Knowledge Acquisition from Assembly Operational Data Using Principal Components Analysis and Cluster Analysis

Difficulties identified in the use of performance measures to monitor and control manufacturing or assembly system behaviour justify the evaluation method presented in this paper. The ability to improve the effectiveness of both evaluation and decision-making tasks, based on the knowledge acquisition from operational data, is an important requirement for an effective planning and systems operation. The proposed evaluation method is based on the use of multivariate techniques-Principal Components Analysis and Cluster Analysis-and will prevent the loss of information in each available set of measures and will promote effective use of data in the analysis of system performance.