A Saddle Point of aD Inventory Problem
lunichi Nakagami, Masami Yasuda
Abstract
lunichi Nakagami, Masami Yasuda
Abstract
When the demand distribution is unknown, that is, most of the informations of it are insufficient for a definite probability distribution, a minimax procedure is sometimes used in an inventory problem. The minimax inventory problem is considered as a 2-person zero-sum game, in which one player (manager) decides his ordering level and other player (nature) chooses her distribution in the prescribed class of distributions. We shall show the necessary and sufficient conditions for the existeuce of saddle points in this problem, and give the explicit formula for a set of saddle points and a saddle value. A multi-period model is also discussed.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
When the demand distribution is unknown, that is, most of the informations of it are insufficient for a definite probability distribution, a minimax procedure is sometimes used in an inventory problem. The minimax inventory problem is considered as a 2-person zero-sum game, in which one player (manager) decides his ordering level and other player (nature) chooses her distribution in the prescribed class of distributions. We shall show the necessary and sufficient conditions for the existeuce of saddle points in this problem, and give the explicit formula for a set of saddle points and a saddle value. A multi-period model is also discussed.
Key concepts: Saddle point, Minimax, Saddle, Mathematics, Mathematical economics, Zero-sum game, Mathematical optimization, Distribution (mathematics)