Optimization Versus Robustness in Simulation: A Practical Methodology, With a Production-Management Case-Study

摘要: Whereas Operations Research has always paid much attention to optimization, practitioners judge the robustness of 'optimum' solution be greater importance.Therefore this paper proposes a practical methodology that is stagewise combination following four proven techniques: (1) discrete-event simulation, (2) heuristic (3) risk or uncertainty analysis, and (4) bootstrapping.This illustrated through case study on production control systems.That defines as system's capability maintain short-term service measure, in variety environments (scenarios).More precisely, measure probability fill rate remaining within prespecified range.Besides satisfying probabilistic constraint, system should minimize long-term work-in-process.Actually, compares systems: Kanban, Conwip, Hybrid, Generic.These systems are studied for well-known example, namely line with stations single product.The conclusion Hybrid best when not ignored, but otherwise Generic best: considerations do make difference.