Operational controllers are a critical component for smart factories and the realization of initiatives like Industry 4.0, where good reliable operational decision-making is critical to success. Contemporary manufacturing execution systems (MES) are proprietary closed systems. Thus, they are not an adequate foundation for building and testing the conceptual models necessary to develop the computational decision-making bridge from planning to execution that is required for smart manufacturing. Open, generic models and implementation examples are a fundamental requirement.
As defined in the ISA-95 standard (International Society of Automation 2021), operational control is the translation of production plans into the execution of production operations. ISA-95 is the current standard for manufacturing operations management controllers, and while it identifies the information requirements for the various functions in operational control, it does not address the design or implementation of operational controllers; in particular it does not address decision-making in operational control.
One of the most appealing promises of 'smart factories', Industry 4.0, and 'digital twins' is the ability to maintain a digital representation of the target system, update its state to reflect the actual state of the target system, and run 'what if' analyses of alternative planning and operational control decisions using the 'digital twin' technology to predict results in the physical system. For systems with 'smart' controllers, this promise simply cannot be met if those simulations cannot correctly reflect the operational control decisions with a high degree of fidelity. To do so, the logic for making those operational control decisions that is coded into those 'digital twins' must precisely reflect the logic of the operational controller in the physical system. In the past, neither the operations research community nor the simulation community has addressed this issue. This paper offers at least one approach to creating the promised 'digital twin'.
The paper provides a concrete example of an operational controller that conforms to ISA-95, embodies the necessary computational decision making to be considered 'smart' and has a plausible path to implementation. With a robotic cell consolidating totes for delivery in a logistics hub as the use case, researchers describe the design of the cell’s operational controller and an implementation approach used in an AnyLogic hybrid agent-discrete event simulation.