Home Innovation SAP Adarsh Vaid Develops Hybrid Fr...

Adarsh Vaid Develops Hybrid Framework for SAP ERP Anomaly Detection

SAP

Business Honor
10 April, 2025

Vaid's hybrid model boosts ERP anomaly detection accuracy while cutting computational costs by 40%.

A SAP architect named Adarsh Vaid has created a hybrid framework for identifying irregularities in SAP ERP systems which is a major breakthrough for enterprise resource planning (ERP) systems.

Vaid oversaw a research team that developed a revolutionary method that included adaptive K-Means clustering, Self-Organizing Maps (SOM), and Gaussian Mixture Models (GMM). The platform achieves impressive anomaly detection accuracy rates up to 94%, addressing challenges in managing complex enterprise data.

Vaid states that corporate systems produce a lot of data that needs constant monitoring to make sure reliability and correctness. Their hybrid architecture addresses this by enabling efficient data processing with a high precision rate of 95.5% through the utilization of distributed computing on Spark clusters.

This invention comes at an important moment as companies face with the increasing difficulty of data in their ERP systems. Vaid's approach stands out for its real flexibility to shifting data patterns, which has generally been lacking from traditional clustering methods.

The International Journal of Computer Science and Mobile Computing published the study which demonstrates remarkable adaptability across a range of dataset sizes. Using short (less than 100,000 records), medium (100,000–1 million records) and big (more than 1 million records) datasets, the hybrid model demonstrated a 94% detection accuracy rate with 95.5% precision in demanding testing.

Most significantly, the framework reduced computational cost by 40% when compared to existing methods while maintaining performance when processing large-scale enterprise data. For businesses using this solution, the framework has demonstrated efficacy in supply chain operations (91% accuracy), inventory management (93% accuracy), and financial transaction anomaly detection (96% accuracy).

Inventions such as this type of hybrid framework will be important in preserving the quality and effectiveness of corporate systems as businesses continue their digital transformation.