The NCDOT Central Office System Timing (COST) Section is charged with developing, evaluating, and maintaining closed-loop signal system timing plans across the state. Currently, signal plan evaluation is a manpower intensive task that involves field observation and travel time runs. This approach to plan evaluation is costly and inefficient. Even when extensive field study is undertaken, the data represent a relatively small observational sample of traffic control and traffic flow conditions. Therefore there is clear motivation for developing performance evaluation methods that are cost effective and comprehensive.
One promising avenue of investigation is exploitation of the signal system detector and signal timing data for generation of performance metrics. Although these signal systems include extensive vehicle detection for use in dynamic system control logic, the data have been used exclusively as inputs to dynamic control logic. A new thread of research activity has begun to explore used of signal control detection and control outputs for estimating system performance. The SMART-SIGNAL system prototype developed by researchers at the University of Minnesota for the Minnesota Department of Transportation is one such effort. As a first step in the proposed project, the project team will thoroughly investigate the research, practice, and technical report archives in the U.S. and internationally to develop a detailed picture of the current state of practice and research.
Beginning with the foundation of relevant results and methodologies from previous research and development activities, the project team will devise a data collection plan to support the rigorous development of closed-loop system performance models. The accuracy and precision of these models will be assessed relative to field observed performance measurements. Recommendations will be developed for the implementation of the best performing models. When implemented, the products of the proposed research will significantly enhance the efficiency and effectiveness of COST section timing plan evaluation activities. Furthermore, the project team will provide a comparative evaluation of system travel time estimates from the proposed models with respect to INRIX arterial surveillance data. Along these lines, it is anticipated that the performance models developed will also support ongoing NCDOT efforts to deploy a comprehensive statewide mobility and reliability monitoring system and to provide timely, accurate, and useful traveler information.
The overarching goal of this research project was to investigate the potential for the NCDOT Central Office Signal Timing (COST) Section to monitor and assess the quality of field deployed closed-loop signal system plans using the data inherent in the systems. The project is complete and has produced recommendations and deliverables that should enhance the COST Section’s ability to achieve its mission of developing and maintaining quality signal coordination plans across the state of North Carolina.
Key findings and conclusions include the identification of a series of monitoring and analysis elements that can be implemented using the OASIS software detector and split monitor logs. In order to analyze dynamic, cycle-by-cycle bandwidth, a tool entitled the Dynamic Bandwidth Analysis Tool (DBAT) was developed and provided as a project deliverable. The DBAT tool was enhanced to provide exhaustive search optimization that identifies offset combinations that maximize dynamic bandwidth for a given set of split monitor log cycle-by-cycle signal indications. DBAT optimization is feasible for systems up to four or five intersections in length. An LP formulation was developed and tested that overcomes the system size limitation. The LP formulation can serve as the basis for future development of an implementable dynamic bandwidth optimization tool.
