MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202641085869 A) filed by Dr. Juhi Mehta; Mr. Prateek Thodkar P; Ms. Shruthi Ramesh; Ms. Ayesha Samreen; Ms. Minu Subramanya; Ms. Nompi Raj; Dr. Megha Garud; and Dr. Vidya Chandrasekar on July 13, 2026, for Ai-Driven Multilingual Digital Twin Framework For Predictive Asset Management And Optimisation Of Smart Industrial Ecosystems.
Inventors include Dr. Juhi Mehta; Mr. Prateek Thodkar P; Ms. Shruthi Ramesh; Ms. Ayesha Samreen; Ms. Minu Subramanya; Ms. Nompi Raj; Dr. Megha Garud; and Dr. Vidya Chandrasekar.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: An AI-driven multilingual digital twin framework is disclosed for predictive asset management and optimisation of smart industrial ecosystems. The framework comprises a data acquisition layer for ingesting structured sensor data and unstructured multilingual textual, audio, and visual data from industrial assets; a multilingual natural language processing engine that identifies, translates, and semantically normalises operational text and speech across multiple languages while extracting structured fault entities; a digital twin core engine that fuses physics-informed models with machine learning surrogate models to represent each asset's real-time and projected state; a predictive analytics engine generating remaining useful life estimates and anomaly scores; and a multi-objective optimisation engine that coordinates maintenance, operational, and resource allocation recommendations across interdependent assets. A hybrid edge-cloud orchestration layer allocates computation based on latency and complexity, while an explainable artificial intelligence subsystem renders natural language explanations in the operator's native language. The framework addresses multilingual data fragmentation, cross-facility generalisation, and multimodal fusion limitations of prior art systems, improving asset reliability, efficiency, and lifecycle performance across multinational industrial ecosystems.
Disclaimer: Curated by HT Syndication.
