A multi-attribute decision model for setting production planning parameters

Production and inventory-related decisions, which significantly influence each other and sometimes involve multiple attributes, trade-off assessment and uncertainties, serve a key role in the performance of make-to-stock (mts) manufacturing systems that are controlled by a constant work in process (conwip) order release policy. To benefit from established production planning methods, a crucial task in this context is to define suitable production parameter settings for a given planning horizon. To address this problem, we present a multi-attribute decision model to determine appropriate settings for the planning parameters, namely, cycle time, throughput rate, holding cost and stockout cost. The proposed model uses discrete event simulation to evaluate the performance of a conwip/mts manufacturing system in relation to the work in process and finished goods inventory. Analysis of variance (ANOVA) and a Kruskal-Wallis test are conducted to verify the significant effect on the analyzed parameters. The compromise solution that is recommended for the conwip/mts problem is obtained by considering a multi-attribute expected utility function that is representative of a decision maker’s preferences and risk attitude regarding the probability distribution of the simulation outputs. In contrast with preview studies on planning parameter setting, the result compensates the low performance of one of the attributes as a result of the high performance of another attribute, based on the axiomatic structure of MAUT.Based on the real data of a multi-product assembly line, a numerical application is employed to visualize the steps of this decision model and to demonstrate its usefulness in practical issues.