Track: Inventory Management

Abstract

The purpose of this study is to investigate how artificial intelligence (AI) and machine learning (ML) methods, are being utilized in Lot Sizing and to anticipate future research directions in this area. The approach to conducting this study was through a systematic literature review mainly using the PRISMA framework. This study analyzes the publications available on Web of Science and Scopus associated with AI and ML to Lot sizing with a focus also on Optimization of order quantities.

Initial keyword search led to a download of 325 papers, which were further reduced to 36 using various filters and eligibility criteria. The included studies are then all listed in tabular format whilst the relevant findings and the main algorithms adopted were discussed. The research found that the use of AI and ML it is still in its infancy, thus providing vast opportunities different areas of lot sizing. While some authors reported significant improvements in systems performance, hence indicating their efficiency, some stated that system is not fully utilised, while some found that the implementation could significantly reduce the bullwhip effect.

Previous research on AI applications in lot sizing has been limited to specific areas such as Demand forecasting which is rather broad. This study offers a unique contribution by providing an in-depth review of AI and ML in Lot sizing.