Optimum Allocation of Transit Signal Priority Deployment Along a Transit Corridor: A Bilevel Optimization Approach
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The central focus of this thesis is to develop a delay cost optimization model based on the cost of total person delay determining optimal Transit Signal Priority (TSP) configuration along a specified corridor based. The optimal configuration of TSP along a corridor allows for TSP to be implemented when it only provides cost benefits and reduces overall delay. Additionally, if there is an implementation or operation restriction for the number of intersections with TSP enabled then this optimization model allows for the immediate selection of the optimal locations. The TSP configuration model as a bilevel approach with the upper level expressed as a delay cost optimization model is a useful tool for current and future transit planning applications. It uses newly available data and provides thorough recommendations for the optimal configuration of TSP based on selected key performance indicators (KPIs) including threshold values for each intersection. The cost delay optimization model was developed to calculate the cost of delay as related to TSP implementation at each intersection along a corridor using KPIs were selected to represent all users of the corridor for peak hour flow, specifically current bus passengers, downstream waiting passengers, and other commuters driving private vehicles. The bilevel approach can be split into the upper-level delay cost optimization model and the lower-level VISSIM model. The upper-level delay cost optimization model may be used to evaluate each intersection along a corridor as to the efficacy of implementing total person delay TSP and is developed from selected KPIs. The cost delay optimization model to minimize delay of all users along a corridor was developed based on the selected KPIs as the input variables. The output of the delay cost optimization model is the configuration of TSP along the corridor which is then input into the VISSIM model during the bilevel approach.