%0 Journal Article
%T An Improved Enterprise Resource Planning System Using Machine Learning Techniques
%A Ahmed Youssri Zakaria
%A Elsayed Abdelbadea
%A Atef Raslan
%A Tarek Ali
%A Mervat Gheith
%A Al-Sayed Khater
%A Essam A. Amin
%J Journal of Software Engineering and Applications
%P 203-213
%@ 1945-3124
%D 2024
%I Scientific Research Publishing
%R 10.4236/jsea.2024.175011
%X Traditional Enterprise Resource Planning (ERP) systems with relational databases take weeks to deliver predictable insights instantly. The most accurate information is provided to companies to make the best decisions through advanced analytics that examine the past and the future and capture information about the present. Integrating machine learning (ML) into financial ERP systems offers several benefits, including increased accuracy, efficiency, and cost savings. Also, ERP systems are crucial in overseeing different aspects of Human Capital Management (HCM) in organizations. The performance of the staff draws the interest of the management. In particular, to guarantee that the proper employees are assigned to the convenient task at the suitable moment, train and qualify them, and build evaluation systems to follow up their performance and an attempt to maintain the potential talents of workers. Also, predicting employee salaries correctly is necessary for the efficient distribution of resources, retaining talent, and ensuring the success of the organization as a whole. Conventional ERP system salary forecasting methods typically use static reports that only show the system’s current state, without analyzing employee data or providing recommendations. We designed and enforced a prototype to define to apply ML algorithms on Oracle EBS data to enhance employee evaluation using real-time data directly from the ERP system. Based on measurements of accuracy, the Random Forest algorithm enhanced the performance of this system. This model offers an accuracy of 90% on the balanced dataset.
%K ERP
%K HCM
%K Machine Learning
%K Employee Performance
%K Pythonista
%K Pythoneer
%U http://www.scirp.org/journal/PaperInformation.aspx?PaperID=133208