Automated train scheduling system using Genetic Algorithm

Abstract

The increment of interest for transport service in rail activity stipulates higher proportion of consumed infrastructure capacity. In this system of activity stream, even minor deviations from the arranged schedule can affect its stability, and this can bring about a noteworthy diminishment of the nature of transport service. Given the way that the railroad business is as of now running without much abundance limit, better arranging and planning instruments are expected to successfully deal with the rare assets, keeping in mind the ultimate goal to adapt to the quickly expanding interest for rail route transportation. The objective of operational scheduling of trains is to safely move about each train, as fast as possible, from its origin to its destination such that the total delay of all trains can be minimized. This paper presents a (Fixed Path + Genetic Algorithm) heuristic model, an optimization-based approach for scheduling of trains. The Fixed Path model assumes that path of the trains is fixed for preparing the train schedule. The Genetic Algorithm is used for selecting the path for each train that takes minimum time to arrive at the destination. Together it presents a schedule that can minimize the travel time of each train maximizing capacity of the network. This paper proves the fact that applying the proposed model, rail traffic can be improved regarding the increase of the timetable stability and maximizing capacity subject to safety constraints.

Artificial Intelligence Applied on Traffic Planning and Management for Rail Transport: A Review and Perspective

“…For minimizing the travel time of each train and maximizing capacity of the network, the authors in [42] presented an optimization-based train scheduling approach, i.e., fxed path + genetic algorithm. Also, the GA was used for selecting the assumed fxed path for each train.…”

Artificial Intelligence in Railway Transport: Taxonomy, Regulations, and Applications

“…defect prediction [80] failure prediction [81] defect detection [82] Y: defect prediction [83] defect detection [84], [85] Safety and security Y: train protection [86], speed error reduction [87] Y: accidents [53] disruptions [88] Autonomous driving and control Y: energy optimization [89] intelligent train control [90] Y: intelligent train control [55] Traffic planning and management Y: train timetabling [91], [92] Y: delay analysis [40], train rescheduling [93] train timetabling [63], [94], train shunting [95] Revenue management P: revenue simulation [96] P: overall revenue management [97] inventory control and prediction [98] Transport policy P: energy network policy making [99] U Passenger mobility P: demand forecasting [100] Y: flow prediction [101], [102] and reinforcement learning for optimal train control. Reference [89] proposed a method for energy optimisation of the train movement applying control based on genetic algorithms.…”

“…Computational experiments were conducted based on a German railway network. Reference [91] presents a heuristic model based on the concept of Fixed Path + Genetic Algorithm. The Fixed Path model assumes that the path of the trains is fixed for preparing the train schedule.…”

Intelligent System for Step-by-Step Control During the Operational Rebuilding of the Schedule

“…In [3], an approach is presented that is most closely approximated to the task of dispatch control. However, it, like [5], is solved only for one type of traffic. Such a restriction simplifies the search for a solution, since in such a model there are no rules and restrictions on various types of traffic, for example, the priority of passing trains or restrictions on the maximum allowed speed depending on the type of traffic.…”