This dissertation was written as a part of the MSc in Data Science at the International Hellenic
University. The aim of the thesis, is to present ways that machine learning methods can aid in
supply chain management, specifically in lead time prediction and reduction. Since lead time has
been acknowledged as a key factor in supply chain planning and in customer’s satisfaction as
well, it has been seen fit to research this topic in depth. These methods were applied to a big multi
product and multi-stage aluminium manufacturing group of companies headquartered in Greece.
In detail two predictive models were examined to predict the lead time of the company’s major
product groups, architectural aluminium profiles and accessories, and one model for the demand
forecasting of aluminium accessories to prevent stock outs which heavily affect customer’s orders
lead time. The results of this case study were more than satisfactory, having outperformed the
performance of existing systems concerning lead time prediction and demand forecasting
Collections
Show Collections
