Research - Vinay Singh | AI & ERP Publications

IEEE Published Research

IEEE PUBLICATION R² 99.24% Accuracy

Artificial Intelligence-Driven Forecasting Models for Demand Forecasting in Inventory Optimization

Groundbreaking research designing and validating ANN and hybrid LSTM–GRU models for demand forecasting, achieving near-perfect accuracy (R² 99.24%) on real-world retail datasets. This work demonstrated the effectiveness of deep learning in modeling non-linear, time-dependent business behavior and laid the foundation for enterprise financial automation and predictive analytics.

Key Contributions:

Hybrid LSTM-GRU architecture for time-series forecasting
Validation on real-world retail demand datasets
Foundation for AI-driven ERP cash flow forecasting models

Deep Learning LSTM-GRU Inventory Optimization Time-Series Forecasting

IEEE RELATED Extended Framework

Deep Learning Models for Time-Series Forecasting in Enterprise ERP Systems

Extended the IEEE-published LSTM-GRU framework to enterprise ERP systems, developing an AI-driven cash flow forecasting model integrated with Oracle Fusion Cloud and EBS. Combined real-time ERP transactions with external economic indicators, enabling organizations to move beyond static spreadsheets toward adaptive, self-learning financial forecasts.

Oracle Fusion Cash Flow Forecasting ERP Integration

Major Research Publications

30-40% Improvement

AI-Driven Cash Flow Forecasting in ERP Systems: Integrating Economic Indicators and Real-Time Transaction Data Using LSTM

Robust LSTM-based time-series forecasting model integrating real-time Oracle ERP transaction data with external economic indicators. Deployed in Oracle Fusion Cloud and EBS treasury modules, improving cash forecast accuracy by 30-40% and supporting real-time treasury dashboards.

LSTM Oracle Treasury Cash Flow

35% Forecast Improvement

AI-Driven ERP Evolution: Enhancing Supply Chain Resilience with Neural Networks and Predictive LSTM Models

Deep learning framework embedding predictive LSTM models into Oracle Fusion Cloud ERP systems. Piloted at McGraw Hill and Cisco, improved forecast accuracy by 35%, reduced carrying costs by 15%, and increased responsiveness to supply disruptions by 40%.

Supply Chain McGraw Hill Cisco

15-20% Cost Reduction

AI and ERP Integration for Adaptive Dynamic Costing Based on Consumer Demand Fluctuations in Manufacturing

Real-time AI model combining LSTM and gradient boosting for dynamic costing and pricing in Oracle ERP. Integrated with Oracle Fusion Cost Management and Manufacturing Cloud, achieved 15-20% reduction in costing errors and enabled adaptive pricing tied to demand patterns.

Manufacturing Gradient Boosting Dynamic Costing

Best Practices

Lessons Learned from Large-Scale Oracle Fusion Cloud Data Migrations

Critical insights from high-impact Oracle Fusion Cloud implementations at McGraw Hill and GE Capital. Covers multi-terabyte migrations from Oracle EBS, migration strategy patterns, data cleansing best practices, automation scripts, and real-time reconciliation dashboards.

Oracle EBS Cloud Migration GE Capital

Peer Review & Editorial Roles

Contributing to Academic Excellence

Springer Journal

International Journal of System Assurance Engineering and Management

Q2 Ranked 'A' by ABCD

Asian Journal of Research

Computer Science

Asian Journal

Economics, Business and Accounting

Asian Research Journal

Mathematics

Research Impact

Real-World Results from Published Research

99.24%

Forecast Accuracy (R²)

IEEE LSTM-GRU Model

40%

Better Disruption Response

Supply Chain Resilience

35%

Forecast Improvement

McGraw Hill & Cisco

20%

Cost Reduction

Dynamic Costing Model

Research & Publications