Source: NHI course on Travel Demand Forecasting (152054A)

1
Session 11 Model Calibration, Validation, and
Reasonableness Checks
Source NHI course on Travel Demand Forecasting
(152054A)
2
Terminology

• Model Calibration
• Model Validation
• Reasonableness checks
• Sensitivity checks
• Special generators
• Screen lines (some modelers do not think this is
  important)

Is the model sensitive to policy options?
3
Key Concepts

• Not enough attention on model evaluation and
  reasonableness checks
• Checks performed after each step
• reduces error propagation

4
Information Requirements for Validation and
Reasonableness

• Demographics and employment
• Highway and transit networks
• Base year survey
• Base year traffic counts

5
Sources of Error

• Network coding
• Sampling
• Computation
• Specification
• Data transfer
• Data aggregation

6
Inputs and Outputs
Number and location of households and employment (demand) Socioeconomic Data
Complete? Level of Detail?
Transportation system (supply) Network Data
Reasonable? Methodology? Source?
TDF Model Specification Model validation and calibration
Travel survey data
Sensitive? Documentation of calibration? Valid
for base year?
Transportation system performance
Current? Reasonable?
7
Model Calibration and Validation
Model Calibration
Independent Data Sets
Feedback Loop
Model Validation
Model Application
8
Socioeconomic Data Check Reasonableness of

• Source for estimates and forecasts
• Population and household size trends (graph 1950
  to present and check trend)
• Household income trends (graph as far back as
  this goes 1990?)
• Dollar values used in forecast (use constant
  dollars)
• If used, trend of automotive availability (S
  curve?)
• Distribution of employment by type (basic,
  retail, service) over time
• Employees per household and per capita rate of
  increase is decreasing
• Household and employment changes by zone
• Area type (density) changes are the growth
  areas reasonable?

9
Travel Survey Data Reasonableness Checks

• travel survey data
• Types of survey conducted
• Year of survey
• If no survey (borrowed)
• Source of trip rates, lengths, TLFD
• Is area similar
• Geographic area?
• pop/HH/emp characteristics?
• Urban density and trans system?
• Trip chaining convention
• Compare to similar regions and to same
• region in earlier times
• Person trip rates by trip purpose
• Mean trip lengths by trip purpose
• HBW longest? HBO shortest?
• TLFDs by trip purpose

10
Network Data Reasonableness Checks

• Facility types, level of detail
• Speed and capacity look-up table (what LOS used
  for capacity?)
• Significant transportation projects narrative?
• Plot (facility types, lanes,
• speeds, area types)

11
Trip Generation Reasonableness Checks

• Trip production and attraction models
• Form?
• sensitivity?
• Trip purposes used
• Person trip or vehicle trip rates used?
• PA balance (0.9-1.1 ok)
• Special generators (be consistent in future
  model)

12
Trip Generation CalibrationTypical Values

• Person trips per household 8.5 to 10.5
• HBW person trips per household 1.7 to 2.3
• HBO person trips per household 3.5 to 4.8
• NHB person trips per household 1.7 to 2.9
• HBW trips 18 to 27 of all trips
• HBO trips 47 to 54 of all trips
• NHB trips 22 to 31 of all trips

13
Trip Distribution Reasonableness Checks

• Mean trip length (increasing or decreasing?)
• TLFDs

14
1st iteration
Calibrate friction factors
15
Mode Split Reasonableness Checks

• Automobile occupancy factors by trip purpose
• Basis? Constant?
• Form of mode split model?
• Variables included in the utility functions?
• Coefficients logical?
• IIA properties of LOGIT models
• Mode share changes over time
• Mode share comparisons with other cities

16
Mode Split Calibration and Validation
Advanced concepts needed to formulate and
calibrate mode split models

• Experienced planning consultant
• Calibration tasks
• Form of LOGIT model
• Variables included in utility functions
• Calibration of coefficients for utility function
  variables
• Testing for IIA properties
• Analysis of household survey data
• Analysis of on-board transit survey data
• Total highway and transit trips
• Ridership by route

17
Trip Assignment Reasonableness Checks
Compare to see where adding capacity may help

• All-or-nothing assignment
• Equilibrium assignment
• Volume delay equation (BPR)
• Screen line volumes
• Time-of-day assignments
• Local VMT

Equil ibrium
All or Nothing
18
Trip Assignment Calibration and Validation
Assignment calibration performed
last

• Overall VMT or VHT check
• 40 to 60 miles per day per HH in large metro
  areas
• 30 to 40 miles per day per HH in medium metro

19
(No Transcript)
20
Other Factors Impacting Forecasted Travel Demand

• Telecommuting
• Flexible work hours
• HB business
• Aging population
• Internet shopping
• Roadway congestion
• New modes