Software:
Links:
|
| Analysis of perishable-inventory systems with censored data |
|
|
| Abstract: |
We consider a multi-period inventory system of a
perishable product with unobservable lost sales. Demand distribution parameters
are unknown and are updated periodically using the Bayesian approach based on
the censored sales data from the previous periods. Using a change of measure
technique, we develop an explicit expression of the first order condition of the
optimality equation. The expression clearly demonstrates the tradeoff between
the inventory cost incurred in the current period and the future benefit of
improved demand information accuracy. It shows that the myopic solution is a
lower bound on the optimal inventory level. It also enables us to quantify the
expected marginal value of information. | |
| Keywords: |
Inventory control, lost sales, censored demand
data, Bayesian update, dynamic programming, value of information | | |
| |
|
|
|
|
