Poznámka o úloze prodavače novin s oceňováním

A Note on the Newsvendor Problem with Pricing

conference paper

en

The main purpose of the paper is to present a specific case of dynamic pricing for the newsvendor problem. Firstly, a short overview of the newsvendor problem is given together with references to selected literature and remarks to its applicability. Then, dynamic pricing principles are discussed together with references to a decision dependent randomness case in stochastic programming. The dynamic pricing problem deals with determination of selling prices over time for a product whose demand is random and whose supply is fixed. We approach this problem by formulating the newsvendor problem, which is introduced as a single period problem in our case. We focus on specific features of the demand function assuming a decision dependent uniform distribution. We assume that its support size linearly decreases with the increase of the price. Under such assumptions, the model has suitable computational features related to the expectation of the objective function. In addition, a possibility to increase the profit by change of the price may appear. The model formulation allows us to use the MAPLE software for symbolic computations and visualization of results. The test results for the selected data set are visualized at the end of the paper.

Hlavní účel tohoto článku je uvést specifický případ dynamického oceňování pro úlohu prodavače novin. Nejdříve je uveden stručný přehled literatury pojednávající o teto úloze a jejích aplikacích. Dále jsou popsány principy dynamického oceňování spolu s odkazy na literaturu zaměřenou na nezávislé náhodné rozhodovací případy ve stochastickém programování. Úloha dynamického oceňování vede na určení prodejní ceny pro produkt, jehož poptávka je náhodná a dodávka je fixní. K této úloze přistupujeme pomocí úlohy prodavače novin, která je uvedena jako úloha s jedním časovým intervalem. Dále náhodnou poptávku názorně popisujeme pomocí rovnoměrného rozdělení. Předpokládáme, že s rostoucí cenou klesá poptávka a naopak. Modely jsou doplněny výpočty a vyzualizací za pomoci softwaru MAPLE.

The main purpose of the paper is to present a specific case of dynamic pricing for the newsvendor problem. Firstly, a short overview of the newsvendor problem is given together with references to selected literature and remarks to its applicability. Then, dynamic pricing principles are discussed together with references to a decision dependent randomness case in stochastic programming. The dynamic pricing problem deals with determination of selling prices over time for a product whose demand is random and whose supply is fixed. We approach this problem by formulating the newsvendor problem, which is introduced as a single period problem in our case. We focus on specific features of the demand function assuming a decision dependent uniform distribution. We assume that its support size linearly decreases with the increase of the price. Under such assumptions, the model has suitable computational features related to the expectation of the objective function. In addition, a possibility to increase the profit by change of the price may appear. The model formulation allows us to use the MAPLE software for symbolic computations and visualization of results. The test results for the selected data set are visualized at the end of the paper.

stochastické programování, úloha kolportéra novin, dynamické oceňování, MAPLE

stochastic programming, newsvendor problem, dynamic pricing, MAPLE

2012

27.06.2012

VUT

Brno

978-80-214-4540-6

1803-3814

18th International Conference of Soft Computing, MENDEL 2012 (id 19255)

2012

1

1

410–415

6