An integrated model for statistical and vision monitoring in manufacturing transitions

Manufacturing transitions have been increasing due to higher pressures for product variety. One dimension of this variety is color. A major quality control challenge is to regulate the color by capturing data on color in real-time during the operation and to use it to assess the opportunities for good parts. Control charting, when applied to a stable state process, is an effective monitoring tool to continuously check for process shifts or upsets. However, the presence of transition events can impede the normal performance of a traditional control chart. In this paper, we present an integrated model for statistical and vision monitoring using a tracking signal to determine the start of the transition and a confirmation signal to ensure that any process oscillation has concluded. We also developed an automated color analysis and forecasting system (ACAFS) that we can adjust and calibrate to implement this methodology in different production processes. We use a color transition process in plastic extrusion to illustrate a transition event and demonstrate our proposed methodology. Copyright (C) 2003 John Wiley Sons, Ltd.