Line 1: Line 1:
[[Category:Incubator topics]][[Category:Autonomous vehicles]]
+
{{Incubator}}[[Category:Incubator topics]][[Category:Autonomous vehicles]]
  
 
<span style="background:lightgrey;padding:10px;border-left: thick double #0000aa;"> This page is part of the Category [[:Category:Autonomous vehicles|autonomous vehicles]].</span>
 
<span style="background:lightgrey;padding:10px;border-left: thick double #0000aa;"> This page is part of the Category [[:Category:Autonomous vehicles|autonomous vehicles]].</span>

Revision as of 21:23, 8 November 2017

This page is open for editing because it is part of the Incubator. Have something to add? Please register so you can contribute. Have an option you would like to share? Please click on the 'Talk' button to enter the dialogue. The TF Resource Volunteers appreciate your feedback and interest.

This page is part of the Category autonomous vehicles.

Use Case: No More new Roadway Capacity

CAV5s have the potential to dramatically improve the effective roadway capacity of existing infrastructure (TODO: ADD REFERENCE). Policy makers are therefore being asked to make very difficult decisions in the context of long-range planning. Specifically: should their jurisdictions continue to invest in additional roadway capacity? Widening roadways is generally very expensive and requires long lead times and allocating resources to widening roadways means not allocating resources to other purposes, such as operational or maintenance purposes. When combined with the political costs of constituents dealing with congestion, policy makers could benefit from as much information as possible to guide them in making these decisions. It is very complicated to think through the competing impacts of population growth and technological change.

Policy Summary

  • Goal: efficiently allocate scarce resources.
  • Policy: reallocate resources from constructing additional physical roadway capacity (i.e., no new lane-miles) in congested corridors to operational/non-physical-capacity-increasing projects.

Behavioral Considerations

Travel models typically assume that congestion in no-build scenarios will increase proportionately with the population if no new capacity is added. In this policy scenario, the increased capacity comes from CAV5 technology rather than new lane-miles. It is an open question as to whether the technology improvements will outpace population growth, and the behavioral responses to this policy could therefore trend down two paths. If technology adds new capacity faster than population grows, then recurring congestion and peak period travel times will decrease. Coupled with lower values of in-vehicle travel time, the following behaviors are likely to emerge immediately:

  • Longer automobile trips
  • Mode shift towards automobiles and away from transit or non-motorized modes

Over time, longer-term behaviors may emerge:

  • People will choose residential and workplace locations that are located further apart
  • People will purchase CAV5s to maximize their mobility and comfort while traveling long distances.

These changing behaviors could mitigate some of the benefits seen from technological improvements. Conversely, if population grows faster than CAV5 technology the behavioral changes are likely to be inverted. For example, the benefit of owning a personal CAV5 is lower for the short trips that are more likely in congested areas, and therefore shared ownership or hired rides from TNCs may be more common. This policy is not explicitly concerned with behaviors such as household tour coordination or drop-and-hide (off-site) parking, though the feasibility of various daily patterns might change depending on vehicle availability.

Modeling Framework

To represent the immediate behavioral responses to this policy in a travel model, you will need to do at least two things:

  • Modify the highway volume-to-capacity functions to reflect higher capacities at higher volumes.
  • Lower the in-vehicle value of time coefficient for automobile trips
  • Feed back congested travel time skims into the destination choice and mode choice models.

The degree to which these coefficients or functions may shift is unknown, and speaks to the need for scenario testing and risk analysis. Understanding the effects of secondary behaviors may require:

  • Automobile ownership models that distinguish between CAV5 and conventional vehicles
  • Residential and firm location choice models
  • Land use models

Use Case: Occupancy-differentiated Vehicle-miles Traveled Tax

A potential drawback to the wide-scale deployment of CAV5s is that they may harm the environment and/or the public realm by avoiding storage fees and/or optimizing the experience of their owner/customer by engaging in passenger-free travel. For example, an automated vehicle may circle the block to avoid paying a parking/storage fee while their owner/customer shops for groceries. Or, a CAV5 may drop its passenger off at work and then move to a remote location far away to be stored for the day, returning later to retrieve the passenger.

In addition to the concerns about zero passenger travel, urban planners have long been frustrated with the inefficiency of single-occupant vehicle travel and have long advocated for schemes such as high-occupancy vehicle lanes that make more efficient use of existing infrastructure.

An occupancy-differentiated vehicle-miles traveled tax has the potential to mitigate the impacts of both of the above behaviors (as well as raise revenue), by increasing the relative price of automobile travel with one or fewer passengers. Travel models can likely usefully contribute to this conversation across a wide range of automated vehicle adoption/penetration rates.


Policy Summary

  • Goal: efficiently utilize existing infrastructure
  • Policy: charge vehicles a usage tax based on distance traveled and number of occupants, with a steep fee charged to zero-occupant vehicle movements.

Behavioral Considerations

The immediate behavioral responses to this policy would be for households to consider shortening their trips and/or increase their vehicle occupancy. This will lead to a series of primary behavioral changes:

  • Greater intra-household tour coordination
  • More shared rides with people outside households
  • Less last-mile access to mass transit stations
  • Fewer dead-head trips, with reduced off-site parking
  • Discretionary activities would move closer

The secondary impacts are important as well:

  • Increased substitution of travel with teleworking and online shopping
  • Fewer people will own personal CAV5s, and will instead hire rides from TNCs
  • TNCs will have higher fares, but will optimize their services to minimize their tax burden
  • Land use patterns

The advantage of short trips will marginally change residential and firm location decisions, changing land use patterns. Alternatively, people may substitute travel with teleworking and online retail. The tax may dissuade some households from purchasing an AV at all. TNCs operating fleets of for-hire AV’s will therefore have a larger potential passenger base, though they will pass through the tax to their customers in higher fares; they will also face strong incentives to minimize their tax burden through network optimization.

Another consideration is people are relatively insensitive to hidden costs. If the tax is assessed in the form of a periodic bill instead of an up-front fare, then the behavioral effects of this tax on individuals will be minimal. TNC’s, however, would still face strong incentives to optimize.

Modeling Framework

Modeling the behavioral changes of this policy will require elements not frequently found in extant travel models. To model the primary behavioral impacts, the mode will need:

  • Mode choice including hiring or sharing CAV5s, either for main the trip or for transit access.
  • Separate travel time and travel cost values that vary by occupancy
  • Coordinated household activity patterns considering mode and destination accessibilities

The secondary behaviors will need a different set of modeling tools:

  • Vehicle allocation model to consider empty trips and inform household coordination
  • Vehicle ownership model considering differential CAV5 accessibilities
  • Residential and firm location choice
  • Land use model


References

Content Charrette: Autonomous Vehicles