Closed-Loop Digital Twin for Energy-Efficient Scheduling in Food Manufacturing Systems

Abstract

Food manufacturing faces critical challenges in autonomous control due to fragmented digitalization, where simulation and optimization tools operate independently, resulting in suboptimal resource utilization and product quality risks. This paper presents a Hybrid Digital Twin Architecture (HDTA) integrating Discrete-Event Simulation (DES), Constraint Programming (CP-SAT), and Edge Computing into a closed-loop control system for bakery production scheduling. The three-layered HDTA architecture employs DES modeling for parameter generation, a CP-SAT solver for combinatorial schedule optimization with dual objectives (makespan and energy cost minimization), and an Edge Computing layer for real-time execution. Validated on a multi-stage bakery production line, the framework achieved a 24.4% reduction in total production time, 23.06% decrease in energy consumption, and zero quality violations compared to First-In-First-Out (FIFO) baselines. The results demonstrate that bidirectional integration of predictive simulation and prescriptive optimization enables autonomous, energy-efficient production control.

More by Shyrailym Adilkyzy

• Edge-deployed digital twin services for real-time industrial sensor orchestration: architecture and case study in smart manufacturingAsia Conference on Edge Intelligence and Service Computing · 2026
• AI-powered meeting transcription and summarization system based on Jitsi Meet, Jigasi, and VoskInternational Conference on Advanced Sensing and Intelligent Systems · 2026
• A Digital Twin Architecture for Integrating Lean Manufacturing with Industrial IoT and Predictive AnalyticsInformation · 2026 · Citations: 4
• Detection of Industrial Process Anomalies Using Isolation Forest and DBSCAN in a Real Manufacturing EnvironmentAsia Conference on Cognitive Engineering and Intelligent Interaction · 2025