QUANTUM-INSPIRED AI METAHEURISTIC FRAMEWORK FOR MULTI-OBJECTIVE OPTIMIZATION IN INDUSTRIAL PRODUCTION SCHEDULING

Abstract

This study presents a quantum-inspired artificial intelligence metaheuristic (QI-AIM) framework for multi-objective industrial production scheduling, integrating algorithmic innovation with empirical validation across real manufacturing environments. Production scheduling inherently involves allocating limited resources—machines, tools, labor, and energy—to optimize multiple conflicting objectives such as makespan, tardiness, and energy cost. Existing evolutionary multi-objective algorithms (EMOAs) achieve competent performance but often struggle to maintain diversity and interpretability under industrial constraints. The proposed QI-AIM introduces a Q-bit–based priority representation, adaptive rotation updates, and episodic local search, augmented by surrogate modeling for energy-aware objectives. Implemented on classical hardware, QI-AIM emulates quantum principles to balance global exploration and local exploitation, enabling the generation of high-quality, well-distributed Pareto fronts within fixed computational budgets. A cross-sectional, multi–case empirical design links algorithmic outcomes to managerial perceptions using validated five-point Likert scales measuring perceived usefulness, ease of use, satisfaction, and intention to continue use. Data from flexible job-shop, flow-shop, and mixed-layout plants confirm that QI-AIM achieved the highest hypervolume (0.72) and lowest IGD (0.154) across 30-run benchmarks, outperforming NSGA-II, MOEA/D, and GA-LS hybrids while remaining computationally efficient. Regression analyses reveal that perceived usefulness significantly mediates the relationship between observed performance improvements and behavioral intention, with tardiness and makespan reductions explaining 45–53% of variance in acceptance outcomes. Survey means above 4.0 indicate strong managerial endorsement of the system’s clarity, decision support, and interpretability. The findings establish that QI-AIM not only enhances operational performance but also strengthens decision confidence through transparent portfolio visualization of trade-offs. By coupling reproducible algorithmic design with human-centered evaluation, this research contributes a scalable, explainable, and adoption-relevant framework for Industry 4.0–aligned production scheduling and provides a blueprint for integrating advanced optimization with managerial decision processes.