Solving the mixed model sequencing problem with reinforcement learning and metaheuristics

被引:5
作者
Brammer, Janis [1 ]
Lutz, Bernhard [2 ]
Neumann, Dirk [2 ]
机构
[1] Volkswagen AG, Berliner Ring 2, D-38440 Wolfsburg, Germany
[2] Univ Freiburg, Rempartstr 16, D-79098 Freiburg, Germany
关键词
Scheduling; Mixed model sequencing; Reinforcement learning; Metaheuristics; Mixed-integer linear programming; ASSEMBLY LINES SURVEY; WORK OVERLOAD; MINIMIZE;
D O I
10.1016/j.cie.2021.107704
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
This study presents a reinforcement learning (RL) approach for the mixed model sequencing (MMS) problem with a minimization of work overload situations. The proposed approach generates the sequence in a constructive way, so that an action denotes the model to be sequenced next. The trained policy quickly creates an initial sequence, which allows us to use the cutoff time to further improve the solution quality with a metaheuristic. Our numerical evaluation based on an existing benchmark dataset shows that our approach is superior to established methods if the demand plan follows its expected distribution from the learning process.
引用
收藏
页数:9
相关论文
共 36 条
[1]   An adaptive genetic algorithm approach for the mixed-model assembly line sequencing problem [J].
Akgunduz, Onur Serkan ;
Tunali, Semra .
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH, 2010, 48 (17) :5157-5179
[2]   Minimising work overload in mixed-model assembly lines with different types of operators: a case study from the truck industry [J].
Aroui, Karim ;
Alpan, Gulgun ;
Frein, Yannick .
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH, 2017, 55 (21) :6305-6326
[3]   Executing production schedules in the face of uncertainties: A review and some future directions [J].
Aytug, H ;
Lawley, MA ;
McKay, K ;
Mohan, S ;
Uzsoy, R .
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 2005, 161 (01) :86-110
[4]   GRASP for sequencing mixed models in an assembly line with work overload, useless time and production regularity [J].
Bautista J. ;
Alfaro-Pozo R. ;
Batalla-García C. .
Progress in Artificial Intelligence, 2016, 5 (01) :27-33
[5]   Solving mixed model sequencing problem in assembly lines with serial workstations with work overload minimisation and interruption rules [J].
Bautista, Joaquin ;
Cano, Alberto .
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 2011, 210 (03) :495-513
[6]   Machine learning for combinatorial optimization: A methodological tour d'horizon [J].
Bengio, Yoshua ;
Lodi, Andrea ;
Prouvost, Antoine .
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 2021, 290 (02) :405-421
[7]   Sequencing mixed-model assembly lines to minimise the number of work overload situations [J].
Boysen, Nils ;
Kiel, Mirko ;
Scholl, Armin .
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH, 2011, 49 (16) :4735-4760
[8]   Resequencing of mixed-model assembly lines: Survey and research agenda [J].
Boysen, Nils ;
Scholl, Armin ;
Wopperer, Nico .
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 2012, 216 (03) :594-604
[9]   Sequencing mixed-model assembly lines: Survey, classification and model critique [J].
Boysen, Nils ;
Fliedner, Malte ;
Scholl, Armin .
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 2009, 192 (02) :349-373
[10]  
Brockman Greg, 2016, arXiv