Scheduling of the Shuttle Freight Train Services for Dry Ports Using Multimethod Simulation–Optimization Approach

Abstract

This paper introduces a simulation–optimization method for addressing scheduling problems for shuttle freight trains (SFTs) in a shared railway corridor between a seaport and dry port. We use dispatching delays for scheduling the SFT trips so as to not disturb the existing scheduled regular train (SRT) paths. The method employs a multi-method microscopic simulation model and an optimization framework. A swarm-based optimization algorithm is used for finding the best dispatching delays to preserve SRT paths. The method is demonstrated for a railway corridor between the Alsancak seaport and a close-distance dry port. The railway corridor is modeled using a simulation model considering single and double railway tracks, stations, and schedules. By running the simulation–optimization, the SFT freight transport capacity and the quality of the SFT and SRT operations were compared using key performance indicators (number of completed trips and station stops, average trip delay, and average station delay) addressing the throughput and punctuality after the application of dispatching delays. The results show that, by preserving the existing SRT paths, freight transport capacity decreased by 11.1% (from 18 to 16 completed SFT trips) and 13.8% (from 36 to 31 completed SFT trips) for single and couple SFT scenarios, respectively. The methodology also decreased the average SFT station delays by 45.2% and 45.6% for the single and couple SFT scenarios comparing with the unoptimized SFT trips. However, the number of SFT station stops increased by 12.5% and 57.1% for the single and couple SFT scenarios for prioritizing the SRTs. Also after the optimization, the average SFT trip delays decreased by 30.7% and 0.58% for the single and couple SFT scenarios. This study successfully demonstrates that the proposed method can be used for scheduling the SFT trips inside a congested railway corridor and can be implemented as a capacity assessment tool for cyclic SFT service using a series of key performance indicators addressing throughput and punctuality.