Cost–Quality–Energy Optimization for a Smart Manufacturing Line Using a Public Industrial Metrics Dataset

Abstract

Modern smart-manufacturing lines face tight energy budgets and stringent quality targets while delivering stable, low-cost output. This study presents a transparent, operations-ready optimization framework: public shop-floor data (production, maintenance, energy, downtime, defects) are distilled into conservative, auditable guardrails via medians and quantiles, and an interpretable linear program selects daily production, maintenance hours, and downtime under time, energy, and quality limits. The model is solved with Gurobi and interpreted through shadow prices, one-at-a-time sensitivity, and ε-constraint Pareto frontiers for cost–energy and cost–quality trade-offs. The baseline indicates demand, rather than the capacity, governs the operating point (binding demand, marginal maintenance minimum, slack time/energy/downtime). Sensitivity and frontier analyses show costs remain flat under moderate tightening of energy or quality guardrails and rise sharply only near the energy feasibility wall or with substantially stricter quality. The outcome is a simple, reproducible planning tool that quantifies the value of extra demand, time, energy, or quality and supports day-to-day scheduling.