A novel Iterative Optimization-based Simulation (IOS) framework: An effective tool to optimize system's performance

- A generic Iterative Optimization-based Simulation (IOS) framework is developed.
- Upon occurrence of trigger events, simulation model configuration is optimized.
- This framework could undertake multiple optimizers within the simulation run.
- In a case study, it is shown that the IOS outperforms the Simulation-Based Optimization (SBO) approach.

In this paper, a unique Iterative Optimization-based Simulation (IOS) framework is presented, which includes a threefold integration of simulation, optimization and database managers. With this IOS model, optimization occurs frequently at the operational level in order to optimize system variables during the simulation run. In other words, a trigger event momentarily pauses the simulation and signals the optimization manager to optimize the status of the system. Meanwhile, a snapshot of the system, which includes the system status parameters, is transferred to a database. The optimization manager uses this data as an input to the analytical modeling and solves the problem optimally by considering the pre-defined objectives, constraints and updated state of the system. The framework automatically re-formulates the mathematical model using updated parameters, and then, the optimization manager finds a solution in a reasonable amount of time. The database manager is then updated based on the current optimal solution and simulation continues from this point in time. This cycle is repeated for each trigger event in the system and lasts until simulation reaches its timespan. Deploying this promising IOS model enables practitioners to take advantage of long-term simulation runs of their system, while it has been optimized multiple times according to the occurrence of any predefined incidents. The performance of the proposed IOS framework is evaluated using a manufacturing case study, and its results are compared with Simulation-Based Optimization (SBO).

93