A practical, distribution-free approach to the classical newsboy inventory stocking problem for highly right-skewed but otherwise unknown demand distributions

Sukanto Bhattacharya, Renato Alas Martins, Kuldeep Kumar*

*Corresponding author for this work

Research output: Contribution to journalArticleResearchpeer-review

1 Citation (Scopus)

Abstract

The reliable approximation of the demand distributions is critically important for most of the existing solution approaches to the classical newsboy inventory problem. This is rendered especially difficult in the absence of adequate historical sales information especially for newly established businesses. Such approximation, as previous research has shown, is also not entirely hazard-free even for businesses that have a sales history. So a closed-form, distribution-free approach definitely warrants exploration and has indeed been the subject of previous research. However, the distribution-free approaches that currently exist in the literature work optimally in idealized situations where demand distributions are approximately normal. But many real-life business situations; where the newsboy-type inventory model can be relevantly applied; face highly right-skewed demand distributions. We posit a very simple and practical solution approach for the classical newsboy-type inventory stocking problem that is designed for situations where the demand distributions are highly right-skewed but are otherwise totally unknown. To demonstrate its versatility, we give an illustrative sample of Poisson demand distributions which if assumed, yield optimal solutions that agree exactly with our posited approach.

Original languageEnglish
Pages (from-to)291-302
Number of pages12
JournalJournal of Interdisciplinary Mathematics
Volume14
Issue number3
DOIs
Publication statusPublished - Jun 2011

Fingerprint

Dive into the research topics of 'A practical, distribution-free approach to the classical newsboy inventory stocking problem for highly right-skewed but otherwise unknown demand distributions'. Together they form a unique fingerprint.

Cite this