A Nonparametric Asymptotic Analysis of Inventory Planning with Censored Demand

We study stochastic inventory planning with lost sales and instantaneous replenishment where, contrary to the classical inventory theory, knowledge of the demand distribution is not available. Furthermore, we observe only the sales quantity in each period and lost sales are unobservable, that is, demand data are censored. The manager must make an ordering decision in each period based only on historical sales data. Excess inventory is either perishable or carried over to the next period. In this setting, we propose nonparametric adaptive policies that generate ordering decisions over time. We show that the T-period average expected cost of our policy differs from the benchmark newsvendor cost---the minimum expected cost that would have incurred if the manager had known the underlying demand distribution---by at most O(1/T0.5).

[1]  John A. Muckstadt,et al.  Inventory Control in Directed Networks: A Note on Linear Costs , 2004, Oper. Res..

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

[3]  Elad Hazan,et al.  Logarithmic regret algorithms for online convex optimization , 2006, Machine Learning.

[4]  David E. Fyffe,et al.  Estimation of Forecast Errors for Seasonal-Style-Goods Sales , 1971 .

[5]  William E. Jones,et al.  Proof of claims , 1997 .

[6]  W. Lovejoy.  Myopic policies for some inventory models with uncertain demand distributions , 1990 .

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

[8]  H. Scarf.  Some remarks on bayes solutions to the inventory problem , 1960 .

[9]  Adam Tauman Kalai,et al.  Online convex optimization in the bandit setting: gradient descent without a gradient , 2004, SODA '05.

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

[11]  S. Nahmias.  Demand estimation in lost sales inventory systems , 1994 .

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

[13]  R. L. Winkler,et al.  Learning, Experimentation, and the Optimal Output Decisions of a Competitive Firm , 1982 .

[14]  Robert D. Kleinberg.  Nearly Tight Bounds for the Continuum-Armed Bandit Problem , 2004, NIPS.

[15]  Evan L. Porteus.  Foundations of Stochastic Inventory Theory , 2002 .

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

[17]  Michael C. Fu,et al.  An Adaptive Sampling Algorithm for Solving Markov Decision Processes , 2005, Oper. Res..

[18]  Martin Zinkevich,et al.  Online Convex Programming and Generalized Infinitesimal Gradient Ascent , 2003, ICML.

[19]  David B. Shmoys,et al.  Provably near-optimal sampling-based algorithms for Stochastic inventory control models , 2006, STOC '06.

[20]  Herbert E. Scarf,et al.  Inventory Models of the Arrow-Harris-Marschak Type with Time Lag , 2005 .

[21]  James H. Bookbinder,et al.  Estimation of Inventory Re-Order Levels Using the Bootstrap Statistical Procedure , 1989 .

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

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

[24]  Paul H. Zipkin,et al.  Foundations of Inventory Management , 2000 .

[25]  Li Chen,et al.  Dynamic Inventory Management with Learning About the Demand Distribution and Substitution Probability , 2008, Manuf. Serv. Oper. Manag..

[26]  T. L. Lai Andherbertrobbins.  Asymptotically Efficient Adaptive Allocation Rules , 1985 .

[27]  R. Jagannathan,et al.  Technical Note - Minimax Procedure for a Class of Linear Programs under Uncertainty , 1977, Oper. Res..

[28]  Apostolos Burnetas,et al.  Adaptive Ordering and Pricing for Perishable Products , 2000, Oper. Res..

[29]  W. Hoeffding.  Probability Inequalities for sums of Bounded Random Variables , 1963 .

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

[31]  Warren B. Powell,et al.  Learning Algorithms for Separable Approximations of Discrete Stochastic Optimization Problems , 2004, Math. Oper. Res..

[32]  Georgia Perakis,et al.  Regret in the Newsvendor Model with Partial Information , 2008, Oper. Res..

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

[34]  문일경.  The Distribution Free Newsboy Problem : Review and Extensions , 1993 .

[35]  Peter Auer,et al.  Finite-time Analysis of the Multiarmed Bandit Problem , 2002, Machine Learning.

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

[37]  Jing-Sheng Song,et al.  Analysis of Perishable-Inventory Systems with Censored Demand Data , 2008, Oper. Res..

[38]  Xiangwen Lu,et al.  Dynamic Inventory Planning for Perishable Products with Censored Demand Data , 2005 .

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

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

[41]  E. Silver,et al.  A Bayesian Analysis of the Style Goods Inventory Problem , 1966 .

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

[43]  A. F. Veinott,et al.  Computing Optimal (s, S) Inventory Policies , 1965 .

[44]  Jing-Sheng Song,et al.  Inventory Control with Unobservable Lost Sales and Bayesian Updates , 2008 .