Intelligent Transportation Systems

Intelligent Transportation Systems
Efficiency, Equity and Fairness

At the Core of the Issues

Heterogeneity
- The relevant population (users and others) is comprised of dissimilar people
Differential Impacts
- As a result, different people/groups may be affected differently

Why Should We Care?

Public Sector Involvement
- The government is/will be a major provider of IVHS services and some services may not be deemed appropriate
Implementability
- Ignoring these issues may result in public opposition
Social Responsibility
- Societies can only succeed if their members worry about these kinds of issues

Framework

Efficient Policies
- Social welfare is increased (a utilitarian concept)
- Kaldor-Hicks criterion
Equitable Policies
- Nobody is worse off (an egalitarian concept)
- Pareto criterion
Fair/Superfair Policies
- People would rather have what they have than what others have
- Envy criterion (Baumol)

Look Good but Inefficient?

Ramp metering
- May result in engine mode shifts that increase emissions (which may have a social cost)
Electronic Toll Collection
- Possible increase in equilibrium travel cost (Braess)
Other Information and Control Systems
- Possible induced demand and associated costs

Efficient but Inequitable?

Congestion Pricing
- Can result in total reductions in travel costs but some groups may gain at the expense of others (even when revenues are redistributed)
Route Guidance
- May reduce travel time on some routes but increase travel time on others

Efficient and Equitable but Unfair?

In-vehicle Signing
- Can result in total reductions in travel costs without anyone's cost increasing but some groups may have it and others may want it

Related Concerns

Should current information, predicted information, guidance, and control be treated differently?
Can we ethically provide information/controls that are consistent with people's desires (e.g., don't route me through a `bad' neighborhood)?
What are the appropriate roles for the public and private sectors (national defense is probably not a good model)?

Conclusions

There's Always More to Learn