| |
"The Bayesian Newsvendors in Supply Chains with Unobserved Lost Sales"
David Glenn, Arnab Bisi, and Martin L. Puterman
|
| |
First Author :
|
|
Second Author :
|
Arnab Bisi
Commerce and Business Administration
University of British Columbia
2053 Main Mall
Vancouver, BC V6T1Z2
Canada
|
|
Third Author :
|
Martin L. Puterman
Commerce and Business Administration
University of British Columbia
2053 Main Mall
Vancouver, BA V6T1Z2
Canada
|
|
| |
| |
| Abstract : |
| |
| We consider two-echelon supply chains with one supplier and two retailers. Retailers are censored newsvendors facing general parametric demand distributions involving unknown parameters. Using a Bayesian MDP formulation, we investigate how the supplier can make use of the combined information gathered from the retailers’ sales data to increase channel profits. We compare among the following three scenarios: (i) supplier shares her (pooled) demand distribution updates with the retailers, (ii) supplier does not use pooled updates, and (iii) integrated firm. We show that buy-back contracts achieve channel coordination in the above system. |
|
| |
| |
| Keywords : |
| |
| Bayesian Markov decision processes, Censoring, Demand estimation, Inventory, Lost sales, Myopic policy, Supply chain |
| |
| |
| Manuscript Received :
2004
|
| Manuscript Published :
August 2, 2004
|
|
| |
| |
This abstract has been viewed 2281 times.
Click here to view the full text of this paper.
|
| |
| |
|
