A Study of Periodic Review Inventory Systems with Record Inaccuracy

• Main Authors: Yu-Jia Chen, 陳玉加
• Other Authors: Shu-Lu Hsu
• Format: Others
• Language: zh-TW
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/79434389855714595836

碩士 === 淡江大學 === 資訊管理學系 === 91 === Inaccurate inventory arises when the stock record is not in agreement with the physical stock. Such discrepancies may be incurred due to time lags between flows of information and materials, pilferage, deterioration or incorrect unit of issues. The impact of these discrepancies is that the system fails to order at right time with right quantity, leading to inappropriate inventory level and increasing operations cost. However, there is little work investigating the effects of record variance on the performance of inventory system. This thesis is to study periodic review inventory systems involving errors in recorded stock level. Three models are presented to investigate the relations between the record variance and the decisions of inventory control under different patterns of variances. The first model is extended the period review model with partial backorders with a constant error rate in recorded stock level. The second one is proposed for the case that the error rate increases with the cycle-count period exponentially. The third model is developed for the case that the error rate is a Beta-distributed random variable. For each model, an efficient algorithm is presented and applied to find the optimal solutions of the cycle-count period and the target level of replenishment. Numerical experiments and sensitivity analysis are also performed for each model to provide further implications of the effects of parameters on decisions and performance of system.