Inventory Policy with Parametric Demand: Operational Statistics, Linear Correction, and Regression

In this paper, we consider data-driven approaches to the problem of inventory control. We first consider the approach of operational statistics and review related results which enable us to maximize a priori expected profit uniformly over all parameter values, when the demand distribution is known up to the location and scale parameters. For the case of the unknown shape parameter, we first suggest a heuristic approach based on operational statistics to obtain improved ordering policies and illustrate the same for the case of a Pareto demand distribution. In more general cases where the heuristic is not applicable, we suggest linear correction and support vector regression approaches to better estimate ordering policies, and illustrate these using a Gamma demand distribution. In certain cases, our proposed approaches are found to yield significant improvements.

[1]  H. Scarf.  Bayes Solutions of the Statistical Inventory Problem , 1959 .

[2]  Min-Chiang Wang,et al.  Expected Value of Distribution Information for the Newsvendor Problem , 2006, Oper. Res..

[3]  Robert H. Hayes,et al.  Statistical Estimation Problems in Inventory Control , 1969 .

[4]  Evan L. Porteus,et al.  Stalking Information: Bayesian Inventory Management with Unobserved Lost Sales , 1999 .

[5]  Dimitris Bertsimas,et al.  A Robust Optimization Approach to Inventory Theory , 2006, Oper. Res..

[6]  Warrren B Powell,et al.  An Adaptive, Distribution-Free Algorithm for the Newsvendor Problem with Censored Demands, with Applications to Inventory and Distribution , 2001 .

[7]  L.W.G. Strijbosch,et al.  Assessing the effects of using demand parameters estimates in inventory control and improving the performance using a correction function , 2009 .

[8]  Woonghee Tim Huh,et al.  A Nonparametric Asymptotic Analysis of Inventory Planning with Censored Demand , 2009, Math. Oper. Res..

[9]  Min-Chiang Wang,et al.  Mean-range based distribution-free procedures to minimize "overage" and "underage" costs , 2007, Eur. J. Oper. Res..

[10]  L. Wasserman,et al.  The Selection of Prior Distributions by Formal Rules , 1996 .

[11]  K. Arrow,et al.  Optimal Inventory Policy. , 1951 .

[12]  Katy S. Azoury.  Bayes Solution to Dynamic Inventory Models Under Unknown Demand Distribution , 1985 .

[13]  J. Kiefer,et al.  The Inventory Problem: I. Case of Known Distributions of Demand , 1952 .

[14]  J. George Shanthikumar,et al.  Solving operational statistics via a Bayesian analysis , 2008, Oper. Res. Lett..

[15]  A. F. Veinott.  Optimal Policy for a Multi-product, Dynamic Non-Stationary Inventory Problem , 1965 .

[16]  J. Kiefer,et al.  The Inventory Problem: II. Case of Unknown Distributions of Demand , 1952 .

[17]  David B. Shmoys,et al.  Provably Near-Optimal Sampling-Based Policies for Stochastic Inventory Control Models , 2007, Math. Oper. Res..

[18]  Martin L. Puterman,et al.  The Censored Newsvendor and the Optimal Acquisition of Information , 2002, Oper. Res..

[19]  S. A. Conrad,et al.  Sales Data and the Estimation of Demand , 1976 .

[20]  David J. Braden,et al.  Informational dynamics of censored observations , 1991 .

[21]  D. Iglehart.  The Dynamic Inventory Problem with Unknown Demand Distribution , 1964 .

[22]  G. Gallego,et al.  The Distribution Free Newsboy Problem: Review and Extensions , 1993 .

[23]  Bernhard Schölkopf,et al.  New Support Vector Algorithms , 2000, Neural Computation.

[24]  S. Karlin.  Dynamic Inventory Policy with Varying Stochastic Demands , 1960 .

[25]  J. George Shanthikumar,et al.  A practical inventory control policy using operational statistics , 2005, Oper. Res. Lett..