Incorporating Reliability Performance Measures into Operations and Planning Modeling Tools (2014)

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.

1The broader goals of the Reliability focus area within the second Strategic Highway Research Program (SHRP 2) are to address unexpected traffic congestion and improve travel time reli- ability. To this end, SHRP 2 research projects have brought forward numerous technical measures and policies for further consideration and development. In parallel with these proj- ects, the L04 project, Incorporating Reliability Performance Measures into Operations and Planning Modeling Tools, is aimed at improving planning and operations models to create suitable tools for the evaluation of projects and policies that are expected to improve reliability. The L04 project addressed the need for a comprehensive framework and conceptually coher- ent set of methodologies to (1) better characterize reliability, and the manner in which the vari- ous sources of variability operate individually and in interaction with each other in determining overall reliability performance of a network; (2) assess the impact of reliability on users and the system; and (3) determine the effectiveness and value of proposed counter measures. In doing so, this project has closed an important gap in the underlying conceptual foundations of travel modeling and traffic simulation, and provided practical means of generating realistic reliability measures using network simulation models in a variety of application contexts. A principal accomplishment of the project is a unifying framework for reliability analysis using essentially any particle-based microsimulation or mesosimulation model that produces vehicle travel trajectories. The framework developed in this study is built on a taxonomy that recognizes demand- versus supply-side, exogenous versus endogenous, and systematic versus random variability. The framework features three components: 1. A Scenario Manager, which captures exogenous sources of unreliability, such as special events, adverse weather, work zones, and travel demand variation; 2. Reliability-integrated simulation models that model sources of unreliability endogenously, including user heterogeneity, flow breakdown, and collisions; and 3. A vehicle Trajectory Processor, which extracts reliability information from the simulation output, namely, vehicle trajectories. The primary role of the Scenario Manager is to prepare input scenarios for the traffic simula- tion models; these scenarios represent mutually consistent combinations of demand- and supply- side random factors and are intended to capture exogenous sources of variation. Endogenous variation sources are captured in the traffic simulation model, depending on the modeling capability of the selected platform and the intended purpose of the analysis. The framework may be used with any “particle-based” simulation model, namely, microscopic and mesoscopic Executive Summary

2simulation models that produce individual vehicle (or particle) trajectories. These trajectories enable construction of any level of travel time distributions of interest (e.g., networkwide, origin–destination pair, path, and link) and subsequent extraction of any desired reliability met- ric. These tasks are performed by the Trajectory Processor, which produces the scenario-specific travel time distribution from each simulation run and constructs the overall travel time distribu- tion aggregated over multiple scenarios. The Scenario Manager allows generation of hypothetical scenarios for analysis and design purposes, while the scenario management functionality allows retrieval of historically occurring scenarios or of scenarios previously constructed as part of a planning exercise (e.g., in conjunc- tion with emergency preparedness planning). Furthermore, the Scenario Manager/Generator facilitates direct execution of the simulation model for a particular scenario by creating the necessary inputs that reflect the scenario assumptions. When exercised in the latter manner (i.e., in random generation mode), the Scenario Manager becomes the primary platform for conduct- ing reliability analyses, as experiments are conducted to replicate certain field conditions, under both actual and hypothetical (proposed) network and control scenarios. In particular, the Sce- nario Generator enables execution of experimental designs that entail simulation over multiple days, thus reflecting daily fluctuations in demand, both systematic and random. Two main approaches may be used to assess the travel time reliability for a given project assessment or application: (1) the Monte Carlo approach and (2) the mix-and-match (or user-defined) approach. In addition to the framework and tool itself, the project also developed the method- ological aspects of conducting scenario-based reliability analysis, including mechanisms for gen- erating scenarios recognizing logical, temporal, and statistical interdependencies among different sources of variability modeled through the scenario approach. The vehicle Trajectory Processor produces and helps visualize reliability performance measures (travel time distributions and indicators) from observed or simulated trajectories. The travel time distributions and associated indicators are derived from individual vehicle trajectories, defined as sequences of geographic positions (nodes) and associated passage times. These trajectories are obtained as output from particle-based microscopic or mesoscopic simulation models. Such trajectories may alternatively be obtained directly through measurement [e.g., probe vehicles equipped with global positioning systems (GPSs)], thus enabling validation of travel time reliability metrics generated on the basis of output from simulation tools. Prototypes of a Scenario Manager and a Trajectory Processor have been developed as project- specific deliverables of this research. The tools are conceptually generic and (simulation) software- neutral. The prototypes were demonstrated for the microsimulation modeling platform Aimsun and the mesosimulation dynamic traffic assignment (DTA) platform DYNASMART-P, both of which are representative of other available options in their respective categories to enable rapid cross-platform adaptation. The prototypes and the overall reliability-analysis framework were demonstrated by applying these microsimulation and mesosimulation models to networks extracted from the New York City regional network. Detailed calibration and validation steps were described using available data sources in addition to a specially acquired sample of actual vehicle trajectories based on GPS traces—highlighting and demonstrating the role and potential of such vehicle trajectories in traffic simulation model development and application, especially for reliability-oriented analysis purposes. In addition to the development and application of this general framework, the study made specific contributions in several related areas, namely: (1) development and validation of a robust relationship between the standard deviation of the trip time per unit distance and the mean of the trip time per unit distance, using both simulated and observed trajectories; (2) a detailed proposal of an approach for incorporating reliability considerations into planning

3 models and practices, using different levels of representational detail and associated computa- tional requirements; and (3) initial development of a new approach to microscopic modeling of driver behavior that can capture endogenously more of the sources of variability than currently available models. In summary, this project developed and demonstrated a unified approach with broad applicability to various planning and operations analysis problems, which allows agencies to incorporate reliability as an essential evaluation criterion. The approach as such is indepen- dent of specific analysis software tools so that it can enable and promote wide adoption by agencies and modeling software developers. The project also developed specific software tools intended to serve as prototypes of the key concepts—namely, a Scenario Manager and a Trajectory Processor—and demonstrated them with two commonly used network modeling software platforms.