Determining capacity of shunting yards by combining graph classification with local search

Arno J.G. van de Ven, Y. Zhang, Wan-Jui Lee, H. Eshuis, A.M. Wilbik

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

Dutch Railways (NS) uses a shunt plan simulator to determine capacities of shunting yards. Central to this simulator is a local search heuristic. Solving this capacity determination problem is very time consuming, as it requires to solve an NP-hard shunting planning problem, and furthermore, the capacity has to determined for a large number of possible scenarios at over 30 shunting yards in The Netherlands. In this paper, we propose to combine machine learning with local search in order to speed up finding shunting plans in the capacity determination problem. The local search heuristic models the activities that take place on the shunting yard as nodes in an activity graph with precedence relations. Consequently, we apply the Deep Graph Convolutional Neural Network, which is a graph classification method, to predict whether local search will find a feasible shunt plan given an initial solution. Our experimental results show our approach can significantly reduce the simulation time in determining the capacity of a given shunting yard. This study demonstrates how machine learning can be used to boost optimization algorithms in an industrial application.

LanguageEnglish
Title of host publication11th International Conference on Agents and Artificial Intelligence (ICAART 2019)
EditorsLuc Steels, Ana Rocha, Jaap van den Herik
PublisherSCITEPRESS-Science and Technology Publications, Lda.
Pages285-293
Number of pages9
Volume2
ISBN (Electronic)9789897583506
StatePublished - 2019
Event11th International Conference on Agents and Artificial Intelligence, ICAART 2019 - Prague, Czech Republic
Duration: 19 Feb 201921 Feb 2019
http://www.icaart.org/

Conference

Conference11th International Conference on Agents and Artificial Intelligence, ICAART 2019
Abbreviated titleICAART2019
CountryCzech Republic
CityPrague
Period19/02/1921/02/19
Internet address

Fingerprint

Learning systems
Simulators
Industrial applications
Neural networks
Planning
Local search (optimization)

Keywords

  • Classification
  • Convolutional Neural Networks
  • Local Search
  • Machine Learning
  • Planning and Scheduling

Cite this

van de Ven, A. J. G., Zhang, Y., Lee, W-J., Eshuis, H., & Wilbik, A. M. (2019). Determining capacity of shunting yards by combining graph classification with local search. In L. Steels, A. Rocha, & J. van den Herik (Eds.), 11th International Conference on Agents and Artificial Intelligence (ICAART 2019) (Vol. 2, pp. 285-293). SCITEPRESS-Science and Technology Publications, Lda..
van de Ven, Arno J.G. ; Zhang, Y. ; Lee, Wan-Jui ; Eshuis, H. ; Wilbik, A.M./ Determining capacity of shunting yards by combining graph classification with local search. 11th International Conference on Agents and Artificial Intelligence (ICAART 2019). editor / Luc Steels ; Ana Rocha ; Jaap van den Herik. Vol. 2 SCITEPRESS-Science and Technology Publications, Lda., 2019. pp. 285-293
@inproceedings{ae9adbd7e4894d79ad0771794c4e9f15,
title = "Determining capacity of shunting yards by combining graph classification with local search",
abstract = "Dutch Railways (NS) uses a shunt plan simulator to determine capacities of shunting yards. Central to this simulator is a local search heuristic. Solving this capacity determination problem is very time consuming, as it requires to solve an NP-hard shunting planning problem, and furthermore, the capacity has to determined for a large number of possible scenarios at over 30 shunting yards in The Netherlands. In this paper, we propose to combine machine learning with local search in order to speed up finding shunting plans in the capacity determination problem. The local search heuristic models the activities that take place on the shunting yard as nodes in an activity graph with precedence relations. Consequently, we apply the Deep Graph Convolutional Neural Network, which is a graph classification method, to predict whether local search will find a feasible shunt plan given an initial solution. Our experimental results show our approach can significantly reduce the simulation time in determining the capacity of a given shunting yard. This study demonstrates how machine learning can be used to boost optimization algorithms in an industrial application.",
keywords = "Classification, Convolutional Neural Networks, Local Search, Machine Learning, Planning and Scheduling",
author = "{van de Ven}, {Arno J.G.} and Y. Zhang and Wan-Jui Lee and H. Eshuis and A.M. Wilbik",
year = "2019",
language = "English",
volume = "2",
pages = "285--293",
editor = "Luc Steels and Ana Rocha and {van den Herik}, Jaap",
booktitle = "11th International Conference on Agents and Artificial Intelligence (ICAART 2019)",
publisher = "SCITEPRESS-Science and Technology Publications, Lda.",

}

van de Ven, AJG, Zhang, Y, Lee, W-J, Eshuis, H & Wilbik, AM 2019, Determining capacity of shunting yards by combining graph classification with local search. in L Steels, A Rocha & J van den Herik (eds), 11th International Conference on Agents and Artificial Intelligence (ICAART 2019). vol. 2, SCITEPRESS-Science and Technology Publications, Lda., pp. 285-293, 11th International Conference on Agents and Artificial Intelligence, ICAART 2019, Prague, Czech Republic, 19/02/19.

Determining capacity of shunting yards by combining graph classification with local search. / van de Ven, Arno J.G.; Zhang, Y.; Lee, Wan-Jui ; Eshuis, H.; Wilbik, A.M.

11th International Conference on Agents and Artificial Intelligence (ICAART 2019). ed. / Luc Steels; Ana Rocha; Jaap van den Herik. Vol. 2 SCITEPRESS-Science and Technology Publications, Lda., 2019. p. 285-293.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Determining capacity of shunting yards by combining graph classification with local search

AU - van de Ven,Arno J.G.

AU - Zhang,Y.

AU - Lee,Wan-Jui

AU - Eshuis,H.

AU - Wilbik,A.M.

PY - 2019

Y1 - 2019

N2 - Dutch Railways (NS) uses a shunt plan simulator to determine capacities of shunting yards. Central to this simulator is a local search heuristic. Solving this capacity determination problem is very time consuming, as it requires to solve an NP-hard shunting planning problem, and furthermore, the capacity has to determined for a large number of possible scenarios at over 30 shunting yards in The Netherlands. In this paper, we propose to combine machine learning with local search in order to speed up finding shunting plans in the capacity determination problem. The local search heuristic models the activities that take place on the shunting yard as nodes in an activity graph with precedence relations. Consequently, we apply the Deep Graph Convolutional Neural Network, which is a graph classification method, to predict whether local search will find a feasible shunt plan given an initial solution. Our experimental results show our approach can significantly reduce the simulation time in determining the capacity of a given shunting yard. This study demonstrates how machine learning can be used to boost optimization algorithms in an industrial application.

AB - Dutch Railways (NS) uses a shunt plan simulator to determine capacities of shunting yards. Central to this simulator is a local search heuristic. Solving this capacity determination problem is very time consuming, as it requires to solve an NP-hard shunting planning problem, and furthermore, the capacity has to determined for a large number of possible scenarios at over 30 shunting yards in The Netherlands. In this paper, we propose to combine machine learning with local search in order to speed up finding shunting plans in the capacity determination problem. The local search heuristic models the activities that take place on the shunting yard as nodes in an activity graph with precedence relations. Consequently, we apply the Deep Graph Convolutional Neural Network, which is a graph classification method, to predict whether local search will find a feasible shunt plan given an initial solution. Our experimental results show our approach can significantly reduce the simulation time in determining the capacity of a given shunting yard. This study demonstrates how machine learning can be used to boost optimization algorithms in an industrial application.

KW - Classification

KW - Convolutional Neural Networks

KW - Local Search

KW - Machine Learning

KW - Planning and Scheduling

UR - http://www.scopus.com/inward/record.url?scp=85064832609&partnerID=8YFLogxK

M3 - Conference contribution

VL - 2

SP - 285

EP - 293

BT - 11th International Conference on Agents and Artificial Intelligence (ICAART 2019)

PB - SCITEPRESS-Science and Technology Publications, Lda.

ER -

van de Ven AJG, Zhang Y, Lee W-J, Eshuis H, Wilbik AM. Determining capacity of shunting yards by combining graph classification with local search. In Steels L, Rocha A, van den Herik J, editors, 11th International Conference on Agents and Artificial Intelligence (ICAART 2019). Vol. 2. SCITEPRESS-Science and Technology Publications, Lda.2019. p. 285-293.