Title: Understanding the Concept of Latent Demand in Traffic
1Understanding the Concept of Latent Demand in
Traffic
- Prof. Patricia L. Mokhtarian
- Civil Environmental Engineering, UC Davis
- plmokhtarian_at_ucdavis.edu
- www.its.ucdavis.edu/telecom/
- (530) 752-7062
2Outline of this Talk
- What are latent and induced demand, and their
implications? - Empirical approaches to assessing induced demand
- Typical results
- Limitations
- UC Davis study using matched pairs
- More recent work Cervero/Hansen Choo/Mokh.
- Summary
- Concluding thoughts
3What is Latent Demand?
- Often used interchangeably with induced demand,
but the two concepts can be technically
distinguished as follows - Latent demand Pent-up (dormant) demand for
travel, travel that is desired but unrealized
because of constraints - Induced demand Realized demand that is
generated (induced, drawn out) because of
improvements to the transportation system
4Induced Demand
- The increment of new vehicle traffic that would
not have occurred at all without the capacity
improvement. - Clear in theory, but difficult in practice!
- Observed increases in traffic on a
capacity-enhanced network link can arise from a
variety of sources
5When is Traffic Growth Induced Demand?
- Shifts in departure time
- Changes in route or destination (no for vehicle
trips but maybe for VMT) - Shifts from shared modes to drive alone
- New or longer trips to existing locations
- Background demographic growth (WHOA)
- Trips generated by new development attracted to
the improved corridor
6Why do we Care about Induced Demand?
- Need to be able to forecast newly-created travel
(that WNHOA) - Affects the cost-benefit calculation for the
improvement - Affects the assessment of environmental impacts
- Legal/political ramifications
- Sierra Club v. MTC, 1989
- UK abandoned predict and provide policy in mid
90s
7Empirical Approaches
- Case studies
- Cross-sectional disaggregate modeling
- Cross-sectional aggregate modeling
- Time series aggregate modeling
- Cross-sectional/time series aggregate modeling
- Time series link/facility level analysis with
controls
8Case Studies
- Change in traffic on single facility measured
- Results mixed, but have generally found observed
volumes higher than forecasts - May highlight idiosyncratic circumstances
- Often short-term difficult to distinguish
induced demand from shifted demand or background
growth
9Cross-sectional Disaggregate Modeling
- Using 1995 NPTS (travel diary data), analyze
association of VMT with speed - Higher speeds associated with greater VMT
- Speed is a more behaviorally-sound influence on
VMT than capacity - Association doesnt guarantee causality cant
identify long-term impacts
10Cross-sectional Aggregate Modeling
- Models impact of lane-miles on VMT for metro
areas in US - Increase of 1 in lane-mi leads to 0.8 increase
in VMT - Potentially represents long-term equilibrium
- Bi-directional causality impossible to untangle
with single equation, no dynamic element
11Time Series Aggregate Modeling
- Decomposed VMT growth (Milwaukee, 1963-1991) into
sources based on assumed relationships - 6-22 of total VMT growth attributable to new
capacity - Regional focus decomposition approach useful
- Still only one direction of causality permitted
12Cross-sectional/Time Series Aggregate Modeling
- Models VMT as function of lane-mi among other
variables, for multiple areas over time - 1 increase in ln-mi ? 0.2 0.9 increase in VMT
(long-run gt short-run) - Advantages
- Covariates help capture background influences
- If area large enough, demand shifts accounted for
- Temporal precedence can be established
13Cross-sectional/Time Series Aggregate Modeling
(contd)
- Disadvantages
- Not all background influences captured
- Facility/metro-level analyses subject to
confounding with changes in classification and
urban boundary over time - Even temporal precedence doesnt guarantee
causality - Effectiveness of lagged variables depends on
whether planning horizon is longer than the lag
14Time Series Link/Facility Level Analysis with
Controls
- Compares growth in ADT on improved links, to that
on matched set of unimproved links - Study of 18 matched prs in CA (UCD faculty) found
no difference in growth rates - Controls for causes of growth common to improved
and comparison segments - Several disadvantages
15Time Series Link/Facility Level Analysis (contd)
- Disadvantages
- Difficult to find suitable controls
- Doesnt control for spatial shifts from nearby
- Cannot establish a control for an entirely new
link - Another possible reason for difference ADT
v. VMT new capacity may affect trip length more
than frequency
16Recent Work Cervero/Hansen
- Cross-sectional/time series aggregate
- state hwys, 34 CA counties, 1976-97
- Simultaneous equations
- Lane-miles ? VMT
- VMT ? lane-miles
- Both directions of causality significant,
lane-miles ? VMT the stronger direction
17Cervero/Hansen (contd)
- Probably the most rigorous published study to
date - Issues
- Did facility reclassification, metro area effects
confound relationships? - What happened to traffic on lower-classification
facilities? - Are the instrumental variables appropriate?
- Is the high goodness-of-fit spurious?
- Were the lags long enough?
18Recent Work Choo/Mokhtarian
- Time series aggregate (USwide, 1951-2000)
- Comprehensive structural model
19Sociodemo-graphics
Economic Activity
Travel Demand (VMT)
Telecom Demand
Transport.Sys. Infrastructure (lane-mi)
Telecom System Infrastructure
Telecom Costs
Travel Costs
Land Use
20Choo/Mokhtarian (contd)
- Time series aggregate (USwide, 1951-2000)
- Comprehensive structural model
- Corrected for high correlations due to similar
temporal trends - Also found both directions of causality
significant, lane-miles ? VMT the stronger
direction
21Summary
- Its a complex issue!
- Each approach has advantages and disadvantages,
something to offer but not definitive answers - To better understand extent to which answer
depends on method, apply multiple methods to same
region - Nevertheless, the most sophisticated analyses
find evidence for induced demand
22Concluding Thoughts
- Transportation demand will continue to grow
- Thus, cant eliminate all system improvements
just because demand will increase - Should rather weigh the costs (increased fuel
consumption, emissions) against the benefits
(increased mobility, economic gain) - Need to continue to improve our measurement and
modeling of both costs and benefits - And continue efforts to more appropriately price
the provision of service