A SINGLE PERIOD INVENTORY MODEL WITH CAPACITATED PRODUCTION UNDER UNCERTAINTY OF BUDGET AND STOCKOUT COST

Nowadays all business scenarios are in a competitive area, while dealing with single-period products, the basic and major problem is to manage its inventory in the best way. In traditional single-period inventory models, a business strategy is described where only a single procurement is made for a specific period under probabilistic demand. But through the developing inventory models, different aspects should be considered such as forecasting the stockout level, budget, the suitable production capacity as well as demand in order to optimize the profit. This paper investigates a single period multi-product inventory model in which products are stochastically defective and demands are random and continuous and the model constrains the production capacity, budget and stockout cost. Initials, the expected profit in general form in terms of density functions of the demand and the percentage of defective is represented; then, the required descriptions for these functions are given. Considering the uncertainty on budget and stockout cost, these two constraints are presented in stochastic and fuzzy forms. Though, the production capacity is deterministic in every constraint. Therefore, these constraints are of three types: 1) budget and stockout cost constraints are both stochastic. 2) One of the budget and stockout cost constraints is stochastic and the other is fuzzy. 3) Both budget and stockout cost constraints are fuzzy. It should be noted that the possibility and necessity representations of fuzzy constraints are considered. By using chance constrained programming and possibility/necessity programming techniques, stochastic and fuzzy constraints are respectively determined. The problem is solved using SQP algorithm utilizing MATLAB. In addition, an example has been provided to clarify the model completely and at last sensitivity analysis is performed on the profit function with respect to various parameters in order to figure out the profit dependency of each parameter that is so useful in implementing the real world models.