Tackling Uncertainty in Online Multimodal Transportation Planning Using Deep Reinforcement Learning

Amirreza Farahani; Laura Genga; Remco M. Dijkman

doi:10.1007/978-3-030-87672-2_38

Tackling Uncertainty in Online Multimodal Transportation Planning Using Deep Reinforcement Learning

Amirreza Farahani, Laura Genga, Remco M. Dijkman

Information Systems IE&IS

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

Abstract

In this paper we tackle the container allocation problem in multimodal transportation planning under uncertainty in container arrival times, using Deep Reinforcement Learning. The proposed approach can take real-time decisions on allocating individual containers to a truck or to trains, while a transportation plan is being executed. We evaluated our method using data that reflect a realistic scenario, designed on the basis of a case study at a logistics company with three different uncertainty levels based on the probability of delays in container arrivals. The experiments show that Deep Reinforcement Learning methods outperform heuristics, a stochastic programming method, and methods that use periodic re-planning, in terms of total transportation costs at all levels of uncertainty, obtaining an average cost difference with the optimal solution within 0.37% and 0.63%.

Original language	English
Title of host publication	Computational Logistics - 12th International Conference, ICCL 2021, Proceedings
Subtitle of host publication	12th International Conference, ICCL 2021, Enschede, The Netherlands, September 27–29, 2021, Proceedings
Editors	Martijn Mes, Eduardo Lalla-Ruiz, Stefan Voß
Place of Publication	Cham
Publisher	Springer, Cham
Chapter	38
Pages	578-593
Number of pages	16
ISBN (Electronic)	978-3-030-87672-2
ISBN (Print)	978-3-030-87671-5
DOIs	https://doi.org/10.1007/978-3-030-87672-2_38
Publication status	Published - 22 Sep 2021
Event	12th International Conference on Computational Logistics, ICCL 2021 - University of Twente, Enschede, Netherlands Duration: 27 Sep 2021 → 29 Sep 2021 Conference number: 12 https://iccl2021.nl/

Publication series

Name	Lecture Notes in Computer Science (LNCS)
Publisher	Springer
Volume	13004
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	12th International Conference on Computational Logistics, ICCL 2021
Country/Territory	Netherlands
City	Enschede
Period	27/09/21 → 29/09/21
Internet address	https://iccl2021.nl/

Keywords

Optimization
Deep Reinforcement Learning
Online planning under uncertainty
Multimodal transport

Access to Document

10.1007/978-3-030-87672-2_38

Cite this

Farahani, A., Genga, L., & Dijkman, R. M. (2021). Tackling Uncertainty in Online Multimodal Transportation Planning Using Deep Reinforcement Learning. In M. Mes, E. Lalla-Ruiz, & S. Voß (Eds.), Computational Logistics - 12th International Conference, ICCL 2021, Proceedings: 12th International Conference, ICCL 2021, Enschede, The Netherlands, September 27–29, 2021, Proceedings (pp. 578-593). (Lecture Notes in Computer Science (LNCS); Vol. 13004). Springer, Cham. https://doi.org/10.1007/978-3-030-87672-2_38

Farahani, Amirreza ; Genga, Laura ; Dijkman, Remco M. / Tackling Uncertainty in Online Multimodal Transportation Planning Using Deep Reinforcement Learning. Computational Logistics - 12th International Conference, ICCL 2021, Proceedings: 12th International Conference, ICCL 2021, Enschede, The Netherlands, September 27–29, 2021, Proceedings. editor / Martijn Mes ; Eduardo Lalla-Ruiz ; Stefan Voß. Cham : Springer, Cham, 2021. pp. 578-593 (Lecture Notes in Computer Science (LNCS)).

@inproceedings{5d06cbd6e54b4b63be546618c55ef9aa,

title = "Tackling Uncertainty in Online Multimodal Transportation Planning Using Deep Reinforcement Learning",

abstract = "In this paper we tackle the container allocation problem in multimodal transportation planning under uncertainty in container arrival times, using Deep Reinforcement Learning. The proposed approach can take real-time decisions on allocating individual containers to a truck or to trains, while a transportation plan is being executed. We evaluated our method using data that reflect a realistic scenario, designed on the basis of a case study at a logistics company with three different uncertainty levels based on the probability of delays in container arrivals. The experiments show that Deep Reinforcement Learning methods outperform heuristics, a stochastic programming method, and methods that use periodic re-planning, in terms of total transportation costs at all levels of uncertainty, obtaining an average cost difference with the optimal solution within 0.37% and 0.63%.",

keywords = "Optimization, Deep Reinforcement Learning, Online planning under uncertainty, Multimodal transport",

author = "Amirreza Farahani and Laura Genga and Dijkman, {Remco M.}",

year = "2021",

month = sep,

day = "22",

doi = "10.1007/978-3-030-87672-2_38",

language = "English",

isbn = "978-3-030-87671-5",

series = "Lecture Notes in Computer Science (LNCS)",

publisher = "Springer, Cham",

pages = "578--593",

editor = "Martijn Mes and Lalla-Ruiz, {Eduardo } and Stefan Vo{\ss}",

booktitle = "Computational Logistics - 12th International Conference, ICCL 2021, Proceedings",

note = "12th International Conference on Computational Logistics, ICCL 2021 ; Conference date: 27-09-2021 Through 29-09-2021",

url = "https://iccl2021.nl/",

}

Farahani, A , Genga, L & Dijkman, RM 2021, Tackling Uncertainty in Online Multimodal Transportation Planning Using Deep Reinforcement Learning. in M Mes, E Lalla-Ruiz & S Voß (eds), Computational Logistics - 12th International Conference, ICCL 2021, Proceedings: 12th International Conference, ICCL 2021, Enschede, The Netherlands, September 27–29, 2021, Proceedings. Lecture Notes in Computer Science (LNCS), vol. 13004, Springer, Cham, Cham, pp. 578-593, 12th International Conference on Computational Logistics, ICCL 2021, Enschede, Netherlands, 27/09/21. https://doi.org/10.1007/978-3-030-87672-2_38

Tackling Uncertainty in Online Multimodal Transportation Planning Using Deep Reinforcement Learning. / Farahani, Amirreza ; Genga, Laura ; Dijkman, Remco M.

Computational Logistics - 12th International Conference, ICCL 2021, Proceedings: 12th International Conference, ICCL 2021, Enschede, The Netherlands, September 27–29, 2021, Proceedings. ed. / Martijn Mes; Eduardo Lalla-Ruiz; Stefan Voß. Cham : Springer, Cham, 2021. p. 578-593 (Lecture Notes in Computer Science (LNCS); Vol. 13004).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Tackling Uncertainty in Online Multimodal Transportation Planning Using Deep Reinforcement Learning

AU - Farahani, Amirreza

AU - Genga, Laura

AU - Dijkman, Remco M.

N1 - Conference code: 12

PY - 2021/9/22

Y1 - 2021/9/22

N2 - In this paper we tackle the container allocation problem in multimodal transportation planning under uncertainty in container arrival times, using Deep Reinforcement Learning. The proposed approach can take real-time decisions on allocating individual containers to a truck or to trains, while a transportation plan is being executed. We evaluated our method using data that reflect a realistic scenario, designed on the basis of a case study at a logistics company with three different uncertainty levels based on the probability of delays in container arrivals. The experiments show that Deep Reinforcement Learning methods outperform heuristics, a stochastic programming method, and methods that use periodic re-planning, in terms of total transportation costs at all levels of uncertainty, obtaining an average cost difference with the optimal solution within 0.37% and 0.63%.

AB - In this paper we tackle the container allocation problem in multimodal transportation planning under uncertainty in container arrival times, using Deep Reinforcement Learning. The proposed approach can take real-time decisions on allocating individual containers to a truck or to trains, while a transportation plan is being executed. We evaluated our method using data that reflect a realistic scenario, designed on the basis of a case study at a logistics company with three different uncertainty levels based on the probability of delays in container arrivals. The experiments show that Deep Reinforcement Learning methods outperform heuristics, a stochastic programming method, and methods that use periodic re-planning, in terms of total transportation costs at all levels of uncertainty, obtaining an average cost difference with the optimal solution within 0.37% and 0.63%.

KW - Optimization

KW - Deep Reinforcement Learning

KW - Online planning under uncertainty

KW - Multimodal transport

UR - http://www.scopus.com/inward/record.url?scp=85116377670&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-87672-2_38

DO - 10.1007/978-3-030-87672-2_38

M3 - Conference contribution

SN - 978-3-030-87671-5

T3 - Lecture Notes in Computer Science (LNCS)

SP - 578

EP - 593

BT - Computational Logistics - 12th International Conference, ICCL 2021, Proceedings

A2 - Mes, Martijn

A2 - Lalla-Ruiz, Eduardo

A2 - Voß, Stefan

PB - Springer, Cham

CY - Cham

T2 - 12th International Conference on Computational Logistics, ICCL 2021

Y2 - 27 September 2021 through 29 September 2021

ER -

Farahani A , Genga L , Dijkman RM. Tackling Uncertainty in Online Multimodal Transportation Planning Using Deep Reinforcement Learning. In Mes M, Lalla-Ruiz E, Voß S, editors, Computational Logistics - 12th International Conference, ICCL 2021, Proceedings: 12th International Conference, ICCL 2021, Enschede, The Netherlands, September 27–29, 2021, Proceedings. Cham: Springer, Cham. 2021. p. 578-593. (Lecture Notes in Computer Science (LNCS)). https://doi.org/10.1007/978-3-030-87672-2_38

Tackling Uncertainty in Online Multimodal Transportation Planning Using Deep Reinforcement Learning

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this