Can Traffic Simulation Models Contribute on Mobility Management Evaluation

1
13th European Conference on Mobility Management
Cost Benefit and Evaluation of Mobility
Management
Can Traffic Simulation Models Contribute on
Mobility Management Evaluation? A Conceptual
Analysis

Kursaal Congress Center 13-15 May
2009 Donostia San Sebastian Spain
Panos Papaioannou Professor
Socrates Basbas Ass. Professor
Ioannis Politis Ph.D Candidate
2
PRESENTATION OUTLINE

• Objectives and Applications of Transport
  Planning Tools
• Transportation Models and Benchmarking
  Evaluation
• Introducing TPT into Mobility Management
  Evaluation
• Conclusions and Discussion
• Annex Case Study Classical Approach

3
KEY QUESTION
Why it is Important to Use Transportation
Planning Software Tools ??
4
REASONS

• Transportation System Complex multi-dimensional
  factors
• not easily determined, measured or estimated
  directly
• Impact Estimations (ex ante!) derived from
• the construction of a new road infrastructure
• or operation of a new transport mode,
• or.implementation of a MM plan!
• Impact Estimations
• - The transportation system itself
• - The environmental effects and the potential
  revenues
• - The redistribution of the land use
• Easier to Introduce Transport Planning Theories

5
OBJECTIVE OF TRANSPORT PLANNING SIMULATION
TOOL

• To represent with accuracy the underlying
  operation of
• the transport system
• (in terms of traffic conditions and travel
  patterns)
• To create reliable mathematical models for
  testing
• different / various schemes at the base year
  (underlying)
• or at future years (planning horizons )
• These schemes pertain to be at the supply (new
  
• infrastructure, new mode, pedestrialization
  of roads etc)
• or the demand (car pooling, flexible working
  hours etc)
• side

6
OBJECTIVE OF TRANSPORT PLANNING SIMULATION
TOOL

• A simulation traffic model can estimate the
  impacts derived from a Mobility Management
  Measure, primarily on the demand changes.
• In fact, a MMM (such as car pooling, van
  pooling, flexible or staggered working hours
  etc.) is translated into changes at the Origin
  Destination Matrices of each respective demand
  segment and changes in travel chain in general.
• An evident disadvantage is that existing
  simulation tools just simulate the anticipated
  improvements of a network. The reality proves
  that when the traffic conditions are improved new
  (generated) traffic is added (the vicious circle
  of the transportation systems)

7
TRAVEL PATTERNS EXAMPLE
Production trips
Attraction Trips
8
APPLICATIONS OF TRANSPORT PLANNING SOFTWARE
TOOLS

• Traffic and Transportation Studies
• Feasibility (Socio Economic) Studies
• Cost Benefit Studies
• Urban Planning Studies
• Environmental Studies

• Mode Choice and Travel Behavior Studies!!

9
Transportation Models and Benchmarking Evaluation
10
Transportation Models and Benchmarking Evaluation

• According to the HCM (2000) a transportation
  model is
• A computer program that uses mathematical models
  to conduct experiments with traffic events on a
  transportation facility or system over extended
  periods of time
• Transportation Models Classification
• According to their application area
• According to the level of presentation of
  the traffic flows
• According to the time period of the
  analysis

11
Transportation Models Classification
12
Macroscopic Models

• Take into account transportation network
  attributes
• such as capacity, speed limit, flow and
  density
• Simulate large scale facilities (highways,
  regions etc)
• No need to track individual vehicles (aggregate
  theory)
• No detailed information about road design and
  signal
• plans is needed
• CUBE, TRIPS and VISUM

13
Mesoscopic Models

• Take into account the actual road geometry and
  signal
• timing plans
• Simulate intersections in a corridor or city
• Simulate individual vehicles
• Describe activities based on aggregate or
  macroscopic
• level
• SATURN, CORSIM, TRANSCAD, EMME/3, AIMSUN

14
Microscopic Models

• Simulate characteristics and interactions of
  individual
• vehicles
• Study area Intersection or a road segment
• (e.g. a corridor )
• Enclose theories and rules for vehicle
  acceleration,
• passing manoeuvres and lane-changing
• PARAMICS, VISSIM, AIMSUN

15
Comparative Analysis of the most commonly used
transportation software
16
Existed Transport and Simulation Models
The analysis is based only on Quantitative
Data/Results !!
17
KEY QUESTIONS

• What are the user needs of the study area?
• How much dependent the users are to their cars?
  
• What will be the overall impacts of a real
  Mobility
• Management Measure (MMM) to the Study area
• Which MMM is the most promising to this
  specific area
• Which are the potential barriers to implement
  them?
• The Qualitative or Quantitative data should be
  taken into
• consideration most? The same?

18
A Conceptual Framework of Introducing
Transportation Models into Mobility Management
Measures Evaluation and Classification
19
(No Transcript)
20
Planning Phase

• The MMM that will be examined should be linked
  with the
• trip purposes of the study area (different
  demand matrices)
• Why not to follow the categorization of MMM
  derived from
• MAX project?
• A well structured questionnaire should
• Estimate the behavioral stage of the
  targeted
• population (why not the diagnostic
  questions?)
• Identify the user needs (that wanted or
  expected) and
• the level of acceptance of the examined MMM
  through
• well knownused techniques

21
Planning Phase

• The criteria of evaluation should be clearly
  determined
• Transportation indices (VKT, Speed, Delays etc.)
• Environmental indices (CO, HC, NOx etc.)
• Level of maturity (Low, Medium, High)
• Change on Behavioral Stage (0 stage, 1 stage, 3
  stages)
• The selection of the appropriate Transportation
  Model
• should be based on
• The criteria of evaluation
• The area under consideration (macro,meso,micro)

22
Analysis Phase

• The criteria and sub-criteria (quantitative
  and qualitative)
• should get an evaluation grade
• All the criteria should also obtain weights
  (experts survey)
• Well know multi criteria decision analysis tools
  (MCDA)
• could easily apply the weights to the grades
  
• ( software HIPRE 3, web-HIPRE, EXPERT
  Choice model)

23
Classification Phase

• The evaluation grade for the qualitative
  criteria are based
• on subjective judgment
• Various techniques can quantify the qualitative
  criteria
• ( e.g. Evidentional Reasoning Approach)
• If the initial evaluation criteria are properly
  selected, then
• the final ranking of the MMM will include
  qualitative
• parameters such as the trip purpose, the
  behavioral stage
• etc. which are not included in conventional
  evaluations
• Alternatively, the proposed methods could be
  classified
• through a cost benefit analysis (all the
  benefits are
• translated into momentary units classical
  approach)

24
CONCLUSIONS

• Mobility Management seems to be adopted more and
  more by local authorities
• It is important to have accurate estimations
  about the most promising MMM before moving out
  of the office
• The classical transportation planning theory
  cannot include qualitative parameters especially
  from the behavioural psychology side

25
CONCLUSIONS

• These parameters are equal important since can
  affect the effectiveness of a measure
• A new framework should be established combining
  the knowledge obtained from transportation
  planning theories and psychology behavioural
  science

26
Thank you for your attention!!
Ioannis K. Politis -------------------------------
------ Ph.D. Candidate Laboratory of
Transportation and Construction
Management Department of Civil Engineering
Aristotle University of Thessaloniki,
Greece pol_at_civil.auth.gr
27
ANNEX
Case Study The use of a mesoscopic traffic
analysis model in order to run alternative road
charging schemes at the Outer Ring Road of
Thessaloniki
28
THE STUDY AREA
29
THE STUDY HIGHWAY

• 35 km length freeway
• Estimated budget of 700 million euros
• Will offer connections to the Inner Ring Road
• 13 Bridges with a total length of 2 km
• 20 Tunnels with a total length of 20 km
• 9 Interchanges
• Completion date 2016

30
THE STUDY HIGHWAY
31
THE EVALUATION MODEL

• Mesoscopic Model SATURN (Simulation and
• Assignment of Traffic to Urban Road
  Networks)
• Extended network was coded (base year 2006)
• 783 simulation nodes including
• 27 external nodes
• 310 priority junctions
• 292 traffic signals
• 154 dummy nodes
• 2508 simulation links
• 6350 simulation turns
• 210 traffic zones
• Morning Peak Period 0800-0900
• Ap. 200 traffic counts were used for
  calibration purposes
• (180 for new O-D matrix estimation and 20
  for validation)

32
THE EVALUATION MODEL
Modeled vs Observed Flows
33
SCENARIOS TESTED
34
2006 BASE YEAR NETWORK
35
2016 PLANNING YEAR NETWORK
36
DETAILED REPRESENTATION OF THE INTERSECTIONS
IC 1-2 Interchange to the Inner Ring Road
37
DETAILED REPRESENTATION OF THE INTERSECTIONS
IC 6 Panorama
38
NUMERICAL RESULTS
39
NUMERICAL RESULTS
40
NUMERICAL RESULTS
41
NUMERICAL RESULTS
42
NUMERICAL RESULTS
43
KEY FINDINGS OF THE STUDY

• The distance based tolls frustrate journeys gt 20
  km
• The average journey length varies between
  12-15 km for
• all the methods and toll rate levels
  examined
• The demand is inelastic (- 1 lt e lt 0) for all
  the examined
• scenarios, especially for the East West
  Direction
• Flat tolls schemes lead into more elastic
  interrelations
• with respect to demand (actual flow)

44
KEY FINDINGS OF THE STUDY
OBTAINED REVENUES

• Flat Tolls The optimum toll value should be
  greater than
• 2 euros
• Higher toll level will lead
  to lower actual flows
• and accordingly to bigger
  obtained revenues
• Distance Based Tolls The optimum toll value
  should be
• lower
  than 0.087 euros/km
• Lower
  toll level will lead to higher
• actual
  flows and accordingly to
• bigger
  obtained revenues