STOCHASTIC CELL LOADING, FAMILY AND JOB SEQUENCING IN A CELLULAR MANUFACTURING ENVIRONMENT

In this paper, a stochastic cell loading, family and job sequencing problem derived from a shoe manufacturing company is addressed. The problem is also observed in several other manufacturing systems such as jewelry, frozen food, medical device, blood sugar where jobs have individual due dates and uncertain processing times. Family splitting is allowed and family setup is applied where two consecutive jobs are not from the same family. Each job has probabilistic processing time and deterministic due date. A Stochastic Non-Linear Mathematical Model is developed and solved with Lingo software. The objective is to minimize the number of tardy jobs (nT) and total probability of tardiness. Jobs are classified as tardy, risky and early based on the obtained probability of tardiness. A job is called “tardy” if the probability of tardiness is 1 and “early” if it is 0. If the job is neither early nor tardy (the probability is between 0 and 1), it is classified as “risky”. Single and multi-cell configurations are experimented with and the results are compared with deterministic model (Suer & Mese, 2011). The proposed approach decreased the risky jobs in all configurations. The number of early jobs increased and the number of tardy jobs stayed the same in single cell configuration. In multi-cell configuration, less number of risky and greater number of early jobs are obtained with proposed approach.