CONTRIBUTION TO MODELING AND IMPROVING QUALITY CONTROL OF FINISHED PRODUCTS IN PRODUCTION SYSTEMS BY USING BAYESIAN NETWORKS AND LEAN SIX SIGMA

Industrial production systems in the Sahelian region of sub-Saharan Africa and the Central African sub-region face numerous challenges, including the lack of control over customer satisfaction levels and the instability and variability of operational quality control processes. This often leads to consumer dissatisfaction and an insufficient product conformity rate. To address this issue, we propose a methodology aimed at reducing variability and improving the operational quality control process of industrial production systems. This methodology combines the use of Lean Six Sigma (LSS) tools, Bayesian Networks (BNs), and multilinear regression analysis. Our combined approach consists of six stages. To implement this combined approach, we selected a tissue production system from the SITRACEL industrial company based in Cameroon. This implementation revealed an insufficient conformity rate of 3.727o, customer dissatisfaction of 16.25% compared to benchmarks, dominant quality defect causes directly related to the machine, and modeled quality control indicators to track variability in scrap and waste rates.