X-bar charts or pre-control: an economic evaluation using alternative cost models

Abstract

The on-line monitoring techniques of pre-control and X-bar control charts are compared, using two cost models that differ in the treatment of process variation cost. The first cost model assumes that process variation results in increased costs only when out-of-tolerance parts are made, while the second cost model uses Taguchi's loss function to quantify the cost of process variation. An example shows that the least-cost alternative depends on the cost model used. The sensitivity of the two techniques to changes in a number of process parameters is analysed for the loss function-based model. The results indicate that the most important consideration in choosing the process monitoring procedure is the assumption regarding the cost of process variation.