Machine Learning-based Predictive Analytics in ERP Systems

Abstract

This thesis studies applying machine learning-based predictive analysis in ERP systems. A literature review is conducted to see what modules and features have had these integrations in ERP systems before, as well as what kinds of advantages they could bring for a company. In the case study of the thesis, a machine learning model is trained for the project management module for predicting project budgets. The thesis shows that machine learning-based predictive analysis is most used in the inventory management module and that it can provide a lot of benefits, mainly by optimizing processes. The trained machine learning model did not achieve an accuracy high enough for an integration to the system but showed potential in ML-based project budgeting in ERPs. Further efforts should be directed to increasing the size of the dataset.