Introduction
Efficient production scheduling is crucial for maximizing resource utilization in eco-industrial parks (EIPs), where factories share energy and materials. Diverging from traditional practices, the park not only relies on external suppliers for raw materials but also fosters resource synergy among its internal factories. This includes the exchange of by-products among factories, promoting a circular economy and reducing external dependencies. Moreover, the park adopts Waste-to-Energy (WTE) technology, converting waste into valuable forms of energy such as heat, electricity, or transport fuels.
However, balancing energy distribution and job scheduling is challenging due to fluctuating power availability, production constraints, and inter-factory dependencies. Traditional scheduling methods often fail to optimize both energy use and production efficiency, leading to delays and resource waste.
This study presents a multimethod constraint programming (CP) model developed in AnyLogic to optimize production scheduling in an industrial park. The CP model generates optimal schedules, which are then validated in AnyLogic’s simulation environment to ensure feasibility and efficiency.
Simulation model
The proposed framework integrates production scheduling and energy allocation through constraint programming and AnyLogic-based simulation. The constraint programming (CP) model defines job start times, generator usage, and energy assignments while incorporating constraints on power availability, factory capacities, and shift schedules. The primary objectives of the model are to maximize energy utilization and schedule the highest number of jobs efficiently. The IBM ILOG CPLEX tool is used to solve the complex scheduling constraints.
The AnyLogic-based simulation presents factories as modular agents. These agents capture real-time material and energy flows. A centralized energy controller aligns production scheduling with available power supply and ensures smooth operations. A feedback loop allows continuous optimization by testing CP-generated schedules under realistic conditions.
The simulation uses discrete-event modeling to capture job execution timelines and agent-based modeling to simulate interactions between factories, power plants, and shared resources. It also includes optimization testing to refine scheduling strategies.
Read also: Explore additional resources on manufacturing simulation.
Results
The simulation provided real-world validation and identified capacity constraints and scheduling bottlenecks for further refinement. The multimethod CP-simulation approach significantly improved production scheduling efficiency:
- Energy utilization increased from 69% to 75%, and waste was reduced.
- Scheduled job completion improved from 89% to 93%, ensuring timely production.
- Optimized job sequencing reduced idle time, enhancing workflow.
This study demonstrates that integrating constraint programming with AnyLogic simulation enhances production scheduling in eco-industrial parks. The multimethod approach optimizes job sequencing, improves energy efficiency, and minimizes delays.
Experimental results showed that factories operating under the optimized scheduling framework completed 6% more jobs on average and reduced idle energy usage by 15% compared to traditional scheduling methods. Additionally, peak production delays were shortened by 20%, contributing to a more efficient and predictable workflow.
Future research will explore real-time adaptive scheduling and AI-driven forecasting to further enhance production scheduling in industrial environments.