Development, Evaluation, and Selection of Advanced Transit Signal Priority Concept Directions

This article presents a process to define the framework for an advanced Transit Signal Priority (TSP) algorithm. For this study, traffic and transit agencies from a broad range of municipalities in Ontario, Canada, provided their views and expertise on various TSP-related issues, including practical needs, design implementation, performance measures, and challenges in developing effective TSP control systems. Based on their inputs as well as the objectives of the project, a set of TSP control concept directions was developed that are characterized with different methodologies and technologies. A listing of selection criteria was also established to evaluate the proposed TSP concept directions. Using these criteria, a ranking and evaluation process was undertaken to select one final TSP control concept that is of interest to potential users of advanced TSP systems. The work described in this article provides a successful example of a process to build consensus among stakeholders for advancing TSP developments. Introduction In September 2001, Transport Canada announced a commitment and solicited proposals for the development of products and services that will accelerate the Journal of Public Transportation, Vol. 9, No. 5, 2006 98 growth of ITS knowledge and skills, and promote the uptake and commercialization of ITS technology in Canada. A joint team from the University of Toronto, LEA Consulting Ltd., and Fortran Traffic Systems was awarded a contract for the development of an advanced Transit Signal Priority (TSP) algorithm, which has the potential to be deployed in the field. The objectives of this project can be summarized as follows: • Develop and evaluate a unique, innovative TSP algorithm, that has the potential to be deployed in the field. • Facilitate the exchange of knowledge and ideas between the academic research community and the industrial sector during the algorithm development process. • Provide a means to improve mobility and transportation efficiency. • Increase operational and regulatory efficiencies for system users and public agencies. • Encourage the development of products and services that will accelerate the growth of ITS knowledge and skills, and promote the uptake and commercialization of ITS technology. This article presents a process to develop the framework of the advanced TSP control algorithm as well as the rationale for the selected algorithm approach, rather than describing the developed algorithm itself. Details of the final products of this research can be found elsewhere (Lee et al. 2005, 2006). This article presents various types of TSP control concepts that are characterized with different methodological and technological components. The design of these multilevel TSP concept directions was based on progressive levels of control concepts for the provision of sophisticated TSP control. This research also provides various TSP-related issues from the perspectives of traffic and transit agencies. Professionals from a broad range of municipalities in Ontario, Canada, representing traffic and transit departments, offered their views and expertise on various TSP-related issues, including practical needs, design implementation, performance measures, and the challenges of developing advanced TSP systems. The work described here provides an example of a process to build consensus among stakeholders for advancing TSP developments. Historical background of TSP is briefly presented in the next section. The cornerstone of most proposed TSP control concepts involves an accurate transit travel time prediction method to support the achievement of more efficient and effecAdvanced Transit Signal Priority Concept Directions 99 tive signal priority for both transit and traffic. The state-of-the-art in the transit travel time prediction methods is provided next. Following that is a detailed description of the proposed set of TSP control concepts, the used evaluation criteria, and the selection process.