Short-term robustness of production management systems: new methodology

This paper investigates the short-term robustness of production planning and control systems. This robustness is defined here as the system’s ability to maintain short-term service probabilities (i.e., the probability that the fill rate remains within a prespecified range), in a variety of environments (scenarios). For this investigation, the paper introduces a heuristic, stagewise methodology that combines the techniques of discrete-event simulation, heuristic optimization, risk or uncertainty analysis, and bootstrapping. This methodology compares production control systems, subject to a short-term fill-rate constraint while minimizing long-term work-in-process (WIP). This provides a new tool for performance analysis in operations management. The methodology is illustrated via the example of a production line with four stations and a single product; it compares Kanban, Conwip, Hybrid, and Generic production control schemes.