A comparative study of production control mechanisms using simulation-based multi-objective optimisation

Abstract

There exist many studies conducted to compare the performance of different production control mechanisms (PCMs) in order to determine which one performs the best under different situations. Nonetheless, most of these studies suffer from the problems that the PCMs are not compared with their optimal parameter settings in a truly multi-objective context. This paper describes how different PCMs can be compared under their optimal settings through generating the Pareto-optimal frontiers, in form of optimal trade-off curves in the performance space, by applying evolutionary multi-objective optimization to simulation models. This concept is illustrated with a bi-objective comparative study of the four most popular PCMs in the literature, namely Push, Kanban, CONWIP and DBR, on an unbalanced serial flow line in which both control parameters and buffer capacities are to be optimized. Additionally, it introduces the use of normalized hypervolume as the quantitative metric and confidence-based significant dominance as the statistical analysis method to verify the differences of the PCMs in the performance space. While the results from this unbalanced flow line cannot be generalized, it indicates clearly that a PCM may be preferable in certain regions of the performance space, but not others, which supports the argument that PCM comparative studies have to be performed within a Paretobased multi-objective context.

Reference point based evolutionary multi-objective optimization with dynamic resampling for production systems improvement

“…In Figure 8, the attainment surface results for the optimistic RP ( z ¯ 1 , closer to the ideal point) have shown that the best performance can be achieved by the R-NSGA-II with DR2 as the DR mechanism, followed by D-OCBA-m and time-based DR. The higher throughput performance obtained with R-NSGA-II DR2 can be statistically verified using the confidence-based dominance post-optimality analysis, first introduced in Ng et al (2008) and later used for a systematic comparison of production control mechanisms in the objective space using NSGA-II (Ng et al , 2012). The basic idea of confidence-based dominance is to consider the statistical significance when determining whether an objective function of a solution, i.e.…”

“…In comparison to the minCT-maxTP, i.e. Minimize(Cycle Time) against Maximize(Throughput) analysis considered in (Ng et al , 2012), the analysis for the minImp-maxTP plots in the current paper is simpler because only the objective function value of system throughput is subjected to uncertainty from stochastic simulation. In other words, the Welch difference test used in (Ng et al , 2012) and other simulation literature is applied only to f 2 ( x ) as below: …”

An Enhanced Simulation-Based Multi-Objective Optimization Approach with Knowledge Discovery for Reconfigurable Manufacturing Systems

“…When dealing with two conflicting objectives, the NSGA-II is known to be one of the most effective MOEA, endowing a proper convergence and spread of solutions [22]. Three main factors drive the outstanding performance of NSGA-II: the fast non-dominated sorting approach that decreases the computational complexity; the elitism mechanism storing the non-dominated solutions; and the crowding distance calculation that ensures a diverse population by comparing and selecting solutions after the non-dominated sorting [63]. In this study, by incorporating simulation components into the fitness function evaluation of NSGA-II, a customized SMO-NSGA-II for RMS is developed, as described below.…”

Single-stage Kanban system with deterioration failures and condition-based preventive maintenance

“…Lavoie et al [4] compare three pull type control policies in homogenous transfer lines. The approach of simulation-based multi-objective optimization for comparing alternative production control policies is adopted in [5]. An integrated framework for Kanban-type production control systems is developed in [6].…”