MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641049899 A) filed by Easwari Engineering College; and Srm Institute Of Science And Technology, Ramapuram Campus, Chennai, Tamil Nadu, on April 20, for 'ai powered concrete mix and strength prediction.'
Inventor(s) include Manikanda Ganapathi T; Alban Johnson; and Ilamurugan G.
The application for the patent was published on May 1, under issue no. 18/2026.
According to the abstract released by the Intellectual Property India: "Concrete mix design and strength prediction are essential yet challenging tasks in civil engineering due to material variability, nonlinear parameter interactions, and environmental influences. Traditional empirical methods often lack generalization across varying conditions, while standard 5 machine learning models require frequent retraining for new datasets. This study proposes an AIdriven framework combining parameter-efficient fine-tuning of large language models (LLMs) with regression-based techniques. A pretrained DialoGPT-medium model was adapted using LowRank Adaptation (LoRA) with 4-bit quantization to generate mix designs for target compressive strengths. Simultaneously, an Extra Trees Regressor was used to predict strength from mix 10 proportions. Results show strong performance, with the LLM achieving an R of 0.9245 and the regression model reaching 0.8979. Both models were deployed via Gradio interfaces, demonstrating practical applicability in construction informatics."
Disclaimer: Curated by HT Syndication.
