An Evolutionary Model for Optimizing Rail Transit Station Locations

Abstract

Optimizing rail transit station locations is a very complex engineering problem. The requirements and constraints that should be considered in locating rail transit stations are complex and interrelated. Although several optimization models have been developed to solve the rail transit station location problem, most of them focus on a single objective and only yield a suboptimal solution to the problem. Multiple-objective models for optimizing rail transit station locations are rare in the literature and their capabilities are very limited. This paper, addresses the limitations in the existing models by developing an evolutionary model, taking into account various local conditions and the multiple planning requirements that arise from passenger, operator and the community to optimize station locations. The model uses an evolutionary solution algorithm (a search algorithm that imitates the natural evolution process) based on genetic algorithm (GA) integrated with geographic information system (GIS) tools to perform the optimal search. The model was applied to an artificial case study and the results demonstrate that the model can optimally locate stations that satisfied the identified planning requirements and constraints.

Keywords

• Rail Transit Station
• Optimization
• Genetic algorithm
• GIS