A Bottom-Up Design Model for Improving Efficiency of Transit System

Abstract

This paper presents a model to design high-performance public transit system, where the optimal solution is directly selected from routes built up on real road structure. The objective function is the travel time for all travelers and the infrastructure cost can also be under consideration. The optimization procedure consists of four components, including a method to simulate downtown travel demand using open-source data on the Internet, a method to narrow down the decision variable space with the help of reasonable restriction and map api, an efficient a lgorithm to compute average travel time using block representation of the urban area and an improved evolutionary algorithm to search the optimal solution. Eventually, this method is applied to design the transit system for Changzhi, China. The optimal solution is compared to the original transit system of Changzhi to test both the effect and reasonability of the algorithm.