18. making decisions based on multiple criteria

Upload: awesomeprussia

Post on 02-Jun-2018

219 views

Category:

Documents


0 download

TRANSCRIPT

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    1/37

    Transportation Decision MakingPrinciples of Project Evaluation and Programming

    Chapter 18

    Evaluation of Transportation Projectsand Programs Using Multiple Criteria

    Kumares C. Sinha and Samuel Labi

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    2/37

    Decision criteria can have multiple dimensions

    Dollars

    Number of crashes

    Acres of land, etc.

    All criteria are not of equal importance

    For a given criterion, different stakeholders may havedifferent weights.

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    3/37

    Typical Steps in Multi-Criteria

    Decision Making

    1. EstablishTransportation

    Alternatives

    3. EstablishCriteria Weights

    4. Establish Scale to be Used forMeasuring Levels of Each Criterion

    5. Using Scale, Quantify Level(Impact) of Each Criterion for Each

    Alternative

    2. EstablishEvaluation Criteria

    6. Determine Combined Impact of allWeighted Criteria for Each Alternative

    Weighting

    Amalgamation

    Scaling

    11. Determine the Best Alternative

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    4/37

    Establishing Weights

    Weights reflect the relative importance attached bydecision makers to various criteria

    In some cases, the decision maker refers to theagency as well as the facility user. In those cases, the

    weight used for each criterion is a weighted average ofthe weights from these two parties.

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    5/37

    Weighting Techniques

    1. Equal Weights2. Direct Weighting

    3. Derived Weights

    4. Delphi Technique

    5. Gamble Method

    6. Pair-wise comparison: AHP

    7. Value Swinging

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    6/37

    Equal Weights - Example

    Project Cost 33.3%

    Travel Time Saving 33.3%

    VOC Saving 33.3%

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    7/37

    Direct Weighting

    1. Point Allocation A number of points areallocated among the criteria according to theirimportance.

    2. Ranking Simple ordering by decreasingimportance.

    Point allocation is preferred because unlike ranking, ityields a cardinal rather that an ordinal scale ofimportance.

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    8/37

    Point Allocation (0-100) Ranking

    (Cardinal) (Ordinal)

    Project Cost 70 1

    TT Saving 50 3

    VOC Saving 60 2

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    9/37

    Regression-Based Observer-Derived Weighting

    1. Survey respondents assign scores of overallbenefit or desirability for a given combinationof criteria levels achieved by each alternative

    2. Weights are then the resulting regressioncoefficients

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    10/37

    ( )

    2

    2

    i

    i j ji i

    j

    Minimize

    TV w V

    = +

    i = alternative

    j = Criterion

    TV = score or desirability

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    11/37

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    12/37

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    13/37

    Regression

    7 Respondents

    21 Data Points

    TV = wcost* Cost + wtime * Time

    wcost = 0.214

    wtime = 0.786

    R2 = 0.98

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    14/37

    Delphi Technique

    Individual responses aggregated

    Effect of assessment of other respondents

    Consensus building

    Iterative, generally 2 rounds to achieve stablevalues

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    15/37

    Scaling Methods

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    16/37

    GAMBLE METHOD

    1. Carry out an initial ranking of all criteria in order of decreasingimportance. set the first criterion at its most desirable level and

    all other criteria at their lest desirable levels

    2. Compare between the following two outcomes: Sure thing: The outcome is that the criterion in question is at its

    most desirable level while all other criteria at their least desirablelevels

    Gamble: In this outcome, all criteria attained their most desirablelevels p% of the time, their least desirable levels (1-p)% of the time

    3. At a certain level of p the two situations (sure thing and gamble)are equally desirable. At that level, the value of p representsthe weight for the criterion in question

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    17/37

    Example:

    Bus Route Assessment

    Headway (from 5 to 15 minutes)

    Population Served (from 5,000 to 10,000)

    Solution:

    1. Sure Thing: Bus headway is 5 minutes and population served is 5,000

    2. Gamble: Two outcomes:a. A p% chance of an outcome that headway is 5 minutes and

    population is 10,000

    b. A (1-p)% chance of an outcome that headway is 15 minutesand population is 5,000.

    Suppose the respondent were found to be indifferent between sure

    thing and gamble at p = 60%, then, the relative weight for busheadway is 0.6.

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    18/37

    Pair Wire Comparison

    Analytical Hierarchy Process (AHP)

    12 1

    12 2

    1 2

    1 ...................1/ 1 ...................

    ... .... .... .... .... .... .......1/ 1/ ..... ... .... 1

    n

    n

    n n

    a aa a

    a a

    = relative importance of two criteria I and j on the basisof a scale of 1 to 9

    =

    ija

    /i jw w

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    19/37

    Table 18.1: Ratios for Pair wise

    Comparison Matrix

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    20/37

    Value Swinging Method

    1. Consider a hypothetical situation where all criteria at

    their worst values2. Determine the criterion for which it is most preferred to

    swing from its worst value to best value, all other

    criteria remaining at their worst values.3. Repeat steps 1 and 2 for all criteria.

    4. Assign the most important criterion the highest weight

    in a selected weighting range (100 for 1-100 scale) andthen assign weights to the remaining criteria in

    proportion to their rank of importance.

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    21/37

    Scaling of Performance Criteria

    Certainty - Value Function

    Risk - Utility Function

    Uncertainty - Scenario Analysis

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    22/37

    Value Function

    a. Direct Rating Method direct assignment

    of value to various levels of a criterion

    b. Mid Value Splitting Technique based onindifference between changes in levels ofcriterion.

    c. Regression based on data from directrating

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    23/37

    Discrete Value Function

    Discrete Dis-Utility Function for Performance Measure

    of Impact on Natural, Socio-Economic, Historical &Cultural Resources

    -100

    -80

    -60

    -40

    -20

    0

    No

    Impa

    ct

    Minor

    Impa

    ct

    Mod

    erate

    Impact

    Major

    Impa

    ct

    Hug

    e

    Impa

    ct

    Extreme

    Imp

    act

    Utility

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    24/37

    Continuous Value Function

    Dis-Utility Function for Single Performance Measure

    of Emissions

    -100

    -80

    -60

    -40

    -20

    0

    0 20 40 60 80 100 120

    Percentage Increase in Emissions

    Utility

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    25/37

    Developed Value Functions

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    26/37

    Utility Function

    Direct Questioning Using the GambleApproach

    Guaranteed prospect of an outcome vs. riskyprospect of a more favorable outcome.

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    27/37

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    28/37

    Example 18.7

    Utility Functions for agency cost, ecological damage, and vulnerablepopulation served.

    Solution:

    For Agency Cost: Ucost ($30 Million) = 0 (Worst)

    Ucost

    ($ 0 Million) = 1 (Best)

    Sure Thing: The outcome is that agency cost is guaranteed to be $20 Million

    Gamble: There is a 50% chance that cost is 0 and 50% change it is $30 Million

    X50 = $20 Million is the Certainty Equivalent because the expected utility is 0.5

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    29/37

    0

    20

    40

    60

    80

    100

    120

    0 5 10 15 20 25 30 35 40 45

    Criteria Level

    Utility

    0

    20

    40

    60

    80

    100

    120

    0 5 10 15 20 25 30 35 40 45

    Criteria Level

    Utility

    Cost (in $millions)

    Wetland lost in acres (in tens)

    Population served (in thousands)

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    30/37

    Combination of Performance Criteria

    Pareto Optimality

    Difference Approach

    Net Utility = U(B) U(C)

    NPV = PV (B) PV(C)

    Ratio Approach Utility Ratio = U(B)/U(C)

    BCR = PV(B) / PV(C)

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    31/37

    Cost Effectiveness

    Costs and Benefits are not necessarilyexpressed in the same metrics

    Indifference Curves

    Tradeoffs marginal rates of substitutionbetween criteria

    TV = 2*TTR + PCC

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    32/37

    Indifference Curves Using Mathematical Form of

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    33/37

    Indifference Curves Using Mathematical Form of

    Utility/Value Function for Combined Performance

    Measures

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    34/37

    Ranking and Rating Method

    i i j ij

    jScore P w O For each i=

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    35/37

    Impact Index Method

    1 2

    1max( , ,........, )

    tan

    ( 0.5 0.5)

    i j j ij j j j ijj

    jj

    j

    j

    jj j nj

    j

    I R S X e R S X

    wR relativeweight for criterion j

    w

    S scaling factor of measurement X of criterion jX X X

    e RN drawn fromarec gular distribution

    e

    = +

    = =

    = =

    =

    +

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    36/37

    Table E18.10.1: Performance of Alternatives

    Figure E18 10: Plot of Confidence

  • 8/10/2019 18. Making Decisions Based on Multiple Criteria

    37/37

    Figure E18.10: Plot of Confidence

    Intervals