A Deep Q-Learning Network for Ship Stowage Planning Problem

Ship stowage plan is the management connection of quae crane scheduling and yard crane scheduling. The quality of ship stowage plan affects the productivity greatly. Previous studies mainly focuses on solving stowage planning problem with online searching algorithm, efficiency of which is significan...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Polish maritime research Ročník 24; číslo 1; s. 102 - 109
Hlavní autoři: Shen, Yifan, Zhao, Ning, Xia, Mengjue, Du, Xueqiang
Médium: Journal Article
Jazyk:angličtina
Vydáno: Gdansk Sciendo 27.11.2017
De Gruyter Brill Sp. z o.o., Paradigm Publishing Services
Témata:
Container terminal
Decision analysis
Deep Q-Leaning Network (DQN)
Evaluation
Generalization
Markov decision process
Planning
Scheduling
Ship stowage plan
Ships
Stowage (onboard equipment)
Training
Value function approximation
World class companies
ISSN:2083-7429, 1233-2585, 2083-7429
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Ship stowage plan is the management connection of quae crane scheduling and yard crane scheduling. The quality of ship stowage plan affects the productivity greatly. Previous studies mainly focuses on solving stowage planning problem with online searching algorithm, efficiency of which is significantly affected by case size. In this study, a Deep Q-Learning Network (DQN) is proposed to solve ship stowage planning problem. With DQN, massive calculation and training is done in pre-training stage, while in application stage stowage plan can be made in seconds. To formulate network input, decision factors are analyzed to compose feature vector of stowage plan. States subject to constraints, available action and reward function of Q-value are designed. With these information and design, an 8-layer DQN is formulated with an evaluation function of mean square error is composed to learn stowage planning. At the end of this study, several production cases are solved with proposed DQN to validate the effectiveness and generalization ability. Result shows a good availability of DQN to solve ship stowage planning problem.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2083-7429
1233-2585
2083-7429
DOI:10.1515/pomr-2017-0111