Wang, Xiang (Beijing Jiaotong University), Luo, Zhengwei (Beijing Jiaotong University), Ma, Qianying (Beijing Jiaotong University), Zhao, Zixu (Beijing JiaoTong University), Lyu, Jidong (Beijing Jiaotong University), Chai, Ming (Beijing Jiaotong University), Li, Kaicheng (Beijing Jiaotong University)

 A Monte Carlo Tree Search Approach for Operating Profile Planning of Freight Trains on the Long Steep Downward Slope Sections

Scheduled for presentation during the Video Session "On-Demand Video Presentations" (VP-VP), Saturday, November 22, 2025, 08:00−18:00, On-Demand Platform

2025 IEEE 28th International Conference on Intelligent Transportation Systems (ITSC), November 18-21, 2025, Gold Coast, Australia

This information is tentative and subject to change. Compiled on April 2, 2026

Abstract

Train speed profile planning is important to ensure safe and efficient operation of freight train premise and foundation. Due to long station spacing of freight railway lines, existence of large gradient long steep downward slope sections, and feature of electric-pneumatic joint cyclic braking, traditional speed profile planning method for small gradient long steep downward slope sections is hard to meet the requirements of safety and other multi-objective constraints in high gradient downhill sections. In this paper, a Monte Carlo Tree Search(MCTS) based freight train cyclic braking profile planning method is proposed to solve the problem of safe and efficient operation on long steep downhill sections. Firstly, the cyclic braking state transfer model of freight train is built up based on Markov decision-making process with optimization objectives of coupler safety, pneumatic braking distance and electric braking time in combination with constraints of speed limit and braking for freight train running on long steep downhill sections. Secondly, train state equations such as braking distance and coupler force are established by Zhai method with multi-mass point model of freight train, and the cyclic braking profile solution algorithm based on MCTS is designed. The paper takes HXD1 locomotive and C70 gondola car as control objects, binds data of typical long steep downhill sections with maximum gradient of 30‰ as basis for simulation analysis. Compared with DQN method, the maximum coupler force of MCTS algorithm is 6.36% lower and the average speed is increased by 0.34km/h, which is able to reduce longitudinal impulsive force of freight train and improve running efficiency.