Dedicating Lanes for Priority or Exclusive Use by Connected and Automated Vehicles (2018)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

135 This report documents the research conducted under Task 8 of NCHRP Project 20-102, which aims to assess the impacts CAVs on state and local transportation agencies. The project aimed at developing guidance on identifying and describing conditions amenable to dedicating lanes for CAV users. To assist in identifying the conditions and parameters that can make or break a case for dedicating lanes to CAV users, the project team conducted simulations to analyze two CAV applications, CACC and DSH, based on two case study sites—I-66 in Northern Virginia and US-101 in San Mateo County, California. The modeling and simulation activity helped the project team (1) identify parameters that are sensitive to dedicating lanes to CAV users and (2) identify expected impacts under various conditions of lane dedication, market penetration, demand conditions, combined deployment of applications, and so forth using virtual computer- based models. To conduct this research, the team started with a comprehensive literature review to identify: • The types of stakeholders benefited (or disbenefited) when dedicating lanes to CAVs, • Factors influencing these benefits, and • The potential measures of the performance. Specifically, the research team identified three types of stakeholders: (1) DL users, (2) GPL users, and (3) facility owners and operators. The following factors also were identified as influencing the benefits and disbenefits to these stakeholders: • CAV market penetration, representing the percentage of vehicles in the traffic mix with CAV capabilities; • Roadway geometry, including access/egress features, lane attributes, number of lanes, and so forth; • Enforcement intensity, which restricts unallowable categories of users to enter the DLs; • Toll collection attributes, such as whether non-CAV vehicles can use the DLs, for a fee; • Operation hours, such as dynamic operations or peak-hour operations; • CAV technology, which represents the type of applications allowed on vehicles using these lanes; and • Functional types, which dictate the type of vehicles allowed on the DLs. The review also documented performance measures specific to mobility, energy and environment, safety, and societal benefits that might be achieved by users and non-users of such DLs. This step was followed by research on various CAV applications. Several CAV applications exist in the research industry today, and assessing all of them was beyond the scope of this study. As a result, the study team down-selected two CAV applications, CACC and DSH, along with C H A P T E R 1 0 Conclusions

136 Dedicating Lanes for Priority or Exclusive Use by Connected and Automated Vehicles suitable modeling techniques for use in this project. The applications were selected based on three factors: • Suitability to DLs, • Suitability to the CAV/AV environment, and • Adaptability to simulation models. Specific modeling frameworks that could represent each of these applications also were selected, based on several criteria. The project team utilized modeling and simulation-based analysis to evaluate potential benefits and parameter sensitivity of CAV applications on overall traffic efficiency and safety. Having identified multiple modeling-based test sites to which the team had access, the research team narrowed the field of candidates to nine case study sites that were then evaluated and ranked based on the following characteristics: • Overall site characteristics (e.g., geography, operational conditions, modes, presence of managed lanes, and so forth); • Managed lane characteristics (e.g., geometry, allowed users, operating rules, access point configurations, and so forth); and • The feasibility of modeling CAV applications for the site. As a result of this evaluation, the team selected two case study sites: the I-66 corridor in Northern Virginia and US-101 corridor in San Mateo County, California. The project team then conducted simulation-based analysis to evaluate the impacts of CAV DL applications under different sensitivity parameters. Primarily, the simulations studied the impacts of: • Priority lane use (where CAVs were permitted on HOV lanes), versus exclusive lane use (where CAVs had exclusive access to DLs); • Rates of market penetration of CAV/AV users, assessed under a DL case and a non-DL case; • A combination of CAV applications; • Varying demand and changing operational conditions on the CAV DL benefits; • Access restrictions to the DL under exclusive CAV lane situations; and • Hypothetical scenarios, such as incident-related lane closures or moving bottlenecks. In addition to the more operational analysis based on modeling and simulation, the project team also conducted a literature review to identify the laws and regulations regarding dedicating lanes to a specific category of users. It was found that, historically, lanes have been dedicated to HOVs, motorcycles and bicycles, buses, AFVs, and trucks. Consequently, the project team developed specific guidance for agencies on operational characteristics and impacts, as well as regulatory and policy guidance for states and local agencies on conditions amenable to dedicating lanes to CAVs. This guidance is summarized as follows: • For CACC DLs, it is advisable to have shared DLs with HOVs at lower levels of market penetration (10%), exclusive DLs at medium levels of market penetration (20% to 45%), and no DLs for higher levels of market penetration (greater than 50%). • For lower levels of market penetration of CACC: – Under shared DL conditions, slight mobility benefits may result for both DL users and GPL users; and – Under exclusive DL conditions, significant mobility and energy/environmental benefits to DL users may result, at the expense of GPL users.

Conclusions 137 • For higher levels of market penetration of CACC: – Under exclusive DL conditions, moderate to significant mobility and energy/environmental benefits to DL users may result, and slight to moderate benefits to GPL users. • For lower levels of market penetration of DSH: – Under shared DL conditions, slight safety benefits may result for both DL and GPL users; and – Under exclusive DL conditions, slight safety benefits may result for DL users, and sig- nificant safety benefits may result for GPL users at the expense of GPL users’ mobility performance. • For higher market penetration of DSH: – Under exclusive DL conditions, moderate to significant improvement in safety and slight improvement in energy/environmental performance may occur for DL users; GPL users would remain unaffected. • Combining DSH-enabled vehicles with CACC-enabled vehicles will improve safety in addition to mobility and energy/environmental performance. • CACC DLs can provide mobility benefits (in terms of throughput improvement), particularly when the corridor is subject to peak or higher than peak demand. • Mobility benefits are more significant when there is continuous access to the DLs because even vehicles taking shorter trips can utilize the DLs. • Speed differentials between DLs and GPLs increase with restricted access to DLs. DLs with exclusive CAV access will have a much higher travel speed than the GPLs. • Average travel speeds on GPLs reduce significantly when access to DLs is restricted. This reduction in speed occurs because the demand on GPLs will be higher with a restricted access DL compared to a continuous access DL. Depending on the placement and convenience of access points, even eligible vehicles may not use the DLs for shorter trips. • Lane friction (the speed differential between the DL and adjacent GPLs) informs guidance on when barrier separated lanes are warranted: – High MPRs of CAVs with shared DLs with HOV demonstrated the lowest lane friction; this scenario does not warrant lane separation barriers or restricted lane access. – Low MPRs of CAVs with shared DLs with HOV also demonstrated relatively low lane friction. – High MPRs of CAVs with exclusive DLs showed medium lane friction, whereby the average speeds of the DL and the adjacent GPL differed by 10 to 15 mph. According to Best Prac- tices: Separation Devices Between Toll Lanes and Free Lanes (Hlavacek et al. 2007), this level of lane friction does not warrant physical separators, but rather buffer-separated double solid lines. – Low MPRs of CAVs with exclusive DLs showed the highest lane friction, on the order of 30 mph. This level of lane friction warrants physical separators for both enforcement and safety purposes. This guidance must be used in conjunction with the type of analysis that went into developing these statements. Scenarios may exist outside the scope of the analysis performed during this project that could potentially enhance or change this guidance.