Title: Solving on-shelf availability by a creative inference of data science imputation models

Authors: Ashok Mahapatra; Srikanta Patnaik; Manoranjan Dash; Ananya Mahapatra

Addresses: Institute of Business and Computer Studies, SOA University, Bhubaneswar, Odisha, India ' Institute of Technical Education and Research, SOA University, Bhubaneswar, Odisha, India ' Institute of Business and Computer Studies, SOA University, Bhubaneswar, Odisha, India ' Faculty of Science, University of British Columbia, Vancouver, British Columbia, Canada

Abstract: Retailers' foremost objective is to ensure right product at right place at right price for the right customer, which predominantly hinges on its competencies to maximise On-Shelf-Availability (OSA). Although many studies are undertaken to identify and resolve root causes for diminished OSA, they always have centred around supply side inventory and replenishment. In this article we ventured out to explore OSA from a sell side perspective and discovered unidentified diminished OSA occurrence. Accordingly, systematic approach to identify and then estimate the value of diminished sales due to diminished OSA is proposed by means of modern data science mechanisms. Furthermore, a new nomenclature 'invisible missing data' is proposed by drawing a parallel to (1) Out-Of-Stock (OOS) as missing data and (2) estimation of OOS impact on sales as a missing value imputation problem. Finally, the solution made lucid enough to cater to a wider audience and encourage further future refinements in this field.

Keywords: OOS; OSA; out of stock; missing sales; missing value imputation; on shelf availability.

DOI: 10.1504/IJGUC.2022.125150

International Journal of Grid and Utility Computing, 2022 Vol.13 No.4, pp.425 - 446

Received: 07 May 2021
Accepted: 15 Jul 2021
Published online: 31 Aug 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article