Closed-loop deterministic inventory optimization with real-time IoT-based demand tracking in sustainable smart manufacturing

Industry 4.0 technologies are reshaping production and inventory management through real-time sensing and decision intelligence. Conventional inventory models, built on stochastic forecasts, ignore reverse logistics, carbon constraints, and energy structures, rendering them unsuitable for modern circular manufacturing. This study proposes a novel closed-loop deterministic inventory optimization model driven by IoT demand sensing, which captures real-time consumption data to eliminate forecast uncertainty and enable dynamic zero-stockout planning. The model jointly optimizes forward production and reverse remanufacturing flows, incorporating deterministic return-lag and recovery yield, renewable-grid energy allocation, carbon emissions pricing, ramping stability, and waste disposal costs. A linear trust-band mechanism preserves robustness against sensor noise by penalizing deviations from IoT measurements. The resulting scalable linear program supports real-time rolling-horizon execution. This work generalizes classical EOQ/EPQ and closed-loop formulations into a unified deterministic framework embedding sustainability, energy decarbonization, and digital supply-chain intelligence, offering actionable insights for smart, circular, low-carbon factory operations.