Energy-efficient distributed permutation flow shop scheduling problem using a multi-objective whale swarm algorithm

Guangchen Wang,Liang Gao,Xinyu Li,Peigen Li,M. Fatih Tasgetiren
DOI: https://doi.org/10.1016/j.swevo.2020.100716
IF: 10.267
2020-09-01
Swarm and Evolutionary Computation
Abstract:<p>Production scheduling is of great significance in improving production effectiveness while the energy-efficient problem is one of most concerned problems for researchers and manufacturers. Thus, this study investigates the energy-efficient distributed permutation flow shop scheduling problem (DPFSP) with the objectives of makespan and energy consumption. The DPFSP is an extension of permutation flow shop problem (PFSP) considering a set of identical factories. This paper presents a multi-objective mixed integer programming model based on the three sub-problems: allocating jobs among factories, scheduling the jobs in each factory and determining speed upon each job. A multi-objective whale swarm algorithm (MOWSA) is proposed to solve this energy-efficient DPFSP. A new problem-dependent local search is developed to improve the exploitation capability of MOWSA. Moreover, the updating exploitation mechanism is presented to enhance energy efficiency without affecting production efficiency. Finally, the extensive comparison experiments are designed to demonstrate the effectiveness of proposed MOWSA, problem-dependent local search and updating exploitation mechanism. The results indicate the effectiveness of MOWSA and the superior performance over NSGA-II, SPEA2, PAES and MDEA, and also demonstrate that the proposed algorithm can significantly reduce the energy consumption compared with other algorithms.</p>
computer science, artificial intelligence, theory & methods
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the optimization problem between energy efficiency and production scheduling. Specifically, it is to simultaneously minimize the total energy consumption (TEC) and the makespan in the Distributed Permutation Flow Shop Scheduling Problem (DPFSP). In the modern manufacturing environment, with the decrease of fuel resources and a large amount of carbon dioxide emissions, energy conservation and emission reduction have become important issues of concern for enterprises and researchers. Therefore, this research proposes a Multi - Objective Whale Swarm Algorithm (MOWSA) to solve the energy - efficient DPFSP considering setup time. By introducing a new problem - related local search and update utilization mechanism, this algorithm aims to significantly reduce energy consumption while improving production efficiency. ### Specific Problem Description - **Distributed Permutation Flow Shop Scheduling Problem (DPFSP)**: This is a problem extended from the Permutation Flow Shop Problem (PFSP), considering multiple identical factories. Each job needs to go through a series of processing stages, and each stage corresponds to a machine. Jobs can be allocated among different factories, and the job order within each factory also needs to be determined. - **Multi - objective optimization**: The research objective is to reduce the total energy consumption (TEC) while minimizing the makespan. Among them, TEC includes three parts: Idle Energy Consumption (IEC), Setup Energy Consumption (SEC), and Processing Energy Consumption (PEC). - **Algorithm design**: A multi - objective optimization algorithm (MOWSA) based on the whale swarm algorithm is proposed, and the performance of the algorithm is enhanced by introducing problem - related local search and update utilization mechanisms. ### Mathematical Model In order to describe the problem more accurately, the following symbols and formulas are defined: - **Idle Energy Consumption (IEC)**: \[ IEC=\sum_{j = 1}^{n}\sum_{k = 1}^{f}\sum_{i = 1}^{m}PI_{i}\cdot t_{ii;k;j} \] where, \[ t_{ii;k;j}=\sum_{c = 1}^{e}\sum_{g = 1}^{n/C0}Y_{j;i;c}\cdot X_{j,g + 1:k}\cdot\left(\frac{C_{g + 1;i;k}-C_{g;i;k}-p_{j;i}/v_{c}-s_{j;i}}{C0}\right) \] - **Setup Energy Consumption (SEC)**: \[ SEC=\sum_{j = 1}^{n}\sum_{g = 1}^{n}\sum_{k = 1}^{f}\sum_{i = 1}^{m}PS_{i}\cdot X_{j,g:k}\cdot s_{j;i} \] - **Processing Energy Consumption (PEC)**: \[ PEC=\sum_{c = 1}^{e}\sum_{j = 1}^{n}\sum_{k = 1}^{f}\sum_{i = 1}^{m}PP_{i;c}\cdot Y_{j;i;c}\cdot t_{pi;k;j} \] where, \[ t_{pi;k;j}=\sum_{g = 1}^{n}\sum_{c = 1}^{e}\sum_{k = 1}^{f}\sum_{i = 1}^{m}\sum_{j = 1}^{n}X_{j,g:k}\cdot Y_{j;i;c}\cdot\frac{p_{j;i}}{v_{c}} \] - **Total Energy Consumption (TEC)**: \[ TEC = IEC+SEC + PEC \] ### Objective Function - Minimize the makespan: