An Analysis of Machine Learning-Based Decision Support Systems for Enterprise Resource Planning

The ERP systems have become part and parcel of effective operations of organizations through integration of business processes in different parts of the organizations like finance, supply chain and human resource. However, rising due to the complexity of the business, the long-established ERP systems have limited ability to help in terms of offering actionable insights, which make it possible to utilize in strategic decision-making. Machine learning (ML) is an explosive technology that has changed ERP systems and improved their performance to assist in decision-making. The paper under review gives a detailed overview of the role of machine learning methods in the ERP systems and their impact on making strategic decisions with specific reference to the integration of machine learning methods in the ERP systems. This paper discusses most used ML algorithms: classification, regression, clustering and reinforcement learning, and their use in the optimization of ERP including demand forecasting, invention management, financial planning, and customer relationship management (CRM). By using ML in ERP systems, it is possible to be able to ensure that the predictability can be done, in the context of the point of anomaly discovery, along with the possibility of providing real-time support of decision-making, which can result in the operational efficiency of enhanced operations as well as the realization of cost savings, and even improved eventual resource allocation. Also discussed in the review is the difficulties associated with the utilization of ML-based decision support systems in ERP such as quality of data, compatibility between systems, and experts. Lastly, the paper discusses future trends and how the capabilities of deep learning and AI-based ERP system will affect the businesses willing to use advanced systems to make better decisions.