Estimation of Fractiles for Order Point Determination.
Abstract
Inventory stocking problems with stochastic demand typically involve an estimate of the location of some fractile of the demand distribution, where the fractile is usually in the 0.8 - .99 range. This fractile is termed the 'service level' and is the probability that demands will be satisfied from stock on hand. The conventional approach of using the Normal model to estimate this location can sometimes be misleading since it uses information about the center of the distribution to predict behavior in the tail of the distribution. A new method based on a mixture model is proposed for directly estimating the location of the appropriate fractile. A formal Bayesian approach is presented and heuristic smoothing methods are developed. The statistics to be collected can be interpreted as the proportion of backorders and the average quantity backordered in each period. The estimate obtained is biased in general but robust in the sense that it works well for a variety of distributions. The performance of this method is superior to the conventional approach with some qualifications.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1975
- Accession Number
- ADA015684
Entities
People
- Uday S. Karmarkar
Organizations
- Massachusetts Institute of Technology