Abstract ▼ AI-Helper

When the nature of measuring a performance variable is destructive or very expensive, it is attractive to use a surrogate variable which is highly correlated with the performance variable and less expensive to measure. In this paper, we propose rectifying screening procedures using the performance a...

When the nature of measuring a performance variable is destructive or very expensive, it is attractive to use a surrogate variable which is highly correlated with the performance variable and less expensive to measure. In this paper, we propose rectifying screening procedures using the performance and surrogate variables. Two screening procedures are considered; a statistically-based screening procedure to reduce the current proportion of nonconforming items to a specified lower value after screening, and a economically-based screening procedure where the screening limit is determined so that the expected cost is minimized. It is assumed that the surrogate variable given the performance variable is normally distributed with known mean and standard deviation. For two screening procedures, methods of finding the optimal solutions are presented and numerical examples are also given.