A Two-Sided Sequential Screening Procedure

• Main Authors: Mao-Yi Cheng, 鄭茂益
• Other Authors: Shu-Fei Wu
• Format: Others
• Language: zh-TW
• Published: 2000
• Online Access: http://ndltd.ncl.edu.tw/handle/68286472567465373530

碩士 === 淡江大學 === 統計學系 === 88 === A screening procedure which selects items whose performance variable is within a two-sided specification based on observing the correlated screening variables according to the individual misclassification error(IME) is proposed in order to reduce the cost and time effort of inspection. This research is an extended study from the one-sized sequential screening procedure proposed by Wu(1989). The criterion of average outgoing quality(AOQ) and total inspection cost(TIC) are still used to search for the optimal allocation. The generalized formulas for the single stage(SSP), double stage(DSP) and triple stage(TSP) sequential screening procedures are derived by the use of multivariate normal distribution and matrix operation. An example allocating four screening variables into SSP, DSP and TSP is illustrated by the computer program via Fortran IMSL using the derived generalized formulas to search for the best allocation. We utilize Monte Carlo simulation method to simulate the desired probability values for any finite stage of sequential screening procedure, especially for more than three stages due to the complexity of obtaining exact results. The coefficients of all screening variables are chosen to get higher correct selection proportion after screening in this study. Comparing our results by the given coefficients with the results of Lu(1999) by their coefficients, we found that our results do reduce the TIC. At last, we do some sensitivity analysis to find out how different correlation structures affect the choice of optimal allocation.