Leveraging Knowledge Management to Drive Innovation and Technological Competence: The Moderating Role of Big Data Analytics

Abstract

The need for strong enterprise resource planning (ERP) software which helps organizations make the right decisions based on real-time data, has become more important as global markets have become more complex and supply chain ecosystems have become more volatile. Yet even the best ERP systems today struggle with agility and advanced predictive analysis. In this paper, we propose a new architecture for AI-enabled data-driven enterprise resource planning (ERP) systems to create efficient systems for marketing operations and resilient supply chain networks. The framework proposal consists of four vertical layers such as data ingestion layer, AI analytics layer (with machine learning, deep learning and natural language processing), ERP integration layer using service-oriented interfaces and business intelligence layer for output as suggestions so that it could promote better decision making. AI contracts (AI modules embedded in the system) allow for advanced capabilities such as customer segmentation based on behavior optimizing value-based marketing campaigns based on sentiment, mix prediction, time-series demand forecasting, inventory optimization and real-time anomaly detection. We implemented and validated the framework through three industrial case studies in the retail, electronics and logistics sectors. The experiments proved a great performance enhancement with 27.05% better forecast accuracy, 19% lower supply chain disruption and 32.15% higher efficiency of marketing campaigns against ordinary ERP deploys. It also talks about the integration challenges such as data quality, model interpretability and compatibility with legacy system. Results bolster believe for potential efficiency of AI-driven ERP to support strategic planning and flexibility of operations. We will analyze the use of explainable AI in combination of federation learning and blockchain-based technology for secure, decentralized ERPs in our future work.