Author:
Wa-Muzemba Anselm Tshibangu
Affiliation:
Morgan State University, Department of Industrial and Systems Engineering, 1701 E Cold Spring Lane, Baltimore, MD 21251 and U.S.A.
Keyword(s):
Robust Design, Optimization, DOE, FMS, Taguchi Loss Function, Simulation, Six Sigma.
Related
Ontology
Subjects/Areas/Topics:
Industrial Engineering
;
Informatics in Control, Automation and Robotics
;
Performance Evaluation and Optimization
;
Production Planning, Scheduling and Control
;
Systems Modeling and Simulation
Abstract:
This paper analyzes a flexible manufacturing system (FMS) and presents a new scheme to find the optimal operational parameters settings of two of the mostly used performance measures in assessing manufacturing and production systems, namely the throughput rate (TR) and the mean flow time (MFT). The scheme uses an off-line model that combines discrete-event simulation, robust design principles and mathematical analysis to uncover the optimal settings. The research suggests a two-level optimization procedure that uses an empirical process followed by an analytical technique. In a first level, the empirical approach serves to derive the near-optimal values of the two individual performance measures of interest. These values are then used as targets in the second level of the optimization procedure in which, a Taguchi quality loss function (QLF) is applied to the FMS mathematical model derived through simulation-meta-modeling to find the optimal parameter settings. As advocated in Six Si
gma Methodology the optimization of the modeled system is implemented and achieved through a minimization of the performance variation followed by an optimal adjustment of the performance’s mean if necessary, in order to minimize the overall loss incurred due to the deviation of the mean from target.
(More)