Hybrid fuzzy genetic algorithm for the integration of process planning and scheduling for distributed flexible job shop

The study explores the effectiveness of a hybrid fuzzy logic-based genetic algorithm model in addressing process planning and scheduling in a distributed flexible job shop. In this paper, the proposed hybrid fuzzy-GA framework presents a novel integrated mechanism of hybridization through integrating and synchronizing the main three function blocks (GA, fitness, and fuzzy logic) in order to achieve an adaptive control mechanism through dynamic adaptation that utilizes fuzzy inference to tune GA main parameters (i.e., mutation and crossover rates) dynamically and in real time within the evolution process, improving the algorithm’s ability to both explore and exploit the search space efficiently and effectively. Fuzzy logic assigns an output based on inserted input probability and human estimation assumptions. The genetic algorithm is a popular search technique used for solving optimization problems. The crossover and mutation rates, being key parameters in the genetic algorithm, are determined using fuzzy logic to improve overall effectiveness. The algorithm employs chromosome ∅s to minimize the makespan. In chromosome ∅s, the decisions are implicitly determined using heuristic rules that ensure load balancing among manufacturing resources. In this paper, the fuzzy genetic algorithm demonstrates its effectiveness to reduce the makespan by 6% and generate better results with a chance of 90% than standard genetic algorithms. Moreover, this study shows that there is 91% chance it yields outputs within 5% of the absolute minimum.