Text

theses

The newsvendor problem has been widely studied since it first appeared in the literature at the end of the XIX century. It is still the subject of further research that addresses more complex and realistic situations based on previous work. The amount of research work done on this model and its applications is so vast that a simple search in Google Scholar under the keyword "newsvendor" will return almost 8,000 entries between 2010 and 2015. The problem, in its basic formulation, aims at finding an optimal replenishment policy of a perishable product in the face of uncertain, stochastic demand. Such a solution is selected in a way that maximizes the expected profit, which is calculated as the difference between the income and the purchase cost of the good in question. This thesis elaborates on the conditions needed to guarantee the existence of a unique maximum of the objective function in the price-setting newsvendor problem with price-dependent demand. This function is presented as a mean-variance trade-off between the expected profit and the variance of the profit, weighted with a risk parameter. The main goal of this thesis is to simplify any instance of the risk-sensitive newsvendor problem for the two most common price-dependent demand functions, namely, additive and multiplicative functions. When possible, we will provide sufficient conditions for the unimodality of the problem. Unlike many other results previously published, we aim at expressing such conditions by using a metric that captures managerial attention. To this end, we use the lost sales rate elasticity as a measure of the level of service provided by the seller and express these sufficient conditions as a function of this metric.

ETD_7534

Ph.D.

Includes bibliographical references

by Javier Rubio Herrero

doi:10.7282/T3GX4DVF

ETD doctoral

The author owns the copyright to this work.