An Adaptive Multi-Echelon Inventory Control Model for Nonstationary Demand Process

In this paper, we deal with an inventory model of a multi-stage, serial supply chain system where a single product type and nonstationary customer demand pattern are considered. The retailer and suppliers place their orders according to and echelon-stock based replenishment control policy. We assume that the suppliers can access online information on the demand history and use this information when making their replenishment decisions. Using a reinforcement learning technique, the inventory control parameters are designed to adaptively change as the customer demand pattern is altered, in order to maintain a given target service level. Through a simulation based experiment, we verified that our approach is good for maintaining the target service level.

[1]  Jing-Sheng Song,et al.  Optimal Policies for Multiechelon Inventory Problems with Markov-Modulated Demand , 2001, Oper. Res..

[2]  Andrew W. Moore,et al.  Reinforcement Learning: A Survey , 1996, J. Artif. Intell. Res..

[3]  Jing-Sheng Song,et al.  Inventory Control in a Fluctuating Demand Environment , 1993, Oper. Res..

[4]  김광재,et al.  품질기능전개에서의 변동성 분석 , 2004 .

[5]  Stephen C. Graves,et al.  Creating an Inventory Hedge for Markov-Modulated Poisson Demand: An Application and Model , 2001, Manuf. Serv. Oper. Manag..

[6]  Robert Goodell Brown,et al.  Smoothing, forecasting and prediction of discrete time series , 1964 .

[7]  Abhijit Gosavi,et al.  Global supply chain management: A reinforcement learning approach , 2002 .

[8]  Ralph Badinelli,et al.  A Model for Continuous-Review Pull Policies in Serial Inventory Systems , 1992, Oper. Res..

[9]  D. W. Trigg,et al.  Exponential Smoothing with an Adaptive Response Rate , 1967 .

[10]  Robin Roundy,et al.  A 98%-Effective Lot-Sizing Rule for a Multi-Product, Multi-Stage Production / Inventory System , 1986, Math. Oper. Res..

[11]  Herbert E. Scarf,et al.  Optimal Policies for a Multi-Echelon Inventory Problem , 1960, Manag. Sci..

[12]  R. Roundy.  98%-Effective Integer-Ratio Lot-Sizing for One-Warehouse Multi-Retailer Systems , 1985 .

[13]  이재영,et al.  XML을 이용한 건축정보관리 시스템 설계 , 2004 .

[14]  Fangruo Chen,et al.  On (R,NQ) Policies in Serial Inventory Systems , 1999 .

[15]  전진,et al.  Simulated Annealing Algorithm의 변형을 지원하기 위한 객체지향 프레임워크 설계 , 2004 .

[16]  Richard S. Sutton,et al.  Reinforcement Learning , 1992, Handbook of Machine Learning.

[17]  김정섭.  유연생산시스템에서 절삭공구 비용절감을 위한 가공시간과 팔렛배분의 최적화 , 1998 .

[18]  Derek R. Atkins,et al.  98%-Effective Lot Sizing for Series Inventory Systems with Backlogging , 1995, Oper. Res..

[19]  Hau L. Lee,et al.  Information distortion in a supply chain: the bullwhip effect , 1997 .