Case Study: India’s Largest Automobile Manufacturer achieves faster, more accurate part failure analysis with Daffodil Software’s AI solution

India’s Largest Automobile Manufacturer achieves 88% failure detection accuracy with Daffodil Software

India's largest automobile manufacturer faced significant challenges with its time-consuming and inconsistent motor part failure analysis process, which relied heavily on expert evaluations. The company engaged Daffodil Software to leverage its AI development and computer vision services to revolutionize this process, aiming to automate analyses and improve accuracy for both brittle and ductile fracture categorization.

Daffodil Software developed a custom AI-driven web portal using ReactJS and Python, built on AWS services to meet strict data compliance requirements. The solution included segmentation and classification models to accurately detect fracture types and propagation direction. Daffodil completed the project within a strict two-month timeline, achieving an 88% accuracy rate in failure mode detection and saving approximately 60 hours per month. The AI model's prediction speed improved significantly, providing results in just 2.5 to 3 seconds.