A Genetic Algorithm with Feasibility-Agnostic Encoding and Three-Phase Decoding for Scheduling Semiconductor Manufacturing Facilities Under Constrained Queue Time

The semiconductor manufacturing industry faces significant challenges in scheduling operations due to constraints such as queue time (Q-time), sequence-dependent setups, and re-entrant flow. In particular, Q-time constraints play a critical role in maintaining product quality and yield, since exceeding allowable Q-time limits leads to defects in semiconductor devices. To address such scheduling challenges, genetic algorithms (GAs) have been extensively developed over the past few decades. Yet, their rigid compliance with job precedence constraints poses a limitation to exploring a broader solution space, making it challenging to achieve high-quality schedules. In this paper, we propose a novel GA for solving scheduling problems in semiconductor manufacturing facilities with sequence-dependent setups and constrained Q-time. Specifically, we introduce a feasibility-agnostic encoding scheme and a three-phase decoding procedure that consists of job assignment, feasibility adjustment, and machine allocation. Furthermore, customized crossover and mutation operators are incorporated to enhance the exploration of the solution space. Comprehensive experiments on six datasets demonstrate the superiority of the proposed method compared to other metaheuristics, providing a viable solution for scheduling semiconductor manufacturing systems while balancing makespan and Q-time-related objectives.

Recommended citation: Park, I., & Huh, J.*, A Genetic Algorithm with Feasibility-Agnostic Encoding and Three-Phase Decoding for Scheduling Semiconductor Manufacturing Facilities under Constrained Queue Time, submitted to Swarm and Evolutionary Computation.
Download Paper