MULTIVARIATE CONTROL CHART WITH VARIABLE DIMENSIONS FOR FLEXIBLE PRODUCTION ENVIRONMENTS

Multivariate statistical control usually applied in industries that operate in continuous production processes. Thus, the objective of this work is to present a procedure for statistical process monitoring in flexible environments, which have a finite production horizon and p correlated observed variables. A systematic review of the literature on control charts conducted. Subsequently, a model proposition built and a simulation process performed; Thus, a method was conceived, which then validated through the Monte Carlo simulation. As an application of the proposed method, a case study was carried out in a metal-mechanic company in southern Brazil, which operates in the vertical transport segment. The results demonstrated the efficiency of the control chart in identifying and signaling out of control points, having a behavior similar to the expected behavior of its performance measure, ARL. The research contributes to the proposition of a multivariate control chart with variable dimensions for flexible production environments.