Elahi Ihsan, Ali Hamid, Asif Muhammad, Iqbal Kashif, Ghadi Yazeed, Alabdulkreem Eatedal
Department of Computer Science, National Textile University, Faisalabad, Punjab, Pakistan.
Department of Computational Sciences, The University of Faisalabad (TUF), Faisalabad, Punjab, Pakistan.
PeerJ Comput Sci. 2022 Mar 22;8:e932. doi: 10.7717/peerj-cs.932. eCollection 2022.
Optimization is challenging even after numerous multi-objective evolutionary algorithms have been developed. Most of the multi-objective evolutionary algorithms failed to find out the best solutions spread and took more fitness evolution value to find the best solution. This article proposes an extended version of a multi-objective group counseling optimizer called MOGCO-II. The proposed algorithm is compared with MOGCO, MOPSO, MOCLPSO, and NSGA-II using the well-known benchmark problem such as Zitzler Deb Thieler (ZDT) function. The experiments show that the proposed algorithm generates a better solution than the other algorithms. The proposed algorithm also takes less fitness evolution value to find the optimal Pareto front. Moreover, the textile dyeing industry needs a large amount of fresh water for the dyeing process. After the dyeing process, the textile dyeing industry discharges a massive amount of polluted water, which leads to serious environmental problems. Hence, we proposed a MOGCO-II based optimization scheduling model to reduce freshwater consumption in the textile dyeing industry. The results show that the optimization scheduling model reduces freshwater consumption in the textile dyeing industry by up to 35% compared to manual scheduling.
即使在众多多目标进化算法被开发出来之后，优化仍然具有挑战性。大多数多目标进化算法未能找出分布的最佳解决方案，并且需要更多的适应度进化值来找到最佳解决方案。本文提出了一种多目标群体咨询优化器的扩展版本，称为MOGCO-II。使用诸如齐茨勒-德布-蒂勒（ZDT）函数等著名的基准问题，将所提出的算法与MOGCO、MOPSO、MOCLPSO和NSGA-II进行比较。实验表明，所提出的算法比其他算法能产生更好的解决方案。所提出的算法在找到最优帕累托前沿时也需要更少的适应度进化值。此外，纺织印染行业在染色过程中需要大量的淡水。染色过程结束后，纺织印染行业会排放大量的污水，这会导致严重的环境问题。因此，我们提出了一种基于MOGCO-II的优化调度模型，以减少纺织印染行业的淡水消耗。结果表明，与人工调度相比，该优化调度模型可将纺织印染行业的淡水消耗减少多达35%。
