Solving High-Speed Rail Planning with the Simulated Annealing Algorithm

Abstract

High-speed rail (HSR) networks require large investments, and the performance of the infrastructure is affected by varying local environments, while subject to tight layout restrictions. This paper presents a fully integrated three-dimensional model to optimize the HSR alignment at a planning scale, which sets boundaries for the final project design. The model considers mandatory and desirable specifics for the locations to link and the geometry in both the plan view and the longitudinal profile. It also allows one to define prohibited and restricted land-use areas. A computational tool has been developed that takes into account the problem specifics using a simulated annealing algorithm to optimize the problem solution. The capabilities of the model and the tool are demonstrated with the application to an intentionally simple and synthetic case study, considering construction costs and problem constraints, for which sound results are obtained. Both the model and the tool can be expanded to incorporate additional complexity, establishing the basis for real applications and for further integration of geotechnical and hydrological risk factors that affect the HSR performance.