Closed-Loop Optimization and Simulation for Rail Freight Yard

EasyChair Preprint 2602

Abstract

Managing a rail freight yard is a very complicated task that is currently performed mostly manually by yard operators. In the optimization literature, this management is typically decomposed in subproblems which are solved separately, considering only a subset of yard resources as constraining. The Optimized Real-time Yard and Network Management (OptiYard) European project aims to the optimization of the comprehensive yard management problem and to the production of a closed-loop optimization and simulation framework to show its applicability in reality. In this paper, we present the OptiYard closed-loop framework. We consider the whole service operated on trains: trains are managed from their arrival at the yard up to their departure. All resources are modelled in detail: not only tracks, but also shunting locomotives and crews. The problem includes constraints on their number, their capacity and their time availability. Train and locomotive movements are modelled microscopically, to detect the occurrence of conflicts (situations in which an unplanned braking is necessary due to traffic) and take into account realistic running times. Finally, the problem is general in the sense that it can represent operations in virtually any yard, independently from its layout and from the type of operations which are carried out in it.

Keyphrases: Optimization, rail freight, simulation, yard management

@booklet{EasyChair:2602,
  author    = {Paola Pellegrini and Pierre Hosteins and Samuel Deleplanque and Norbert Adamko and Miloš Zat'ko and Joaquin Rodriguez},
  title     = {Closed-Loop Optimization and Simulation for Rail Freight Yard},
  howpublished = {EasyChair Preprint 2602},
  year      = {EasyChair, 2020}}