development of ciudad juarez traffic simulation model in ... · of ciudad juarez will grow to 1.42...

Development of Ciudad Juarez Traffic Simulation Model in Paramics by Jose Osiris Vidana Bencomo Graduate Assistant Research Center of International Intelligent Transportation Research Doctoral Student University of Texas at El Paso Rajat Rajbhandari Associate Research Engineer Center of International Intelligent Transportation Research In cooperation with Center for International Intelligent Transportation Research Project Number: 186040-00003 Project Title: Development of Ciudad Juarez Traffic Simulation Model in Paramics December 2010 Report prepared by Center for International Intelligent Transportation Research Texas Transportation Institute 4050 Rio Bravo, Suite 151 El Paso, TX 79902 TEXAS TRANSPORTATION INSTITUTE The Texas A&M University System College Station, Texas 77843-3135

Center for International Intelligent Transportation Research Texas Transportation Institute Page v

TABLE OF CONTENTS

Page

List of Figures ............................................................................................................................... vi

List of Tables ............................................................................................................................... vii

Disclaimer and Acknowledgments ........................................................................................... viii

Executive Summary ...................................................................................................................... 1

Chapter 1: INTRODUCTION ..................................................................................................... 2 Ciudad Juarez- El Paso Borderplex ............................................................................................ 2 Socio Economy of Ciudad Juarez- El Paso ................................................................................. 2 Manufacturing Industry (Maquiladoras) ..................................................................................... 3 Movement of Trucks and Passenger Vehicles Crossing the Border ........................................... 4 Mobility Issues in Ciudad Juarez ................................................................................................ 5 Problem Statement ...................................................................................................................... 5 Goals and Objectives of the Project ............................................................................................ 6

Chapter 2: LITERATURE REVIEW ......................................................................................... 7 MICROSCOPIC Traffic Simulation ........................................................................................... 7 Large Scale Microscopic Traffic Simulation .............................................................................. 7 Paramics .................................................................................................................................... 10

Chapter 3: METHODOLOGY .................................................................................................. 11 Introduction ............................................................................................................................... 11 Data Collection ......................................................................................................................... 12

Aerial Images ........................................................................................................................ 12 Street Geometry Data ............................................................................................................ 12 Traffic Signals ....................................................................................................................... 12 Speed Data ............................................................................................................................ 12 Traffic Counts ....................................................................................................................... 14

Modeling ................................................................................................................................... 16 Coding the Network .............................................................................................................. 16 Calibration and Validation .................................................................................................... 19

Chapter 4: RESULTS ................................................................................................................. 21 Measurements of Effectiveness ............................................................................................ 21 Analysis................................................................................................................................. 25

Chapter 5: CONCLUSION ........................................................................................................ 34

References .................................................................................................................................... 35

Center for International Intelligent Transportation Research Texas Transportation Institute Page vi

LIST OF FIGURES

Page Figure 1. Population Projection of the Ciudad Juarez. ................................................................... 2 Figure 2. Location of Industrial Parks in Ciudad Juarez. ............................................................... 4 Figure 3. Flowchart of the Methodology. ..................................................................................... 11 Figure 4. Street Network Configuration of Ciudad Juarez. .......................................................... 13 Figure 5. Speed Limit of Streets in Ciudad Juarez. ...................................................................... 14 Figure 6. Intersection Movement Count Locations in Ciudad Juarez. ......................................... 15 Figure 7. Traffic Count Locations in Ciudad Juarez. .................................................................... 16 Figure 8. Snapshot of Ciudad Juarez Roadway Network in Paramics. ........................................ 18 Figure 9. Snapshot of Bermudez Ave. and Ejercito Nacional Ave. Intersection. ......................... 19 Figure 10. Industrial Parks Coded in the Model of Ciudad Juarez. .............................................. 22 Figure 11. Comparison of Simulated Travel of Commercial Vehicles between Individual

Industrial Parks and Both Ports of Entry. ......................................................................... 25 Figure 12. Comparison of Delay of Commercial Vehicles between Individual Industrial

Parks and Both Ports of Entry. .......................................................................................... 26 Figure 13. Percentage of Delay Incurred to Trucks Originating from Different Industrial

Parks and Both Highway Entrances to Ysleta (Zaragoza Bridge) Port of Entry. ............. 27 Figure 14. Percentage Delay Incurred to Trucks Originating from Different Industrial

Parks and Both Highway Entrances to BOTA (Americas) Port of Entry. ........................ 27 Figure 15. Path of Trucks Traveling from the Chihuahua Highway Entrance to the Ysleta

Port of Entry. ..................................................................................................................... 28 Figure 16. Delay on Segments of a Path from the Chihuahua Highway Entrance to the

Ysleta Port of Entry. ......................................................................................................... 29 Figure 17. Path of Trucks Traveling from Gema to the Ysleta Port of Entry. .............................. 29 Figure 18. Delay on Segments of a Path from Gema to the Ysleta Port of Entry. ....................... 30 Figure 19. Path of Trucks Traveling from Independencia to the BOTA Port of Entry. ............... 31 Figure 20. Delay on Segments of a Path from Independencia to the BOTA Port of Entry. ......... 31 Figure 21. Path of Trucks Traveling from Aztecas to BOTA Port of Entry. ................................ 32 Figure 22. Link Delay of the Path of Trucks Traveling from I.P. Aztecas to BOTA Port of

Entry. ................................................................................................................................. 32 Figure 23. Modeled Travel Time Originating from Different Industrial Parks and Both

Ports of Entry. ................................................................................................................... 33 Figure 24. Delay Time Originating from Different Industrial Parks and Both Ports of

Entry. ................................................................................................................................. 33

Center for International Intelligent Transportation Research Texas Transportation Institute Page vii

LIST OF TABLES

Page Table 1. Large-Scale MTS Projects Developed in Urban Areas. ................................................... 9 Table 2. MOEs between Individual Industrial Parks and the Ysleta (Zaragoza Bridge)

Port of Entry. ..................................................................................................................... 23 Table 3. MOEs between Chihuahua and Casas Grandes Highways Entries and Ysleta

(Zaragoza Bridge) Port of Entry. ...................................................................................... 23 Table 4. MOEs between Individual Industrial Parks and the BOTA (Americas Bridge)

Port of Entry. ..................................................................................................................... 24 Table 5. MOEs between Chihuahua and Casas Grandes Highways Entries and BOTA

(Americas Bridge) Port of Entry. ...................................................................................... 24

Center for International Intelligent Transportation Research Texas Transportation Institute Page viii

DISCLAIMER AND ACKNOWLEDGMENTS

This research was performed by the Center for International Intelligent Transportation Research, a part of the Texas Transportation Institute. The contents of this report reflect the views of the authors, who are responsible for the facts and the accuracy of the data presented herein.

The research team would like to thank Rosario Diaz Arellano from the Instituto Municipal de Investigación y Planeación de Ciudad Juarez (IMIP), Victor Hernández from the civil engineering department at Autonomous University of Ciudad Juarez (UACJ), and Hector Gonzalez from the Ciudad Juarez traffic control office for their assistance in the project. Finally, the research team would also like to thank Jon Williams, Rodrigo Chavez, and Noel Chavez for their contribution in this research.

Center for International Intelligent Transportation Research Texas Transportation Institute Page 1

EXECUTIVE SUMMARY

As the city of Ciudad Juarez grows, ports of entry are now located farther from the newly developed urban areas. The housing developments located around the ports of entry have avenues and streets that have a limited capacity for connecting traffic from the new housing areas to the ports of entry, and the industrial parks that were once outside the city are now inside the city core. Trucks have to use congested inner city streets to transport products and supplies from/to the maquiladoras causing delays that have been exacerbated by the increasing number of accidents, limited roadway capacity, and other traffic controls.

This project develops a traffic simulation model for the Ciudad Juarez that could be combined with an existing traffic model for El Paso into a bi-national traffic simulation model for the Ciudad Juarez-El Paso region. The combined traffic simulation model has numerous short- and long-range benefits, not only for planning but also in operation of transportation infrastructure, including port of entry. This effort facilitates the planning and construction of roadways to connect with the ports of entry by providing an accurate estimate of traffic movement in the area and enables analyses and evaluation of commercial freight movement between various industrial parks maquiladora in Ciudad Juarez to ports of entry on the U.S.-Mexico border.

In this research project, Paramics was used to create a large-scale simulation model of Ciudad Juarez because (1) the software creates an origin-destination matrix from screen line counts and intersections counts since Ciudad Juarez does not have a travel demand model; and (2) it uses a customized application within Paramics to input large number of variables and parameter data for sensitivity analysis. Data from previously installed ITS field devices at border-crossings can be injected into the Paramics model to influence the model, improve the visualization, and evaluate truck behavior while crossing the port of entry.

The simulation model demonstrated various scenarios of truck paths connecting industrial parks to ports of entry. The analysis clearly showed that the delay on truck paths is attributed to congestion in the inner urban areas, while freeways surrounding the city provide reliable access to the ports of entry. The delay contributes to 20–60% of the travel time of trucks moving between industrial parks and the ports of entry, which results in higher costs for shipping and moving goods.

As part of the future development of the model, researchers recommend several improvements, one of which is to include public transportation network in the model. Also, calibration of this model was performed with a small number of field traffic counts and researchers believe that additional counts will definitely improve the reliability of the model.


CHAPTER 1: INTRODUCTION

CIUDAD JUAREZ- EL PASO BORDERPLEX The City of El Paso, Texas, together with Ciudad Juarez, Chihuahua, comprises the largest

metropolitan area on the United States-Mexico border. The estimated population of these two cities is 2,043,514 (742,062 and 1,301,452 in El Paso and Ciudad Juarez, respectively) (U.S. Census Bureau) (INEGI, 2009). The Chihuahua State Government estimates that the population of Ciudad Juarez will grow to 1.42 million by 2010 (Figure 1).

Figure 1. Population Projection of the Ciudad Juarez.

The El Paso/Ciudad Juarez borderplex economy depends highly on the trade between both cities. This commercial trade is categorized in two groups: industrial and commercial. The commercial is related to the stores and services derived from Mexican visitors from Ciudad Juarez that demand cheaper, up-to-date, and high variety products that are not available in their city. The second group is related with the maquiladoras industry. This group demands supplies and services from El Paso companies to be processed in the Ciudad Juarez Maquila industry. Both groups generate economic benefits to El Paso through jobs, hotel services, gasoline, taxes, restaurants, attractions, etc.

SOCIO ECONOMY OF CIUDAD JUAREZ- EL PASO The geographical proximity between the two cities provides opportunities for commercial

and social activities. Hence, the three ports of entry between the two countries provide important


gateways between the two cities and the two countries. The ports of entry experience huge delays and long wait times, due to the extremely high demand of vehicles crossing the border for private and commercial purposes. Due to the North American Free Trade Agreement (NAFTA), a large number of twin manufacturing plants (maquiladoras) have sprung up in Ciudad Juarez and surrounding areas. These twin plants manufacture goods for U.S. companies and ship the goods back to different parts of the U.S. via El Paso. Both Mexican visitors and commercial trucks have an important role in improving the economic development of the borderplex. The benefits of this interchange are shown in retail, taxes, employment, warehouses, and industries on both sides of the border.

In the case of commercial trucks, vehicles cross through the ports of entry in both northbound and southbound directions. In the northbound direction, trucks transport final products to warehouses located in El Paso. In the southbound direction, trucks return empty to Ciudad Juarez or transport supplies stemming from the manufacturing strategy of “production sharing” adopted by companies on both sides of the border. The production sharing consists of sub-assembly on one side of the border and finished products on the other side. Because the borderplex is the seventh largest manufacturing center in North America (in terms of workers) (El Paso Regional Economic Development Corporation, 2010), the demand of truck transportation is one of the most important factors on the U.S./Mexico border.

Mexican visitors travel to El Paso for shopping, tourism, work, schools, and social activities. In the process, a considerable amount of money is spent on a multitude of items (groceries, clothing, appliances, furniture, etc.) and services (hotels, restaurants, medical facilities, etc.). The sales also make a significant contribution in employment and local government revenues via retail sales tax. The El Paso Branch of the Federal Bank of Dallas estimates that around 11 to 14% of all retail sales in El Paso are from Mexican visitors (Federal Reserve Bank, 2009).

MANUFACTURING INDUSTRY (MAQUILADORAS) The maquiladora industry has an important role in the economy of the borderplex. Ciudad

Juarez has 29 industrial parks where 350 maquiladoras are located. The production areas include automotive (36%), appliance (14%), biotechnology (4%), clothing (2%), electronics and telecommunications (14%), information technology (6%), forestry and furniture (8%), and others (16%). (Instituto Municipal de Investigación y Planeación, 2007). By origin, the maquiladoras are mainly from the U.S. (70%) and the rest from Canada, Europe, and Asia. The industrial parks are located in three main industrial zones in the city: Poniente (Western) with 11 industrial parks, Centro (Downtown) with 6 industrial parks, and Sur-Oriente (southeastern) with 12 industrial parks. Some maquiladoras are located outside of the industrial parks. Figure 2 shows the location of different industrial parks in Ciudad Juarez.

Maquiladoras transport their production to warehouses or directly send their finished products into the U.S. through the two ports of entry. This phenomenon has an important impact in El Paso. The payroll of American citizens that live in El Paso and commute to work in Ciudad Juarez Maquilas was $247.8 million in 2005. The other benefit is the attraction of industries to El Paso as consequence of the “production/share” practice with companies in Ciudad Juarez. This industry attraction has contributed with direct and indirect jobs, retail sales, companies’ providers of services, transportation, warehouses, etc. (Coronado, 2005).


Figure 2. Location of Industrial Parks in Ciudad Juarez.

MOVEMENT OF TRUCKS AND PASSENGER VEHICLES CROSSING THE BORDER Since 2001, border crossing times of northbound traffic coming from Ciudad Juarez through

three ports of entries has increased significantly, even though the number of private vehicles entering the U.S. is slowly declining. Increasing crossing times have also prevented Mexican visitors from crossing the border. As a result, retail activities have declined. As an example, after September 11, 2001, the crossing wait time for Texas border cities increased up to four hours and the retail was reduced by 15 to 50% (Federal Research Bank, 2001).

Higher delay at the ports of entry has a negative impact on demand for travel, which ultimately affects the economy of both cities. In addition, longer wait times increase lost hours of work, lost productivity, wasted fuel, air pollution, etc. Also noticeable is the occupancy or


number of passengers per vehicle, which has been decreasing since 2000. The occupancy has decreased from 3.12 persons per vehicle in 2000 to 2.21 persons per vehicle in 2008. However, the number of people crossing the border either walking or using buses has increased slightly. According to the Texas Center for Border Economic and Enterprise Development at the Texas A&M University, an estimated loss of $76 million in retail sales in Texas border cities could be attributed to 1% of Mexican visitors declining to cross the border into Texas border cities (Patrick, 2006).

Freight border crossing delays have a higher significance due to their effect on the maquiladoras industry; a key element of the borderplex economy. Trucks cross the border into the U.S. via two ports of entry in the El Paso area. Unlike private cars, the demand for trucks crossing through the ports of entry in El Paso is high due to the commercial activities between the two countries. Thus, the demand for trucks crossing the border is directly related to high or low production by the maquiladoras. In the last 12 years, the number of tucks crossing the border into El Paso has had a positive increase.

MOBILITY ISSUES IN CIUDAD JUAREZ Due to the growth of the maquiladoras industry after 1960, Ciudad Juarez could not keep up

with the urban planning process. The lack of planning resulted in an overloaded traffic network with considerable decrease in mobility. More specifically, decrease in mobility is mainly due to inadequate capacity, frequent accidents, etc. In spite of the growing security issues, new companies continue to set up factories in Ciudad Juarez and more people migrate to Ciudad Juarez for jobs. Most likely, the population growth will result in more congestion. Hence, the city needs to evaluate the current traffic network and evaluate how it affects the accessibility of personal and commercial vehicles, especially to industrial parks and to the ports of entry.

PROBLEM STATEMENT The growth and expansion of the city of Ciudad Juarez to the south and southeast directions

has made it so that the ports of entry are now located far from the newly developed urban areas. On the other hand, the fast expansion of the city, a result of the maquiladoras industry between the 1960s and 1990s, did not provide an ideal environment for planning and managing an urban mobility plan. Consequently, the housing developments located around the ports of entry have avenues and streets that have a limited capacity for connecting traffic from the new housing areas to the ports of entry.

In the same way, the initial location of industrial parks was defined in strategic sites with feasible access to high capacity avenues and distributed around the urban area to provide easier access to and from the ports of entry. However, because of the growth of the urban areas, the industrial parks that were once outside the city are now inside the city core. This has resulted in trucks having to use local and often congested inner city streets to transport products and supplies from/to the maquiladoras. Hence, delays for truck traffic within Ciudad Juarez have increased significantly. The delay for trucks has been exacerbated by the increasing number of accidents, limited roadway capacity, and other traffic controls.


The efficient and reliable transportation of goods is an integral part of the production process of the maquiladoras industry, so this situation affects the whole maquila industry. Worsening mobility within the city imposes a high transportation cost to companies producing and moving goods across the border.

GOALS AND OBJECTIVES OF THE PROJECT A larger goal of the project is to develop a bi-national traffic simulation model for the Ciudad

Juarez-El Paso region. The El Paso region already has a macroscopic travel demand model and mesoscopic simulation-based model, while Ciudad Juarez does not. The model developed in this project can then be combined with the existing El Paso regional models. The combined traffic simulation model has numerous short-and long-range benefits, not only for planning but also in operation of transportation infrastructure, including port of entry. This effort will produce such a model to facilitate the planning and construction of roadways to connect with the ports of entry by providing an accurate estimate of traffic movement in the area.

The objective of this particular project is to develop a traffic simulation model for the Ciudad Juarez. The model will not attempt to simulate traffic on every single roadway, but focus on major arterials in the city. This will enable analyses and evaluation of commercial freight movement between various industrial parks maquiladora in Ciudad Juarez to ports of entry on the U.S.-Mexico border. This kind of large scale traffic simulation model has never been developed for the Ciudad Juarez.


CHAPTER 2: LITERATURE REVIEW

MICROSCOPIC TRAFFIC SIMULATION Microscopic traffic simulation (MTS) models simulate the behavior of individual vehicles

within a predefined road network. They are used to estimate the likely changes in traffic patterns in the network resulting from changes to traffic flow patterns, physical improvements, etc. Microscopic traffic simulation models represent each vehicle unit (car, bus, train, bicycle, and pedestrian) with their own behavioral characteristics. Majority of MTS models employ Dynamic Traffic Assignment (DTA) to select the route choice of vehicles.

Microscopic traffic simulation employs a computer analysis based on several theories related to car following behavior, lane changing behavior, and gap acceptance. They generate pseudo-random numbers in order to produce stochastic behaviors in the model (Margison & Collins, Paramics Microsimulation Modelling Manual, 2009).

This differs from macroscopic static models, which are based on fluid-flow theory and assumes that traffic behaves like a fluid. This fluid-flow assumption averts the possibility of evaluating in detail the outcome of some traffic phenomena like delays, queues, intersections, etc., where stochastic car/diver behavior is one of the main contributing factors. A combination of the movement theories and the characterization of each vehicle create a simulation of the real network scenario in microscopic detail, with different vehicles types and driving behavior.

LARGE SCALE MICROSCOPIC TRAFFIC SIMULATION The MTS programs initially were developed to model small networks due to the software or

hardware limitations of processing the numerous events in a detailed model of a large network. MTS models are used to evaluate street networks for a relatively shorter period (e.g., peak hour) when the data can easily be obtained from using field data collection. Nevertheless, increased computer memory and speed has enabled the simulation of relatively large-scale networks. Hence, the models can be used to evaluate performance measures of the roadway network, predict scenarios under different variables such as accidents or traffic demand, optimize traffic signal design, realize accurate transportation projects, and execute evacuation and emergency detour plans. However, developing MTS for larger-scale street networks requires a much larger data set, which is costly and difficult to obtain.

The accuracy of MTS models depends on the quality of the model (road geometry and traffic control), calibration process, and the traffic demand input data (e.g., origin-destination matrix). The availability of accurate origin-destination (O-D) matrices that represent traffic demand in a MTS model (O-Dm) is crucial. However, converting O-D matrices to be used with microscopic traffic simulation is not easy to undertake. Therefore, the current practice of using large-scale MTS models employs alternatives for importing the O-Dm into the network. The most frequently used methodology is the combination of a seed O-Dm matrix and traffic counts. This methodology is called “O-D Estimation” and has the advantage of updating the seed O-Dm


matrix, which is obtained from a prior planning analysis. However, the disadvantage is the need of the seed O-Dm. (Federal Highway Administration, 2003).

The use of microscopic traffic simulation in large scale projects has been used to evaluate freeway corridors where the critical element—traffic demand—is easily obtained through counts located in ramps. The results obtained in this scenario have been accurate where a different calibration process has been developed successfully (Chu, Liu, Oh, & Recker, A Calibration Procedure for Microscopic Traffic Simulation, 2004). However, the use of microscopic traffic simulation in urban networks has been used only briefly due to different complexities. A compilation of large-scale microscopic traffic simulation projects in urban areas is mentioned in a report by John Hourdos et al., (Hourdos & Michoalopoulos, Twin Cities Metro-Wide Traffic-Simulation Feasibility Investigation, 2008). The main objective of these simulations is to support the traffic management systems and conduct feasibility studies for public transportation. Table 1 condenses compilation of the projects mentioned in this report.

From these data, the different modeling difficulties discovered were:

• Data collection is laborious and time consuming. • Modeling the geometric and signaled intersections is time consuming. • Lack of O-D matrices to be used directly in the models. • Dependency upon using travel demand planning O-D matrices to convert into MTS O-D

matrices. • Higher quantity of traffic count data and point traffic counts are needed to achieve precise

and realistic results. • Travel demand model generates inconsistencies. • Calculation processes due to low computer RAM memory is time consuming. • Feeding traffic between microscopic traffic simulation models and travel demand models.

mismatch • Travel paths generated by the O-D do not match with the real traffic pattern.

There are different software tools in the market that provide the environment to create MTS

models. Among the most common are VISSIM, Transmodeler, CUBE, AIMSUN/2, DynusT-VISSIM Converter,1 and Paramics.

1 The DynusT-VISSIM Conversion tool was developed by researchers from Texas Transportation Institute and the University of Arizona.

Table 1. Large-Scale MTS Projects Developed in Urban Areas. Project /City Year Characteristics Data collection source MTS

Software Duration/Budget Difficulties

Singapore 2006 Area: 683sq km, Length: 4,483km network,

Intersections: 10,580 (1,000 controlled), Zones: 3,153

TDM from EMME/2. Peak Hour matrices, 380 traffic count detectors to adjust OD AIMSUN 3 Years OD Calibration

Madrid, Spain Intersections: 2,109, Zones: 207, Traffic:

400,00veh/day

TDM from EMME/2. Peak Hour matrices and 15 minutes periods, 2,146 traffic count

detectors to adjust. Statistical analysis to identify and classify 15min OD matrices

AIMSUN

Columbus, Ohio, U.S. 2005

Area: 30sq miles, Intersections: 450 (330 signalized), Freeway interchanges: 25 , VISUM

zones: 294, VISSIM zones: 65 , Traffic: 100,000cars/morning peak hour

TDM from CUBE translated to VISUM, Traffic and turn counts, VISSIM

Proficient in several tools, extensive data to used in large-scale, collection data

laborious and time consuming, 720 hours to achieve convergence with 3GB of

RAM processor.

Kansas City, Kansas (HNTB) TDM from EMME/2 to VISUM/VISSIM VISSIM

Hassen, Germany 2003 Length: 1,515Km network, Zones:

245,Intersections: 462, Controlled: 200, Traffic: aprox 100,000

700 traffic detection stations AIMSUN 2 Years

TDM OD matrices were outdated. Matrix adjustment was a time consuming and difficult task. Running the model with

regular computers was a initial problem that was improved later with 5GB RAM

and AIMSUM updated version.

London, UK 2006 Length: 220 Miles, Intersections 2,000 , Controlled: 250 TDM from VISUM VISSIM

Lausanne, Switzerland 2004

length: 671km, Area: 100km2, Intersections: 1,476, Controlled:148, Zones: 292, Links: 4,100,

1,600 nodes TDM from EMME/2 AIMSUN

Several inconsistencies from TDM, low traffic count number to compare traffic

(90 points).

Sydney, Australia 2002 Area:12 Km, Intersections: 600, controlled:200, TDM and link and car park volumes Paramics

2,000 person-hours

(modeling)

Berlin, Germany (Berlin Traffic

Management Center, Siemens)

Length: 6,016km VISUM, 272km VISSIM, Intersections:1602 VISUM, 217 VISSIM,

Controlled:98, Zones: VISUM: 1,118, VISSIM: 152

VISEM (activity-based demand model), Traffic counts: 250urban and 200 Freeway. VISUM-VISIM

9 persons months for the micro-

model

Time consuming intersection coding in VISSIM,

Milwaukee, Wisconsin (FSOA)

Length: 150miles, Intersections: 40, Traffic:300,000veh/day, Zones:1000 Paramics

Munich, Germany (PTEV, German

Ministry of Research, Daimler-Chrysler,

BMW and VW)

2005 Length: 1,113km, Links: 11,080 VISEM (activity-based demand model), Model

WIVER (Freight transport survey), Traffic counts: 98 during 6 months.

VISUM-VISIM 3 Years

Collection of traffic control data, Signal groups were not modeled (pedestrian and bicyclists), OD demand calibration on hourly level, OD centroids and inflow

for microsimulation.

Manhattan, New York Area: 5sq mile

OD from TDM. Volumes detectors. Fine-tuning matching manually queue sizes in some

intersections Paramics

1.5Yrs $200K-

$300K (50% data

collection)

Calibration path choice due to grid network (different alternatives)


PARAMICS Paramics (Parallel Microscopic Simulator) is a suite of MTS tools. The different modules

within the suite are Modeller, Processor, Analyser, Estimator, Designer, Converter, Monitor, Programmer, and Urban Analytical Framework (UAF) (Quadstone, 2009).

Paramics Modeller is the main module where the network is built and simulated. The network structure in PARAMCIS is based on nodes (intersections) and links. Links that are interfaced with traffic zones are called connectors. Vehicles are loaded from traffic zones into the road network via the connectors, and travel to other traffic zones. Once the vehicles are loaded into the network, their route choice behavior to the respective destinations may follow the all-or-nothing, stochastic, dynamic feedback, or a combination of these assignment techniques as specified by the users. The traffic demands are specified in several O-D matrices, one for each class of vehicles (cars, trucks, buses, etc.).

Paramics Processor has similar functions as the Modeller but has the capability of performing simulation at a higher speed for a large network bypassing the visualization.

Paramics Analyser provides post-simulation statistical analysis of the Modeller’s results. Paramics Estimator estimates O-Dm matrices from link traffic counts. Paramics Designer is a 3D tool to develop objects to be used in the Modeller to generate 3D visualizations. Paramics Converter takes existing geometric network data from a range of digital sources and converts it into a basic Paramics network.

Paramics Programmer is an application programming interface (API) that allows users to customize new features and also develop new ones that are not provided by the Modeller such as simulation of lane closures and incidents using real-time and historic traffic conditions data. Paramics Monitor calculates the levels of vehicle emission pollution on a road network. (Aitken, Quadstone Paramics V5.1 Modeller User Guide, 2005) (Quadstone, 2009).

In this research project, Paramics was used to create a large-scale simulation model of Ciudad Juarez. Paramics was chosen due to two main reasons. Firstly, the software application has the capacity to create an O-Dm matrix from screen line counts and intersections counts since Ciudad Juarez does not have a travel demand model. Secondly, Paramics was capable of using an API to input large number of variables and parameter data for sensitivity analysis. For example, TTI has installed field devices at the BOTA (Bridge of the Americas) port of entry, which can measure crossing times and wait times of commercial vehicles entering the U.S. These data can be injected into the Paramics model using the API to influence the model, improve the visualization, and evaluate truck behavior while crossing the port of entry.


CHAPTER 3: METHODOLOGY

INTRODUCTION The methodology used to develop the MTS model of Ciudad Juarez using Paramics consists

of different stages: data collection, modeling, and calibration/validation. The flow chart shown in Figure 3 describes the process and the tasks involved. The methodology consisted of data collection from different agencies in Ciudad Juarez, coding the network and inputting traffic characteristics, generating the morning peak-hour O-Dm based on traffic intersection turn movements and traffic link flows, and finally, evaluating and calibrating the network comparing observed and simulated traffic counts data.

Figure 3. Flowchart of the Methodology.


DATA COLLECTION The data needed for developing the traffic simulation were obtained from three different

agencies in Ciudad Juarez: Instituto de Investigacion y Planeación de Ciudad Juarez (IMIP) which is a metropolitan planning agency, Centro de Información Geográfica (CIG) of the Universidad Autónoma de Ciudad Juarez, and the traffic control office of the municipality of Ciudad Juarez.

Aerial Images Aerial images covering the entire Ciudad Juarez urban area were collected to create the street

network in Paramics. These images were obtained from the Centro de Información Geográfica from the Universidad Autónoma de Ciudad Juarez. A scale picture mosaic was created with the images to provide the base of the street network coding.

Street Geometry Data The street network and street names were obtained from the Ciudad Juarez traffic control

office. Figure 4 shows a map of the street network. The geometric configuration of streets and intersections was obtained from field studies and aerial picture views from web sources such as Google Earth® and Google Maps®.

Traffic Signals The data on signal timing of traffic lights were obtained from the traffic control office. The

data covered 204 pre-timed traffic lights in the street network modeled. Currently, the city has 238 traffic lights in operation.

Speed Data Speed data were also obtained from Ciudad Juarez traffic control office, which provided a

map with speed limits. The maximum speed range in the network is 80 km/h (~50 mph) to 15 km/h (~9 mph). School zones have speed limits of 15 km/h (~9 mph), downtown streets have 30 km/h (~19 mph), local or residential streets have 40 km/h (~25 mph), collector streets have 50 km/h (~30 mph), main streets or avenues have 60 km/h (~37 mph), independence freeway and Avenida de las Torres have 70 km/h (~44 mph), and border and Camino Real freeways have a maximum speed of 80 km/h (~50 mph). Figure 5 shows a map of streets in Ciudad Juarez with speed limits.


Figure 4. Street Network Configuration of Ciudad Juarez. Source: Traffic Control department, Municipality of Ciudad Juarez


Figure 5. Speed Limit of Streets in Ciudad Juarez. Source: Traffic Control department, Municipality of Ciudad Juarez

Traffic Counts In order to track the flow of trucks traveling between maquiladoras and ports of entry, 19

turning movement data collections were done in main intersections spread in the Ciudad Juarez street network. Figure 6 shows the location of these intersections. The intersection selection criterion was based on the locations with truck routes from maquiladoras and ports of entry with access to trucks. The data were obtained during typical days in the morning peak-hour (7:00-8:00). Besides the number of vehicles in each of the legs that comprise each intersection, a classification of them was done including the categories of cars, buses, trucks, and semi-trailer.

A database of traffic counts containing 36 locations throughout the city was provided by IMIP. Figure 7 shows the location of traffic counts. These data were not sorted by vehicle; therefore they were used on links where trucks are not allowed to travel such as


downtown. Finally, traffic from BOTA and Ysleta ports of entry were obtained from field studies. The counts were done in both directions and for vehicle and truck access.

Figure 6. Intersection Movement Count Locations in Ciudad Juarez. Source: Google Maps


Figure 7. Traffic Count Locations in Ciudad Juarez. Source: Traffic Control department, Municipality of Ciudad Juarez

MODELING The modeling process involved two stages: coding the street network and obtaining the

demand (O-D) matrix to assign the traffic flow in the network.

Coding the Network The steps to create the roadway network in Paramics were: (1) create an aerial picture

mosaic, (2) draw the street and intersection configuration, (3) input signal timing control, (4) input speed information, and (5) input the right-of- way. An aerial image of the entire Ciudad Juarez was created using images collected. The process consisted of scaling the images in Paramics to match the actual dimensions of roadways and intersections.


The street network along with the proper configuration was drawn over the aerial mosaic. Once the mosaic scale was performed in Paramics, the links and nodes were added. Figure 8 shows a snapshot of the street network in Paramics. The number of lanes, width, directionality, and horizontal curvature of the streets were based on online maps (e.g., Google Maps). Turning movements were obtained from online maps as well as field data collection. The network contains 5,513 links with a total distance of 560 miles (902 km).

Signal timings and right of way information was added to traffic lights and other traffic control locations. During this process, turning movement of each lane of intersections was defined. Roundabouts were also added to the network. Finally, as part of the street network configuration, traffic restrictions on certain vehicle types were added.

The traffic demand in the network was obtained as a function of an estimated O-Dm. In order to create the O-Dm estimation, these data were used: the 126 traffic counts obtained from the intersection movement field data collection, ports of entry traffic counts, and the database provided by IMIP. The task involved a visual and traffic flow calibration process. Two O-Dm s were obtained to provide a simulation of the traffic flow of vehicles and trucks in the network.


Figure 8. Snapshot of Ciudad Juarez Roadway Network in Paramics.


Calibration and Validation A calibration process is needed to validate a MTS model. The objective of calibration is to

produce the most realistic simulation model based on the real traffic pattern/behavior as observed in the field. The calibration process involved the visual calibration and the traffic flow calibration.

The visual calibration consisted of evaluating the behavior of the traffic in the network with an initial and not precise O-D that could simply represent movements of the traffic in the network to detect inappropriate behavior of vehicles in the network. This evaluation applied to intersections, freeways, forbidden movements, and initial traffic control phasing and cycles. During the visual calibration, error checking and parameters were adjusted in the model according to observed parameters. For example, the aerial view of the “Bermudez/Ejercito Nacional” intersection was compared to the modeled intersection and validated through visual calibration as shown in Figure 9.

Figure 9. Snapshot of Bermudez Ave. and Ejercito Nacional Ave. Intersection.

Source: Google

The traffic flow calibration step consisted of comparing and fine-tuning traffic flow in the

model. The process implied the iterative calculation of an O-Dm matrix, capable of reproducing a vehicle flow pattern in the network similar to that observed in the field. The process is known as O-D estimation and employs an initial O-Dm matrix, traffic links, and turn movements. The estimation algorithm redistributes the trips among the traffic zones through an iterative process until achieving the O-Dm that mimics the modeled traffic as it was observed.


The GEH (Geoffrey E. Havers) empirical formula was used to evaluate the differences between the observed and the modeled traffic flow. This equation is a modified Chi-square test that is employed to prove the hourly traffic volumes. The representation of the equation is below:

0.5

The equation provides an auto-scaling value that makes it possible to evaluate an acceptance variance among observed (O) and modeled (M) hourly traffic flow through a link/street. The Federal Highway Administration (FHWA) suggests certain GEH criteria to validate a model (Federal Highway Administration, 2003). The FHWA criteria state that GEH values lower than 5 represent a good fit, GEH values between 5 and 10 require further investigation, and GEH values higher than 10 are considered a bad fit.

Currently there is not a consensus in regards to the calibration and validation criteria among the different modelers (AASHTO, 2010). In the current model, the result reproduced by the MTS model was considered acceptable when the network achieved the GEH value of 5 or less for at least 78% of all links with observed data.


CHAPTER 4: RESULTS

After the calibration and validation of the Ciudad Juarez network was completed, the next step was to evaluate the commercial freight movement between various industrial parks in Ciudad Juarez and the two main entrances for trucks from the city to both ports of entry. In the simulation model, origins were represented by traffic zones, which included the industrial parks while destinations were represented by traffic zones located at the entrance of the Mexican customs facilities at the ports of entry.

Measures of effectiveness (MOEs) representing traffic delay and congestion incurred to trucks on pre-defined routes connecting different origins and destinations were obtained and evaluated. These routes represent the shortest path available. The MOEs selected to evaluate the truck movement were simulated travel time and delay. Modeled travel time represents the total truck average travel time from origin to the destination. Delay represents the difference between the modeled travel time and free-flow travel time. Delay includes the average delay due to traffic congestion and traffic control devices.

Results from the analysis allowed identification of how vehicle congestion in the network affects truck movement. Traffic congestion has an especially significant effect on trucks traversing through routes that go through the inner core of the city. Also, MOEs were obtained to analyze the difference in travel time of trucks originating at different industrial parks going to the ports of entry.

Measurements of Effectiveness The MOEs were: modeled travel time and delay measurements for the different paths of

trucks traveling from industrial parks of Ciudad Juarez to the entrance of Mexican custom facilities in the Ysleta (Zaragoza bridge) and BOTA (Bridge of the Americas) ports of entry. These were extracted from different model runs. The truck traffic from Chihuahua and Casas Grandes highways entering Ciudad Juarez with destination to both ports of entry also were evaluated.

Figure 10 shows in red the path of trucks entering from other parts of the State of Chihuahua toward the BOTA port of entry. In a similar way, the paths characteristics were obtained from each of the industrial parks and highway entry points (Chihuahua and Casas Grandes) to both ports of entry. The paths obtained are the shorter routes between origins (industrial parks) and destinations (entrances to ports of entry).

A summary of the MOEs obtained from the simulation is presented in Tables 2 through 5. Table 2 includes MOEs obtained from individual industrial parks and the Ysleta port of entry. Table 3 integrates MOEs acquired between Chihuahua and Casas Grandes highways entries to Ysleta port of entry. Table 4 constitutes MOEs between individual industrial parks and the BOTA port of entry. Table 5 includes MOEs between Chihuahua and Casas Grandes highways entries to BOTA port of entry.


Figure 10. Industrial Parks Coded in the Model of Ciudad Juarez.


Table 2. MOEs between Individual Industrial Parks and the Ysleta (Zaragoza Bridge) Port of Entry.

Industrial Park Modeled Travel Time

Free Flow Travel Time Delay Length

(miles)

Juan Gabriel 0:30:55 0:19:04 0:11:51 11.33 Juarez 0:25:40 0:17:31 0:08:09 9.83 Gema 0:26:08 0:17:17 0:08:51 9.91 Gema II 0:24:14 0:15:16 0:08:58 8.91 Aztecas 0:25:00 0:15:49 0:09:11 8.94 North Gate 0:28:05 0:18:17 0:09:48 10.57 Aeropuerto 0:28:25 0:19:12 0:09:13 11.13 Panamericana 0:28:29 0:19:16 0:09:13 12.22 Industrial Center Juarez 0:25:54 0:16:57 0:08:57 10.28

AeroJuarez 0:21:30 0:14:10 0:07:20 8.09 Electrolux 0:26:23 0:21:16 0:05:07 12.23 Blvd Independencia 0:18:49 0:14:43 0:04:06 8.73 Intermex Blancas 0:17:28 0:11:28 0:06:00 6.49 Intermex torres 0:18:23 0:11:58 0:06:25 6.88 Henequen 0:21:22 0:08:34 0:12:48 4.94 Paseo de la Victoria 0:14:55 0:08:59 0:05:56 5.05 Rio Bravo 0:02:46 0:02:45 0:00:01 1.45 Satelite 0:11:44 0:07:39 0:04:05 4.32 Rivera Lara 0:26:25 0:17:18 0:09:07 9.81 Bermudez 0:16:20 0:11:57 0:04:23 6.85 Los Fuentes 0:18:55 0:16:03 0:02:52 11.03 Omega 0:19:47 0:17:18 0:02:29 12.63

Table 3. MOEs between Chihuahua and Casas Grandes Highways Entries and Ysleta

(Zaragoza Bridge) Port of Entry.

Arriving to Juarez from:

Modeled Travel Time


(miles)

Chihuahua 0:22:35 0:18:14 0:04:21 11.91 Casas Grandes 0:28:34 0:24:11 0:04:23 15.57


Table 4. MOEs between Individual Industrial Parks and the BOTA (Americas Bridge) Port of Entry.

Industrial Park Modeled Travel Time


(miles) Juan Gabriel 0:21:59 0:15:11 0:06:48 8.51 Juarez 0:20:26 0:14:51 0:05:35 7.60 Gema 0:21:59 0:16:35 0:05:24 8.65 Gema II 0:24:47 0:16:53 0:07:54 9.06 Aztecas 0:28:28 0:15:34 0:12:54 8.51 North Gate 0:34:59 0:19:34 0:15:25 11.11 Aeropuerto 0:32:38 0:17:46 0:14:52 9.94 Panamericana 0:32:10 0:20:14 0:11:56 11.51 Industrial Center Juarez 0:35:49 0:22:49 0:13:00 13.01

AeroJuarez 0:33:05 0:23:04 0:10:01 12.44 Electrolux 0:36:59 0:29:20 0:07:39 20.15 Blvd Independencia 0:29:25 0:22:48 0:06:37 16.65 Intermex Blancas 0:28:04 0:19:33 0:08:31 14.42 Intermex torres 0:32:03 0:20:16 0:11:47 11.81 Henequen 0:32:13 0:22:09 0:10:04 15.67 Paseo de la Victoria 0:23:35 0:14:39 0:08:56 8.17 Rio Bravo 0:26:24 0:19:43 0:06:41 13.98 Satelite 0:28:59 0:12:38 0:16:21 8.28 Rivera Lara 0:17:31 0:10:58 0:06:33 5.87 Bermudez 0:11:25 0:09:51 0:01:34 5.94 Los Fuentes 0:06:55 0:05:25 0:01:30 2.84 Omega 0:07:31 0:03:59 0:03:32 1.93

Table 5. MOEs between Chihuahua and Casas Grandes Highways Entries and BOTA (Americas Bridge) Port of Entry.

Arriving to Juarez from:

Modeled Travel Time


(miles)

Chihuahua 0:33:36 0:21:40 0:11:56 12.73 Casas Grandes 0:40:21 0:28:24 0:11:57 16.40


Analysis The analysis of MOEs between different industrial parks and highway entrances from Ciudad

Juarez to both ports of entry indicates that the longer the travel distance connecting them, the longer the travel time is as shown in Figure 11. However, from the delay MOE it is possible to detect that the delay for commercial vehicle traffic is not correlated with the travel distance as illustrated in Figure 12.

The delay is caused by the traffic of the network particularly in minor streets. Some trucks with longer trips use freeway corridors where the delay is lower than for shorter trips because trucks travel through urban streets that are usually congested.

Figure 11. Comparison of Simulated Travel of Commercial Vehicles between Individual Industrial Parks and Both Ports of Entry.


Figure 12. Comparison of Delay of Commercial Vehicles between Individual Industrial Parks and Both Ports of Entry.

Figure 13 shows that the average percentage of delay incurred to trucks traversing between industrial parks and the Ysleta port of entry is approximately 32% of the total travel time. In the case of the BOTA port of entry, delays represent a 35% of the total travel time. Figure 14 shows the proportion of delay for trucks from each industrial park to the BOTA port of entry.


Figure 13. Percentage of Delay Incurred to Trucks Originating from Different Industrial

Parks and Both Highway Entrances to Ysleta (Zaragoza Bridge) Port of Entry.

Figure 14. Percentage Delay Incurred to Trucks Originating from Different Industrial

Parks and Both Highway Entrances to BOTA (Americas) Port of Entry.

The simulation model also provides delay incurred by trucks on different segments that form a path of trucks traveling from the industrial parks and highway entrances to both ports of entry. Using the simulation results, segments of the route with higher delays were identified to target the segment for possible traffic improvements in order to reduce delay over the entire route (path) between industrial parks and the ports of entry. Figure 15 shows the path of the trucks traveling from the Chihuahua highway entrance to the Ysleta port of entry. This is one of the


longest routes, however, it has lower delay. Once trucks leave the freeway and enter surrounding arterials, delays are substantially increased due to intersection impedances caused by congestion, queuing, and pedestrians. The specific delay is a result of the queue at the intersection of Ramon Rayon and Clouthier as shown in Figure 16.

On other hand, the truck route that connects the industrial park Gema with the Ysleta port of entry is affected by the traffic because the trucks travel on different urban streets and have to traverse through several intersections. Figure 17 illustrates the route traversing through the inner and congested area of the city, and Figure 18 shows different locations in the city where traffic delays are prevalent.

Figure 15. Path of Trucks Traveling from the Chihuahua Highway Entrance to the Ysleta

Port of Entry.


Figure 16. Delay on Segments of a Path from the Chihuahua Highway Entrance to the

Ysleta Port of Entry.

Figure 17. Path of Trucks Traveling from Gema to the Ysleta Port of Entry.

0

0.5

1

1.5

2

2.5

3

2541

:131

5 13

15:131

3 13

13:179

3 17

93:127

5 12

75:79

79:698

69

8:25

49

2389

:87

87:261

7 26

17:266

3 26

63:290

3 29

03:287

8 28

78:290

4 29

04:299

3 29

93:288

2 28

82:288

3 28

83:288

8 28

88:289

5 28

95:288

5 28

85:240

3 24

03:289

3 28

93:289

2 28

92:89

89:90

90:158

9 15

89:158

6 15

86:158

5 15

85:158

4 15

84:91

91:158

2 15

82:93

93:94

94:156

3 15

63:156

15

6:18

1 18

1:18

2 18

2:18

3 18

3:14

21

1421

:156

6 15

66:299

4 29

94:142

0 14

20:141

9 14

19:157

1 15

71:141

8 14

18:157

0 15

70:141

7 14

17:141

6 14

16:141

5 14

15:141

4 14

14:141

3 14

13:148

8 14

88:141

2 14

12:146

4 14

64:141

1 14

11:141

0 14

10:140

8 14

08:140

9 14

09:173

7 17

37:149

1 14

91:148

9 14

89:140

0 14

00:140

2 14

02:142

2 14

22:139

3 13

93:139

5 13

95:142

4 14

24:142

5 14

25:142

6 14

26:142

8 14

28:142

9 14

29:143

0 14

30:143

1 14

31:143

2 14

32:585

58

5:58

6 58

6:13

84

1384

:588

58

8:13

83

1383

:587

58

7:58

9 58

9:59

0 59

0:59

1 59

1:59

2 59

2:59

3 59

3:59

4

Minutes

Route Links

Blvd Independencia ramp to connect Av.Torres

Blvd Independencia & J Porvenir

R. Rayon &Clouthier


Figure 18. Delay on Segments of a Path from Gema to the Ysleta Port of Entry.

In the same context but applied to the BOTA port of entry case, the graph above shows how urban street traffic affects the delay of trucks as they traverse to their destinations. Figure 19 shows the path of the trucks traveling from the industrial park Independencia to the BOTA port of entry. Delays on this path occur on freeway segments and at the intersection of Bermudez Avenue, as illustrated in Figure 20. Figure 21 shows a truck path from the industrial park Aztecas to the BOTA port of entry. This second path has a shorter distance and higher delay than the first one, which is mostly due to traffic congestion in the inner core of the city as depicted in Figure 22.

0

0.5

1

1.5

2

2.5

3

3493

:184

4 18

44:184

5 18

45:186

5 18

65:186

4 18

64:186

3 18

63:177

7 17

77:177

2 17

72:176

9 17

69:179

2 17

92:160

9 16

09:176

5 17

65:161

0 16

10:160

8 16

08:437

43

7:16

63

1663

:431

43

1:16

18

1618

:534

53

4:29

13

2913

:541

54

1:54

0 54

0:32

32

:535

53

5:28

05

2805

:533

53

3:28

04

2804

:532

53

2:53

1 53

1:53

0 53

0:52

9 52

9:52

8 52

8:52

7 52

7:52

6 52

6:52

5 52

5:52

4 52

4:52

3 52

3:52

2 52

2:52

0 52

0:52

1 52

1:51

9 51

9:51

8 51

8:51

7 51

7:27

98

2798

:516

51

6:27

99

2799

:515

51

5:51

4 51

4:62

3 62

3:51

3 51

3:51

1 51

1:50

5 50

5:50

4 50

4:46

9 46

9:25

4 25

4:57

2 57

2:57

3 57

3:57

4 57

4:57

5 57

5:57

6 57

6:57

7 57

7:57

8 57

8:57

9 57

9:58

0 58

0:14

47

1447

:581

58

1:58

2 58

2:14

46

1446

:583

58

3:58

4 58

4:58

5 58

5:58

6 58

6:13

84

1384

:588

58

8:13

83

1383

:587

58

7:58

9 58

9:59

0 59

0:59

1 59

1:59

2 59

2:59

3 59

3:59

4

Minutes

Route Links

J.Gabriel Cordillera de los Andes

Oscar Flores Clouthier & Ixtazihuatl

Clouthier & Platano

Clouthier & Hiedra

Clouthier & Ave Torres

Clouthier & G. Morin

Clouthier & Rayon


Figure 19. Path of Trucks Traveling from Independencia to the BOTA Port of Entry.

Figure 20. Delay on Segments of a Path from Independencia to the BOTA Port of Entry.

0

0.5

1

1.5

2

2.5

3

1561

:155

4 15

54:124

12

4:12

5 12

5:12

6 12

6:12

7 12

7:12

8 12

8:18

1 18

1:15

6 15

6:15

63

1563

:156

4 15

64:157

9 15

79:110

11

0:15

78

1578

:120

12

0:15

62

1562

:156

3 15

63:156

15

6:18

1 18

1:18

2 18

2:18

3 18

3:14

21

1421

:156

6 15

66:299

4 29

94:142

0 14

20:141

9 14

19:157

1 15

71:141

8 14

18:157

0 15

70:141

7 14

17:141

6 14

16:141

5 14

15:141

4 14

14:141

3 14

13:148

8 14

88:141

2 14

12:146

4 14

64:141

1 14

11:141

0 14

10:140

8 14

08:140

9 14

09:173

7 17

37:140

7 14

07:140

6 14

06:138

9 13

89:138

8 13

88:138

7 13

87:137

7 13

77:137

6 13

76:138

2 13

82:137

5 13

75:137

2 13

72:137

1 13

71:137

0 13

70:136

5 13

65:161

7 16

17:136

4 13

64:136

3 13

63:136

2 13

62:136

8 13

68:136

7 13

67:136

6 13

66:136

1 13

61:136

0 13

60:135

9 13

59:135

8 13

58:135

7 13

57:135

6 13

56:135

5 13

55:135

4 13

54:134

8 13

48:134

7 13

47:134

5 13

45:134

9 13

49:134

6 13

46:134

4 13

44:708

70

8:13

32

1332

:133

1 13

31:133

6 13

36:101

6 10

16:101

5 10

15:101

7 10

17:101

8 10

18:101

4 10

14:102

2 10

22:103

7 10

37:102

8 10

28:104

6 10

46:106

7 10

67:103

9 10

39:106

6 10

66:104

0

minutes

Route Links

Bvld Independencia laterals & Return 4 Siglos &Fco Villarreal

Overpass :4 Siglos&Bermudez (queue)


Figure 21. Path of Trucks Traveling from Aztecas to BOTA Port of Entry.

Figure 22. Link Delay of the Path of Trucks Traveling from I.P. Aztecas to BOTA Port of

Entry.

0

0.5

1

1.5

2

2.5

3

2335

:176

6 17

66:174

2 17

42:241

24

1:16

13

1613

:161

2 16

12:161

1 16

11:160

9 16

09:176

5 17

65:161

0 16

10:160

8 16

08:179

1 17

91:651

65

1:27

03

2703

:11

11:270

5 27

05:321

5 32

15:321

8 32

18:321

4 32

14:189

8 18

98:131

8 13

18:12

12:13

13:131

9 13

19:14

14:269

3 26

93:269

2 26

92:197

19

7:32

43

3243

:15

15:290

1 29

01:268

6 26

86:268

9 26

89:38

38:268

8 26

88:47

47:51

51:50

50:49

49:39

39:40

40:414

7 41

47:16

16:267

4 26

74:17

17:267

6 26

76:267

5 26

75:267

9 26

79:268

1 26

81:18

18:268

3 26

83:267

3 26

73:267

2 26

72:120

8 12

08:52

52:286

7 28

67:53

53:245

24

5:19

19

:20

20:21

21:993

99

3:22

22

:988

98

8:23

23

:24

24:25

25:411

4 41

14:27

27:69

69:997

99

7:26

71

2671

:100

1 10

01:998

99

8:26

70

2670

:100

9 10

09:100

2 10

02:100

3 10

03:101

4 10

14:102

2 10

22:103

7 10

37:102

8 10

28:104

6 10

46:106

7 10

67:103

9 10

39:106

6 10

66:104

0

Minutes

Route Links

J. Gabriel&Ponciano Arriaga

O. Flores&Ponciano Arriaga

C Santos& Tecnologico

Tecnologico & T. Borunda

Tecnologico & R. Lara

Tecnologico & Chinameca

Tecnologico & P.Rosales

Tecnologico & V.Guerrero

Tecnologico & P.Rosales

P Serna & H Escobar


Other results provided by the MOEs were the travel time comparison between each industrial park and both ports of entry. Figure 23 shows both options. Similarly, Figure 24 illustrates the comparison of delay between each industrial park and both ports of entry.

Figure 23. Modeled Travel Time Originating from Different Industrial Parks and Both

Ports of Entry.

Figure 24. Delay Time Originating from Different Industrial Parks and Both Ports of

Entry.

0:00:00

0:07:12

0:14:24

0:21:36

0:28:48

0:36:00

0:43:12Juan

Gabriel

Juarez

Gem

a

Gem

a II

Aztecas

North Gate

Aerop

uerto

Panamericana

Indu

strial Cen

ter Juarez

AeroJuarez

Electrolux

Blvd

Inde

pend

encia

Interm

ex Blancas

Interm

ex torres

Hen

eque

n

Paseo de

la Victoria

Rio Bravo

Satelite

Rivera Lara

Berm

udez

Los Fue

ntes

Omega

Chihuahu

a

Casas Grand

es

Mod

eled

Travel (hr:m

in:sec)

Ysleta

BOTA

0:00:00

0:02:53

0:05:46

0:08:38

0:11:31

0:14:24

0:17:17

Juan

Gabriel

Juarez

Gem

a

Gem

a II

Aztecas

North Gate

Aerop

uerto

Panamericana

Indu

strial Cen

ter Juarez

AeroJuarez

Electrolux

Blvd

Inde

pend

encia

Interm

ex Blancas

Interm

ex torres

Hen

eque

n

Paseo de

la Victoria

Rio Bravo

Satelite

Rivera Lara

Berm

udez

Los Fue

ntes

Omega

Chihuahu

a

Casas Grand

es

Delay

(hr:min:sec)

Ysleta

BOTA


CHAPTER 5: CONCLUSION

The maquiladoras industry is an important element in the economy of the El Paso/Ciudad Juarez borderplex, and a crucial element in its process for reliable transportation of goods and people. The growth of the urban area and population of Ciudad Juarez has worsened truck mobility within the city due to the delays incurred by accidents, inadequate capacity, and traffic controls. The worsening traffic congestion affects the entire maquiladora industry imposing high transportation costs to companies produce and move goods across the border.

The objective of the research project was to highlight the importance of using a large scale microscopic traffic simulation model to analyze the mobility of commercial vehicles within the city, especially between the industrial parks in Ciudad Juarez and various ports of entry.

This project developed a large-scale traffic simulation model for Ciudad Juarez to analyze commercial freight routes from industrial park areas in Ciudad Juarez to the ports of entry on the U.S-Mexico border. The simulation model included major roadways that connect industrial parks, where the majority of the maquiladoras are established, to ports of entry. Traffic demand of autos and trucks was produced through an O-D estimation process that was based on traffic counts and traffic intersection movement studies in the city. However, for simplicity, the public transportation network was not included in the simulation model.

The simulation model developed in this research demonstrated various scenarios of truck paths connecting industrial parks to ports of entry. The analysis clearly showed that the delay on truck paths is attributed to congestion in the inner urban areas, while freeways surrounding the city provide reliable access to the ports of entry.

The research also showed that the delay contributes to 20–60% of the travel time of trucks moving between industrial parks and the ports of entry. This is unacceptable for carriers and shippers since the burden of delay translates into a higher cost for shipping and moving goods. Hopefully, the traffic simulation model and the results derived from the model will relay a strong need for improved planning and operation of transportation infrastructure in Ciudad Juarez.

As part of the future development of the model, it is highly recommended that several improvements be made, one of which is to include public transportation in the network. Also, calibration of this model was performed with a small number of field traffic counts and the researchers believe that additional counts will definitely improve the reliability of the model.


REFERENCES

AASHTO. (2010). Best Practices in the Use of Micro Simulation Models. American Association of State highway and Transportation Officials. Aitken, S. (2005). Quadstone Paramics V5.1 Modeller User Guide. Edinburgh: Quadstone Limited. Chu, L., Liu, H., Oh, J.-S., & Recker, W. (2004). A Calibration Procedure for Microscopic Traffic Simulation. Washington D.C.: Transportation Research Board. Coronado, R. (2005). Maquila Impact on Regional Manufacturing. 2005 El Paso Regional Economic Summit. El Paso Regional Economic Development Corporation. (2010). El Paso Regional Economic Development Corporation. Retrieved 2010, from Production Sharing: http://www.elpasoredco.org/ElPaso-ProductionSharing.aspx Federal Highway Administration. (2003). Traffic Analysis Toolbox Volume III: Guidelines for Applying Traffic Microsimulation Software. Washington, D.C. Federal Research Bank. (2001, November/December). Southwest Economy Regional Update. Retrieved 2009, from Federal Reserve Bank of Dallas: http://dallasfed.org/research/swe/2001/swe0106d.pdf Federal Reserve Bank. (2009, July). Federal Reserve Bank of Dallas, El Paso Branch. Retrieved 2009, from Economic Update El Paso: http://www.dallasfed.org/research/update-ep/2009/0902epupdate.pdf FHWA . (2003). Traffic Analysis Toolbox Volume III: Guidelines for Applying Traffic Microsimulation Software. Washington, D.C.: Federal Highway Administration. FRB. (2009, July). Federal Reserve Bank of Dallas, El Paso Branch. Retrieved 2009, from Economic Update El Paso: http://www.dallasfed.org/research/update-ep/2009/0902epupdate.pdf FRB. (2001, November/December). Southwest Economy Regional Update. Retrieved 2009, from Federal Reserve Bank of Dallas: http://dallasfed.org/research/swe/2001/swe0106d.pdf Hourdos, J., & Michoalopoulos, P. (2008). Twin Cities Metro-Wide Traffic-Simulation Feasibility Investigation. University of Minnesota. St. Paul, MN: Minnesota Department of Transportation. INEGI. (2009). Conteo de Población y Vivienda 2005. Retrieved 2009, from Instituto Nacional de Estadistica y Geografia: http://www.inegi.org.mx/est/contenidos/espanol/sistemas/conteo2005/localidad/iter/default.asp?s=est&c=10395


Instituto Municipal de Investigación y Planeación. (2007). Resumen Ejecutivo del Analisis Estructural del Empleo en Ciudad Juarez. Ciudad Juarez. Margison, P., & Collins, P. (2009). Paramics Microsimulation Modelling Manual. Roads and Traffic Authority. Patrick, M. (2006). The Impact of 9/11 and the U.S. VISIT Program on U.S. Border Retailing. San Antonio, Texas: Federal Reserve Bank of Dallas. Quadstone. (2009). Quadstone Paramics. Retrieved 2009, from http://www.paramics-online.com/ REDCo. (2009). El Paso Regional Economic Development Corporation. Retrieved 2009, from Production Sharing: http://www.elpasoredco.org/ElPaso-ProductionSharing.aspx U.S. Census Bureau. (n.d.). Population Estimates. Retrieved August 2009, from County Population Estimates: http://www.census.gov/popest/counties/CO-EST2008-07.html

development of ciudad juarez traffic simulation model in ... · of ciudad juarez will grow to 1.42...

Documents