Travel Demand Modeling

1
Travel Demand Modeling

• Champaign Urbana Urbanized Area Transportation
  Study (CUUATS)

2
Champaign County Regional Planning Commission
(CCRPC)

• Planning agency for county
• Major technical entity for local governments
• Metropolitan Planning Organization (MPO) for the
  urbanized area
• Administrative, intergovernmental policy
  development, planning, economic development,
  social services and Head Start, regional
  (multi-county) services

3
Champaign Urbana Urbanized Area Transportation
Study (CUUATS)

• Transportation entity of CCRPC
• Member agencies - Champaign, Urbana, Savoy,
  University, CUMTD, Champaign County, IDOT
• Primary activities Long Range Transportation
  Plan, Transportation Improvement Program (TIP)
  and Unified Technical Work Program (UTWP)
• Other activities traffic impact, travel time,
  intersection and corridor studies, crash
  analysis, warrant analysis, greenways, pedestrian
  and bike plans

4
What is travel demand modeling?

• Forecast traffic flows on a transportation system
• Tool for evaluating major infrastructure
  investments (roadway improvements or improvements
  to bus service) in Long Range Transportation
  Plans (LRTPs), regional transportation and
  corridor studies
• Practiced by regional transportation agencies,
  MPOs and State DOTs

5
History of travel demand modeling

• Practice of travel-demand forecasting is roughly
  35 years old
• In U.S., modeling started with area wide
  transportation studies in Chicago and Detroit
• NCHRP 187 Quick-Response Urban Travel Estimation
  Techniques and Transferable Parameters, TRB, 1978
• NCHRP 365 Travel Estimation Techniques for Urban
  Planning, TRB, 1998

6
Defining Questions

• Why we travel?
• Where we travel?
• When we travel?
• How we travel?
• What direction we travel?

7
Four Step Modeling

• Most commonly used and simple modeling process
• Four Steps
• Trip Generation
• Trip Distribution
• Mode Choice
• Trip Assignment

8
Major Inputs

• Travel is Spatial
• Define model area
• Define highway network
• Define transit network

9
Model Area
10
Traffic Analysis Zones (TAZs)

• Geographical socio-economic data blocks
• Defined based on land use characteristics
  (population and employment), physical boundaries,
  census blocks and major roadways
• Current CUUATS model has 206 TAZs

11
CUUATS Model TAZs
12
Highway Network

• Highway network
• Node
• Link
• Centroids
• Centroid connectors
• External stations
• Nodes represent intersection points or other
  changes in the roadway geometry, functional
  classification and operational characteristics
• Links represent the roadway segments. There is
  one link for each direction of the roadway
  segment

13
Highway Network

• Centroid is a point which represents the
  concentration of population and employment within
  the TAZ
• Centroids are connected to the major roadways on
  the boundary of the TAZ through centroid
  connectors. Centroid connectors represent
  collection of local access streets
• External stations are points of entry and exit
  for vehicles entering, exiting or traveling
  through the study area. External stations are
  located on major highways bringing traffic to the
  model area

14
CUUATS Model Highway Network
15
Transit Network

• Representation of all transit routes in the
  region
• Transit network
• Transit lines
• Access/egress links
• Stops
• Based on the highway network
• Uses the highway network links for transit lines
  and nodes for stops

16
CUUATS Model Transit Network
17
Socio-Economic Data

• Data at TAZ level
• Data source Census, local survey
• Population data (each TAZ)
• of residents
• of one, two, three and four person households
• Employment Data (each TAZ)
• of retail employees
• of service employees
• of other employees (government, industries and
  institutions)

18
CUUATS Model Socio-Economic Data
19
CUUATS Model Socio-Economic Data
20
Highway Network Data

• Node data
• Turn penalty
• Link data
• of lanes
• Facility type
• Area type
• Distance
• Speed
• Capacity
• External station data
• ADT (Average Daily Traffic)
• Thru trips percentage

• Highway network data
• Node data
• Link data
• External station data

21
CUUATS Model Highway Network Data

• Node Data Turn penalty
• from node, thru node, to node and delay
• 3 sec. for right turn and 6 sec for left turn
• Link Data Facility type
• Type of roadway facility
• Eight facility types currently used
• 0 centroid connector
• 1 local roadway
• 2 collector roadway
• 3 minor arterial roadway
• 4 major arterial roadway
• 5 ramp
• 6 interchange ramp
• 7 freeway

22
CUUATS Model Facility Type
23
CUUATS Model Highway Network Data

• Link Data Area type
• Indicates the area type of the TAZ closest to the
  link
• Six area types currently used
• 1 very high density commercial/residential
• 2 high density commercial
• 3 moderate density commercial
• 4 high/moderate density residential
• 5 very low density commercial/residential

24
CUUATS Model Area Type Data

• TAZ area type is based on one mile radius
  floating population/employment density

CUUATS Model TAZ Area Type
25
CUUATS Model Area Type Data
26
CUUATS Model Speed/Capacity Data

• Link Data Speed/Capacity
• It is based on HCM 2000 chapter 30 and other
  similar model literature
• Speed and capacity are function of facility type
  and area type
• Free flow speed and level of service E capacity

27
CUUATS Model Speed/Capacity Data
Speed (mph) / Capacity (pcphpl)
28
CUUATS Model Transit Network Data

• Mode data
• Mode 1 CUMTD Local System
• Mode 2 CUMTD Express System (used only for
  future conditions)
• Mode 3 CUMTD Campus System
• Mode 11 Walk Access/Egress
• Vehicle data
• Vehicle Type 1 - 40 ft. New Flyer, seating
  capacity 36, standing capacity 25
• Vehicle Type 2 - 60 ft. New Flyer, seating
  capacity 45, standing capacity80
• Vehicle Type 3 - 30 ft. EZ Rider, seating
  capacity 33, standing capacity15

• Transit network data
• Mode
• Vehicle type
• Headway
• Capacity
• Speed

29
Four Step Modeling Process
30
CUUATS Model Trip Generation

• Trip is a journey between two points
• Trips are produced at households and attracted
  towards employment centers
• Production and attraction trips in CUUATS model
  are further classified into five different
  classes
• Home Based Work
• Home Based School
• Home Based Shopping
• Home Based Other
• Non-Home Based

31
Production and Attraction Trips Classification
32
CUUATS Model Trip Generation

• The household data for each TAZ is converted into
  number of trips produced using trip production
  rates
• Trip production rates used in CUUATS model were
  obtained from the household travel diaries
  (survey) conducted by CUUATS in 2003
• Trips attraction rates given in NCHRP 365 were
  used in CUUATS model to convert employment data
  for each TAZ into number of trips

33
CUUATS Model Trip Generation

• The number of trips generated from each external
  station is based on the ADT data and the trip
  generation rates estimated from the
  Origin-Destination survey conducted by CUUATS in
  2003
• Finally, the number of trips attracted is
  balanced to the number of trips produced
• The final product of this step is an estimation
  of the number of person trips for each zone

34
Trip Generation Process
35
CUUATS Model Trip Generation
HBWork 0.166.1(X1) 0.149.8(X2)
0.1811.4(X3) 0.1412.7(X4) HBSchool
0.156.1(X1) 0.049.8(X2) 0.0311.4(X3)
0.0412.7(X4) HBShopping 0.086.1(X1)
0.149.8(X2) 0.1311.4(X3) 0.1212.7(X4) HBOth
er 0.276.1(X1) 0.359.8(X2)
0.3111.4(X3) 0.4112.7(X4) NHB
0.346.1(X1) 0.339.8(X2) 0.3511.4(X3)
0.2912.7(X4)   X1 Total number of 1-person
households X2 Total number of 2-person
households X3 Total number of 3-person
households X4 Total number of 4person
households
36
CUUATS Model Trip Generation
Total Daily Trips by Purpose in 2005
37
CUUATS Model Trip Distribution

• Of the trips estimated in step 1, estimate the
  number of trips with origin and destination in
  each zone
• Estimation of the number of trips going from one
  TAZ to another TAZ

38
CUUATS Model Trip Distribution

• Friction factors are estimates of the reduction
  in trips with the increase in trip length
• Friction factors in the CUUATS model were
  obtained based on the travel time between zones
  and the number of trips between zones obtained
  from the Origin-Destination survey
• A trip matrix or zone to zone person trip table
  is generated by this step

39
Trip Distribution Process
40
Trip Exchange Between Major Areas
41
CUUATS Model Mode Choice

• Estimate the mode of travel (car and bus) for the
  trip origin and destination estimates created in
  step 2
• Person trips between the TAZ pairs are split
  between transit and auto
• Mode split is based on the ratio of highway
  travel time and the transit travel time (utility
  coefficient) and the transit percentage
  (Diversion Curves)
• An existing overall transit percentage of 6 is
  used in CUUATS Model

42
Mode Choice Curve
43
Mode Choice Process
44
Transit Trip Distribution
45
CUUATS Model Trip Assignment

• The person trips (auto drivers and auto
  passengers) are converted to vehicle trips using
  an auto-occupancy rate of 1.4 persons per vehicle
• The vehicle trips between each zone pair is then
  assigned on the highway network
• The transit trips are assigned to the transit
  network

46
CUUATS Model Trip Assignment

• Equilibrium Assignment
• Wardrop equilibrium principle
• For each origin-destination pair of zones, all
  used routes have equal travel times, and no
  unused route has a lower travel time.
• Iteration Process
• Step 1 Find shortest travel time path between
  TAZ pairs
• Step 2 Assign some percentage of trips on the
  shortest travel time path

47
CUUATS Model Trip Assignment

• Step 3 Compute new travel times based on the
  initial assignment and BPR (Bureau of Public
  Roads) curves
• Step 4 Check for convergence, if satisfied end
  assignment, else go to step 1
• The final product of this step is estimated
  traffic volume on every roadway segment of the
  network
• BPR curve Congested Speed (Free-Flow
  Speed)/(10.15volume/capacity4)

48
Trip Assignment Process
49
Model Validation and Calibration

• Model outputs have to be validated and parameters
  calibrated
• Model Validation and Reasonableness Checking
  Manual, FHWA-EP-01-023
• Model outputs checked for reasonableness and with
  real world data
• Real world data obtained from census, traffic
  counts, NPTS and other data sources
• The best data sources are the ones collected at
  the regional level
• Every step in the modeling process contains
  sources of error

50
Model Validation and Calibration
51
Model Outputs

• Congestion maps, volume bandwidth maps, trip
  purpose maps, trip distribution maps, travel
  route maps
• Vehicle miles traveled 2,293,050 miles
• Vehicle hours traveled 90,400 hours
• Congested Speed 14.99 mph

52
Model Outputs
Congestion
53
Model Outputs
Volume Bandwidth
54
Model Outputs
Trip Purpose
55
Model Outputs
Trip Distribution
56
Model Outputs
Travel Route
57
A look into the near future for CUUATS model

• Junction (intersection) modeling
• account for signal delays instead of turn
  penalties
• include intersection lane configurations and
  signal timings into the model
• Time of day assignment
• obtain trip percentages by trip purpose for
  different times of the day
• split trip distribution table for various times
  of day based on these trip percentages
• Build micro simulation models
• input travel demand model outputs to a micro
  simulation model
• do micro analysis for smaller areas
• use simulation as a public participation tool

58
A look into the near future for CUUATS Model