Human-centric optimization of multi-row facility layouts: an NSGA-II-based approach balancing noise exposure and space utilization
摘要
In the context of rapid technological advancements and intensifying market competition, the optimization of production systems has become essential for improving operational performance. This study addresses the Multi-Row Facility Layout Problem (MRFLP), which seeks the optimal arrangement of facilities across multiple rows. While traditional approaches primarily emphasize efficiency and material-handling costs, ergonomic concerns despite their significant influence on worker health, safety, and productivity–are often overlooked. This paper proposes a comprehensive methodology that integrates ergonomic considerations, specifically noise exposure, into the MRFLP. A mathematical formulation is developed to explicitly model acoustic exposure based on physical sound propagation principles, and a metaheuristic solution framework is introduced. In the first phase, a Genetic Algorithm (GA) is employed to minimize noise exposure in a single-objective setting. In the second phase, a Non-dominated Sorting Genetic Algorithm II (NSGA-II) is used to solve a bi-objective model that jointly minimizes noise exposure and maximizes space utilization. To enable rigorous and reproducible evaluation, experiments are conducted on synthetically generated MRFLP instances. All machine dimensions, clearances, and noise emission levels were calibrated using documented industrial ranges and manufacturer specifications, ensuring that the test cases reflect realistic industrial behavior. Numerical results confirm the effectiveness of the proposed methodology, demonstrating its ability to produce ergonomic and space-efficient layouts while providing a robust foundation for future real-world applications.