Auxiliary Road Design and Optimization for Railway Construction in Mountainous Areas

Abstract

Designing and optimizing auxiliary roads to support railway construction in complex mountainous areas is challenging. This paper presents a two-stage optimization model for auxiliary road design. The first stage designs the network layout of auxiliary roads based on the minimum spanning tree method to reduce link costs among various railway auxiliary construction projects. The second stage involves designing and optimizing the alignment of the auxiliary roads based on the deep reinforcement learning approach to minimize construction costs. In addition, the different logistical relationships among various railway auxiliary construction projects are considered to optimize the total turnover volume of the auxiliary road. Finally, a real-world case study of auxiliary road design for a railway construction project in mountainous areas was conducted to verify the proposed method.