Pricing decision in a newsvendor model with partial backorders under normal probability distribution for the demand.

This paper presents a newsvendor model with backorders for customers who are willing to wait to be served. Demand follows a normal probability distribution, with the particularity that the expected value depends on the sale price and the variation coefficient is fixed. Three parameters are considered to characterize this dependence on the expected demand and the sale price: the population size of potential customers, the unit production cost of the item and an elasticity parameter with an isoelastic type. Backorders and lost sales are combined with a fixed proportion for the backorders. The quantities to be determined are the sale price and the order quantity. The goal is the maximization of the expected profit. The optimal solution is obtained in a closed form if there are no lost sales. In the case of a mixture of partial backordering and lost sales, a methodological proposal based on a numerical algorithm is given. The study reveals that the maximum expected profit and the optimal quantities to be determined are highly influenced by the unit purchasing cost and the degree of dependence of the demand concerning the sale price. Other parameters, such as the proportion of backorders and the variation coefficient of demand, are less influential. Numerical examples are used to illustrate the model, and a sensitivity analysis of the optimal solution regarding the nine initial parameters is presented. Some managerial insights deduced from the obtained results are also proposed. • A newsboy model with demand fixed partial backlogged is presented. • Demand follows a normal probability distribution with price-dependent mean. • The decision variables are the lot size and the selling price. • A numerical algorithm is given to obtain the optimal policy. [ABSTRACT FROM AUTHOR]