Accelerating lightweight Structural Design in Automotive Engineering through PLR TRIZ -KBE and AI-driven Generative Design

¹ Université Hassan II, ENSAM, Casablanca, Maroc
² Université Marie et Louis Pasteur, UTBM, CNRS, institut FEMTO-ST, F-90010 Belfort, France

^* Corresponding author: Hassan.rouane-etu@etu.univh2c.ma

Abstract

This study introduces a systematic workflow that combines Patent Literature Review TRIZ (PLR-TRIZ), Knowledge-Based Engineering (KBE), and generative design to optimize compressor brackets for electric vehicles. By extracting inventive principles from global patents and embedding them into automated design rules, the methodology enables rapid exploration of innovative, manufacturable, and patentable solutions. Generative design algorithms, guided by multi-physics constraints and material selection using the Ashby index, produced over 26 design variants. The optimized bracket achieved a 66% reduction in mass, a 22% increase in first natural frequency, and a 22% decrease in material cost compared to the legacy design. Finite element analysis confirmed that all critical criteria— static strength, vibration resistance, fatigue life, and thermal stability—were satisfied, with safety factors above two. The workflow is fully automated and scalable, supporting eco-design and industrial innovation. Unlike classical optimization approaches, this method integrates inventive problem-solving, automated rule-based design, and advanced computational exploration, resulting in solutions that are both high-performing and industrially feasible. Beyond automotive brackets, the methodology applies to the aerospace, energy, rail, marine, and medical sectors, enabling the development of lightweight, robust, and sustainable components. These findings highlight the potential for accelerating innovation and deployment in advanced engineering.