May 22, 2019

Self-driving truck platoons are delivering goods, auto manufacturers and ride-share providers are racing to be the first to distribute autonomous vehicles to customers, and connected and autonomous (CAV) pilot facilities and vehicle-to-infrastructure pilots are launching across the country.

CAVs are already transforming our transportation systems, and they will bring even more capacity impacts in the future. But how will agencies respond to, or rather plan for a connected and autonomous future?

How will CAVs impact capacity? Will shorter headways increase capacities? Or will those potential gains be offset by safety margins at intersections and interchanges? Will the capacity change be proportional for both freeways and arterial streets?

A future with CAVs is still shrouded with questions, but agencies need information now to make investment and policy decisions to plan for future needs.

We’re working on finding answers through a pooled fund study, measuring how CAVs will impact capacity at different levels of market penetration and system aggressiveness. Agencies will be able to use this information in their long-range planning to estimate the range of scenarios they might see in the future.

About the Pooled Fund Study

Long-range transportation plans, municipal transportation plans, regional and system plans, corridor studies, and even traffic impact analyses all rely on multi-year forecasts of travel demands and roadway capacities. Assumptions about future roadway capacities also impact transportation investments and project prioritization.

Written and regularly updated based on years of data collection and millions of dollars of funded research, the Highway Capacity Manual is the national guide for these plans and studies, reflecting the latest in capacity and quality of service analysis across a full spectrum of facility types.

Today, however, agency planners are asking questions that go beyond what we can learn from looking at past data. The data from CAVs operating in real-world, mixed traffic is extremely limited, so we can’t use observed data to predict future impacts yet.

Putting CAVs into future scenarios means we have to use microscopic traffic simulations to arrive at reasonable capacity predictions.

Simulations need to be calibrated with national best practice on CAV algorithms, and need to reflect realistic transportation and bottleneck conditions to help inform agency practices. If those simulations only take place on idealized “pipe” highways or in best-case-scenario conditions, they will not help us realistically prepare.

Oregon DOT spearheaded a pooled fund study to make these simulations possible. Through funding from DOTs in Arkansas, Connecticut, Florida, Indiana, Maryland, North Carolina, Oregon, Texas, Utah, and Washington State, the three of us – Abby Morgan, Burak Cesme, and Bastian Schroeder – along with Jiaqi Ma of the University of Cincinnati are leading the charge to simulate challenging and realistic conditions that will face CAVs in the future. We’re employing the same tools we’re using to evaluate USDOT connected vehicle pilot projects in Tampa, NYC, Wyoming, and Columbus. Our simulations are exploring and testing how CAVs interact with non-CAVs so we can provide guidance to agencies about the sensitivity of key parameters (allowable headways, acceleration, deceleration, etc.) on the final capacities, resulting in a collection of recommended approaches, best practices, and known limitations.

We’re excited about the benefits to agencies from this pooled fund study:

  • It will provide the bookends for planning-level capacities for CAVs based on state-of-the-practice algorithms;
  • It will provide guidance for planning for a connected and autonomous future across a range of facility types, including freeways and surface streets;
  • It will promote consistency among agencies in a topic full of unknowns;
  • It will put the HCM at the forefront of CAV research in terms of operational readiness of CAV applications and technologies;
  • Ultimately, it will inform agencies about how CAVs will affect capacity at different levels of market penetration and different levels of system aggressiveness, leading to better informed investment decisions in area freeways, new roundabouts and signals, and safety projects.

There are two upcoming sessions in which you can learn more about this research and what it means for your long-range planning initiatives.

TRB Highway Capacity Committee Mid-Year Meeting in San Diego, CA

In June, we’re updating the pooled fund study panel and the TRB Committee on Highway Capacity and Quality of Service with an initial set of results. The initial results will focus on capacity estimates for freeways and uninterrupted flow facilities, include merge areas and weaving segments. Our hope is that the committee will seek to adopt these initial results, and formally provide a recommendation to agencies for CAV freeway capacities at this mid-point to our project. We will also seek concurrence on our next phase of research, where we will focus on intersections and surface streets.

Automated Vehicles Symposium in Orlando, FL

In July, we’re presenting at the Automated Vehicles Symposium, which will be held in Orlando. Spearheaded by Erik Ruehr of VRPA Technologies and along with our teaming partner Jiaqi Ma, we’ll be part of an international panel to summarize the research to-date and explain how CAV capacity impacts will be quantified for use in long-range transportation planning.

This information will help agencies estimate the range of CAV impacts they might see in the future. Our aim is that our audience leaves better prepared to discuss the future of CAV roadway capacity effects with decision makers, clients, and the public.

You can learn more about the Automated Vehicles Symposium and register here.

Equipping Planners to Find Answers

Important questions still face agency planners. Should roads be widened? Do we need additional lanes in the future, or can we re-capture space for multimodal users? How will following distance be regulated? What have we not considered as we prepare for a more connected and automated future? The range of scenarios and realistic conditions modeled in this research project will move DOTs closer to answers and equip planners to more reasonably predict capacity demands. Feel free to reach out to any of us to further discuss this project and its implications.

Kittelson Project Team: