A simulation‐expert‐system‐based approach for machine sizing of production systems

Abstract

PurposeThe purpose of this research is to focus on the solution of the resource sizing problem for production systems (PSs), defined as the specification of the number of each type of resources to be used in a production process for a given time period.Design/methodology/approachThe resource sizing problem is tackled by a simulation‐expert‐system‐based approach, coupling an expert system (ES) with a simulation tool. Hence, a number of “simulation – ES optimization” cycles are realized until obtaining non‐improvable levels of performance. The main performance measures considered in this work are related to the manufacturing orders due dates (DD).FindingsThrough the approach proposed in this work, it is possible to size machines in order to optimize DD related performance measures for a PS belonging to a specific application domain. PSs of this domain are characterized by a functional layout and feature no labor constraints. In addition, machines belonging to a same department are considered to be identical.Originality/valueThe developed approach allows studying the machine sizing problem realistically, through the use of stochastic simulation. Also, by coupling an ES to the simulation tool, it avoids the try and error aspect characterizing most simulation‐based approaches. It hence features a well‐structured reasoning mechanism for the search of the best solution.