

TBG121212233731SCO G-23

FIGURE G-22: SCREENLINE 2008: EAST OF INGRA STREET – 3RD AVE. TO 15TH AVE.






COUNT15thAve 11717 16600 15thAve 9174 166003rdAve 3674 16434 3674 5674 3rdAve 2486 8466 2486 56885thAve 27437 23000 27437 21891 5thAve 27213 23000 27213 21588Totals 42828 56034 31111 27565 38873 48066 29699 27276

MTP 2007 Base - E of Ingra - 3rd to 15th - Eastbound-Daily

0

5000

10000

15000

20000

25000

30000


MTP 2007 Base - E of Ingra - 3rd to 15th - Westbound-Daily

0

5000

10000

15000

20000

25000

30000


APPENDIX G

TBG121212233731SCO G-24

FIGURE G-23: SCREENLINE 2010: NORTH OF FIREWEED AVENUE/NORTHERN LIGHTS BOULEVARD – MINNESOTA DR. TO MULDOON RD.

FIGURE G-24: SCREENLINE 2013: EAST OF NORTHWOOD DRIVE – NORTHERN LIGHTS BLVD. TO INTERNATIONAL AIRPORT RD.





VolumeSouthbound

CapacitySouthbound


COUNTASt 21113 24900 21113 14620MinnesotaDr 11432 16600 11432 13795

MinnesotaDr 12661 24900SpenardRd 1316 12000

CSt 15733 24900ArcticBlvd 1789 5000 ArcticBlvd 1453 5000NewSewardHwy 30465 33200 30465 25434 NewSewardHwy 29585 33200 29585 22068LakeOtisPkwy 7834 10200 LakeOtisPkwy 10318 19800BragawSt 8925 15000 BragawSt 10414 15000BonifacePkwy 13013 16600 13013 11299 BonifacePkwy 13823 16600 13823 11363BaxterRd 3180 5000 BaxterRd 3210 5000ChandalarDr 1587 5000 ChandalarDr 1666 5000MuldoonRd 11207 23000 11207 13548 MuldoonRd 12140 23000 12140 13635Totals 111861 166500 87230 78696 111003 172400 55548 47066

MTP 2007 Base - N of Fireweed/Northern Lights - Minnesota to Muldoon - Northbound-Daily

0

5000

10000

15000

20000

25000

30000

35000


MTP 2007 Base - N of Fireweed/Northern Lights - Minnesota to Muldoon - Southbound-Daily

0

5000

10000

15000

20000

25000

30000

35000


APPENDIX G

TBG121212233731SCO G-25






COUNT48thAve 819 5500 48thAve 852 5500IntlAirportRd 20329 30000 IntlAirportRd 21825 30000SpenardRd 11521 15000 SpenardRd 7651 15000McRaeRd 1373 5000 McRaeRd 1961 5000NorthernLightsBlvd 11228 16600 11228 11088 NorthernLightsBlvd 11127 16600 11127 11146Totals 45270 72100 11228 11088 43416 72100 11127 11146

MTP 2007 Base - E of Northwood - Northern Lights to Int'l Airport - Eastbound-Daily

0

2000

4000

6000

8000

10000

12000


MTP 2007 Base - E of Northwood - Northern Lights to Int'l Airport - Westbound-Daily

0

2000

4000

6000

8000

10000

12000


APPENDIX G

TBG121212233731SCO G-26

FIGURE G-25: SCREENLINE 2014: NORTH OF INTERNATIONAL AIRPORT ROAD – SPENARD RD. TO LAKE OTIS PKWY.





VolumeSouthbound

CapacitySouthbound



CSt 13777 24900 13777 13506 CSt 12463 24900 12463 13345MinnesotaDr 23010 58500 23010 20949SpenardRd 8705 7500 SpenardRd 5148 7500NorthwoodSt 818 5000 NorthwoodSt 899 5000CambridgeWay 446 5000 CambridgeWay 393 5000ArcticBlvd 5873 20750 ArcticBlvd 4551 20750NewSewardHwy 27784 39000 27784 27965

HomerDr 2546 18000 2546 1253BraytonDr 4994 18000

NewSewardHwy 30364 39000 30364 27678OldSewardHwy 14391 16600 14391 10499 OldSewardHwy 11904 16600 11904 10453LakeOtisPkwy 15192 23000 LakeOtisPkwy 17638 23000Totals 114990 218250 78962 72919 112453 218250 83824 73926

MTP 2007 Base - N of Int'l Airport - Spenard to Lake Otis - Northbound-Daily

0

5000

10000

15000

20000

25000

30000


MTP 2007 Base - N of Int'l Airport - Spenard to Lake Otis - Southbound-Daily

0

5000

10000

15000

20000

25000

30000

35000


APPENDIX G

TBG121212233731SCO G-27

FIGURE G-26: SCREENLINE 2016: WEST OF MINNESOTA DRIVE – RASPBERRY RD. TO KLATT RD.






COUNT100thAve 6463 15000 100thAve 5766 15000DimondBlvd 15966 24900 15966 14658 DimondBlvd 16159 24900 16159 14633KlattRd 3078 13000 KlattRd 2982 13000RaspberryRd 9081 12400 RaspberryRd 9627 12400Totals 34588 65300 15966 14658 34534 65300 16159 14633

MTP 2007 Base - W of Minnesota - Raspberry to Klatt - Eastbound-Daily

0

2000

4000

6000

8000

10000

12000

14000

16000

18000


MTP 2007 Base - W of Minnesota - Raspberry to Klatt - Westbound-Daily

0

2000

4000

6000

8000

10000

12000

14000

16000

18000


APPENDIX G

TBG121212233731SCO G-28

FIGURE G-27: SCREENLINE 2017: SOUTH OF DIMOND BOULEVARD/ABBOTT ROAD – MINNESOTA DR. TO HILLSIDE DR.


VolumeSouthbound

CapacitySouthbound





COUNTArcticBlvd 1736 5000 ArcticBlvd 1633 5000KingSt 1246 6500 KingSt 1301 6500OldSewardHwy 14477 16600 14477 12071 OldSewardHwy 12949 16600 12949 12136CSt 2384 16600 2384 3372 CSt 2766 16600 2766 3374DimondMallWest 2884 449955 DimondMallWest 3026 449955MinnesotaDr 14394 39000 14394 11214

FrontageRd 2251 18000MinnesotaDr 11130 39000 11130 11238

NewSewardHwy 17937 39000 17937 18226NewSewardHwy 20798 39000 20798 18095BraytonDr 3187 18000

IndependenceDr 4372 5000 IndependenceDr 2407 5000LakeOtisPkwy 8841 16600 LakeOtisPkwy 8679 16600AcademyDr 1458 7500 AcademyDr 958 7500AbbottRd 1917 7400 AbbottRd 1706 7400 1706 1551BirchRd 1463 7400 BirchRd 1221 7400Totals 73109 616555 49192 44883 74012 652555 49349 46394

MTP 2007 Base - S of Dimond/Abbott - Minnesota to Hillside - Southbound-Daily

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000


MTP 2007 Base - S of Dimond/Abbott - Minnesota to Hillside - Northbound-Daily

0

5000

10000

15000

20000

25000


APPENDIX G

TBG121212233731SCO G-29

FIGURE G-28: SCREENLINE 2020: SOUTH END OF STUDY AREA – SEWARD HWY.


VolumeSouthbound

CapacitySouthbound





COUNTNewSewardHwy 5504 7400 NewSewardHwy 5840 7400Totals 5504 7400 0 0 5840 7400 0 0

MTP 2007 Base - S End of Study Area - Seward Hwy - Southbound-Daily

0

0

0

0

0

1

1

1

1

1

1


MTP 2007 Base - S End of Study Area - Seward Hwy - Northbound-Daily

0

0

0

0

0

1

1

1

1

1

1


APPENDIX G

TBG121212233731SCO G-30

FIGURE G-29: SCREENLINE 2100: WEST OF A STREET – SHIP CREEK DR. TO TUDOR RD.






COUNT36thAve 12570 15000 36thAve 16105 15000CalaisSt 4220 6500 CalaisSt 2480 6500BensonBlvd 31681 33200 31681 25777FireweedLn 6659 15000 FireweedLn 4051 15000

NorthernLightsBlvd 32760 33200 32760 2243316thAve 856 6500 16thAve 436 650015thAve 1999 12450 15thAve 4329 12450TudorRd 18660 23000 TudorRd 16998 230007thAve 314 5500 314 390 7thAve 312 5500 312 3859thAve 5793 15000 5793 3588 9thAve 3847 15000 3847 35458thAve 526 5500 8thAve 522 5500

5thAve 13865 24900 13865 236504thAve 8326 22500 8326 70436thAve 13133 24900

3rdAve 3623 15000 3623 5666Totals 104737 185050 46114 36798 99328 177550 54407 55679

MTP 2007 Base - W of A - Ship Creek to Tudor - Eastbound-Daily

0

5000

10000

15000

20000

25000

30000

35000


MTP 2007 Base - W of A - Ship Creek to Tudor - Westbound-Daily

0

5000

10000

15000

20000

25000

30000

35000


APPENDIX G

TBG121212233731SCO G-31

FIGURE G-30: SCREENLINE 2101: NORTH OF SHIP CREEK DR. – C ST. TO MULDOON RD.





VolumeSouthbound

CapacitySouthbound


COUNTLoopRoad 7245 16600

LoopRoad 7084 16600PostRd 3555 15000 3555 2429 PostRd 3185 15000 3185 2505Totals 10639 31600 3555 2429 10430 31600 3185 2505

MTP 2007 Base - N of Ship Creek - C St to Muldoon - Northbound-Daily

0

500

1000

1500

2000

2500

3000

3500

4000


MTP 2007 Base - N of Ship Creek - C St to Muldoon - Southbound-Daily

0

500

1000

1500

2000

2500

3000

3500


APPENDIX G

TBG121212233731SCO G-32

FIGURE G-31: SCREENLINE 2200: EAST OF INGRA STREET – 3RD AVE. TO TUDOR RD.

Westbound Westbound VolumeWestbound



Count Eastbound Eastbound Volume Eastbound CapacityEastbound Volume


COUNTNorthernLightsBlvd 19663 24900 19663 18764 NorthernLightsBlvd 18912 24900 18912 1857836thAve 8660 15000 8660 9471 36thAve 7642 15000 7642 9490TudorRd 22283 23000 22283 20395 TudorRd 21864 23000 21864 2032115thAve 9174 16600 15thAve 11717 166005thAve 27213 23000 27213 21588 5thAve 27437 23000 27437 218913rdAve 2685 8466 3rdAve 3893 16434Totals 89678 110966 77819 70218 91465 118934 75855 70280

MTP 2007 Base - E of Ingra - 3rd to Tudor - Westbound-Daily

0

5000

10000

15000

20000

25000

30000

Westbound Volume Westbound 2007 Count

MTP 2007 Base - E of Ingra - 3rd to Tudor - Eastbound-Daily

0

5000

10000

15000

20000

25000

30000

Eastbound Volume Eastbound 2007 COUNT

TBG121212233731SCO 6-1

6.0 Bibliography

CH2M HILL. 2005a. Anchorage Travel Model Calibration and Validation Final Report. February.

CH2M HILL. 2005b. 2025 AMATS Long-Range Transportation Plan.

CH2M HILL. 2010. Anchorage and Matanuska-Susitna Borough 2035 Land Use Allocation and Forecast. July.

Institute of Social and Economic Research (ISER). 2009. Economic and Demographic Projections for Alaska and Greater Anchorage 2010-2035. http://www.iser.uaa.alaska.edu/Publications/EconDemProjectionsAnchorage_v4.pdf

NuStats. 2002. Anchorage Household Travel Survey.

HDR Alaska, Inc. (HDR). 2006. Knik Arm Crossing Land Use and Transportation Forecasting.

HDR Alaska, Inc. (HDR). 2007. Mat-Su Borough Long-Range Transportation Plan (draft). Prepared for the Matanuska-Susitna Borough.

Travel Model Improvement Program (TMIP), Federal Highway Administration (FHWA), Barton-Aschmann Associates, Inc., and Cambridge Systematics, Inc. 1997. Model Validation and Reasonableness Checking Manual. February.

amats 2011 model user guide: documentation and validation...amats 2011 model user guide:...

Documents