MOEA/D com Busca Local para o Flow Shop Multiobjetivo

The flow shop is a combinatorial problem for which a set of tasks should be sequentially processed by a set of machines. If the sequence of tasks is the same for all machines, then it is called permutation fow shop (PFSP) which is frequently found in assembly lines of industrial plants. This work considers the multiobjective PFSP that minimizes makespan and tardiness. The makespan is the time interval necessary to complete all tasks in all machines and tardiness is the time delay experienced by a task for a given duedate. In this work we consider a multi-objective framework called MOEA/D-DRA (Multi-objective Evolutionary Algorithm based on Decomposition with Dynamic Resource Allocation). In additon, a NEH heuristic is included into a local search embeded into MOEA/D-DRA (differently from other approaches which consider NEH only for the initial population. Experiments have been made for 11 instances of PFSP with 20 to 200 tasks and 5 to 20 machines. The obtained results are compared with those of MOEA/D-DRA using NEH only at the begining. These results show that MOEA/D-DRA with no local search outperforms the proposed approach only for two instances. Resumo. O flow shop é um problema combinatorial em que um conjunto de tarefas devem ser processadas sequencialmente em um conjunto de máquinas. Se a sequência de tarefas é a mesma para todas as máquinas, tem-se o flow shop de permutação (PFSP), frequentemente encontrado em linhas de montagem de plantas industriais. Este trabalho considera o PFSP multiobjetivo que minimiza “makespan” e “tardiness”. O “makespan” é o tempo necessário para processar todas as tarefas em todas as máquinas e o “tardiness” é o atraso para finalizar uma tarefa em um dado prazo. A plataforma multiobjetivo utilizada é o MOEA/D-DRA (“Multi-objective Evolutionary Algorithm based on Decomposition with Dynamic Resource Allocation”). Para tanto, é utilizada a heurı́stica (NEH) em uma busca local acoplada ao MOEA/D-DRA (diferentemente de outros trabalhos que utilizam NEH apenas na inicialização). Experimentos com esta abordagem proposta são realizados para 11 instâncias do PFSP com 20 a 200 tarefas e 5 a 20 máquinas. Os resultados obtidos são comparados com os do MOEA/D-DRA utilizando o mecanismo NEH apenas na inicialização da população. Estes resultados mostram que a versão do MOEA/D-DRA sem busca local supera a abordagem proposta em apenas duas instâncias.

[1]  Chongguo Li,et al.  Diversity Metrics in Multi-objective Optimization: Review and Perspective , 2007, 2007 IEEE International Conference on Integration Technology.

[2]  Éric D. Taillard,et al.  Benchmarks for basic scheduling problems , 1993 .

[3]  Dipti Srinivasan,et al.  A Survey of Multiobjective Evolutionary Algorithms Based on Decomposition , 2017, IEEE Transactions on Evolutionary Computation.

[4]  Kenneth R. Baker,et al.  Principles of Sequencing and Scheduling , 2018 .

[5]  F. Liu,et al.  An improved multi-objective memetic algorithm for bi-objective permutation flow shop scheduling , 2017, 2017 International Conference on Service Systems and Service Management.

[6]  Gary B. Lamont,et al.  Evolutionary Algorithms for Solving Multi-Objective Problems , 2002, Genetic Algorithms and Evolutionary Computation.

[7]  Rubén Ruiz,et al.  Restarted Iterated Pareto Greedy algorithm for multi-objective flowshop scheduling problems , 2011, Comput. Oper. Res..

[8]  Li-Chen Fu,et al.  Multiobjective Permutation Flow Shop Scheduling Using a Memetic Algorithm with an NEH-Based Local Search , 2009, ICIC.

[9]  David J. Groggel,et al.  Practical Nonparametric Statistics , 2000, Technometrics.

[10]  Qingfu Zhang,et al.  The performance of a new version of MOEA/D on CEC09 unconstrained MOP test instances , 2009, 2009 IEEE Congress on Evolutionary Computation.

[11]  Qingfu Zhang,et al.  Multiobjective evolutionary algorithms: A survey of the state of the art , 2011, Swarm Evol. Comput..

[12]  Mehmet Mutlu Yenisey,et al.  Multi-objective permutation flow shop scheduling problem: Literature review, classification and current trends , 2014 .

[13]  Xiangtao Li,et al.  Multiobjective Local Search Algorithm-Based Decomposition for Multiobjective Permutation Flow Shop Scheduling Problem , 2015, IEEE Transactions on Engineering Management.

[14]  Qingfu Zhang,et al.  Adaptive Operator Selection With Bandits for a Multiobjective Evolutionary Algorithm Based on Decomposition , 2014, IEEE Transactions on Evolutionary Computation.

[15]  El-Ghazali Talbi,et al.  Design of multi-objective evolutionary algorithms: application to the flow-shop scheduling problem , 2002, Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600).

[16]  Victor Fernandez-Viagas,et al.  A new vision of approximate methods for the permutation flowshop to minimise makespan: State-of-the-art and computational evaluation , 2017, Eur. J. Oper. Res..

[17]  Qingfu Zhang,et al.  MOEA/D: A Multiobjective Evolutionary Algorithm Based on Decomposition , 2007, IEEE Transactions on Evolutionary Computation.

[18]  Inyong Ham,et al.  A heuristic algorithm for the m-machine, n-job flow-shop sequencing problem , 1983 .

[19]  Rong-Qiang Zeng,et al.  Solving bi-objective flow shop problem with hybrid path relinking algorithm , 2013, Appl. Soft Comput..

[20]  Ravi Sethi,et al.  The Complexity of Flowshop and Jobshop Scheduling , 1976, Math. Oper. Res..