A hybrid simulation technique to predict the effects of human deterioration and learning in an industrial environment

Abstract

Process simulation is an effective tool when used to simulate a system where a great deal of data exists for the process. This technique is however limited when it comes to simulating certain non-deterministic parts such as human behaviour and interaction, for which there may not be a great amount of data available. This work creates a unique hybrid model through the combination of process simulation with agent based simulation that simulates the non-deterministic parts of the process, as well as the deterministic parts. An actual industrial system forms the basis for the research, and the hybrid model is used to understand the effects that human deterioration has on the productivity of this system, as well exploring different scenarios that could lead to improved performance.