Supply Chain: Optimize the Production Cost Using Machine Learning

Abstract

Purpose: The objective of this study is to know how we use can machine learning by applying different algorithms on the supply chain to produce in the cheapest site in the world considering different parameters. Theoretical framework: the study has highlighted the iterative queries of digital revolution in the supply chain. The literary view in this article has illustrated the significant role of machine learning and dark data in reducing the production costs, enhancing delivery performance. Design/Methodology/Approach: The company concerned by the case study is a multinational company specializing in flooring and sports surfaces. It operates in 33 production sites, with 520 sites in more than 100 countries. One of the important factors underlying this complexity is the customer base that expects the product at the same cost all over the world, which forces the system that is currently not centralized to produce at a high cost in some countries. Findings: In this article, we use machine learning by applying different algorithms to unstructured data stored in company servers, where the feedback loop is implemented. The expected result is produced in the cheapest site in the world considering delivery costs. Research, Practical & Social implications: We suggest a future research to use all the remaining dark data saved during ordering on the supply chain and to reduce more the costs in the world . Originality/Value: This article provides insights into how dark data analytics can be used to reduce supply chain costs and offers recommendations for organizations looking to leverage dark data in their supply chain operations.