From SEM weights to AIGC prompts: A design translation method for age-friendly smart cockpits

The growing aging population poses critical challenges for the design of age-friendly smart cockpits. To address this, the present study proposed a data-driven design method based on the “Perception-Cognition-Behavior” (PCB) framework, while incorporating established ergonomic principles as fundamental design constraints. Using Structural Equation Modelling (SEM), user preferences were translated into quantifiable design indicators. By extracting and categorizing user needs through interviews, SEM quantified the influence of each element in a three-dimensional coupling model. A weight-based keyword mapping mechanism was then developed to convert key design elements into structured keywords for Artificial Intelligence Generated Content (AIGC) recognition, thereby generating preliminary design solutions through human–machine collaboration. Throughout the design translation process, ergonomic requirements related to visibility, control accessibility, cognitive workload, and driving safety were maintained as essential engineering constraints. Multi-dimensional validation was conducted through a Pugh decision matrix, user preference testing, and comparative experiments with traditional design methods. The results indicated that the proposed approach significantly outperformed traditional experience-driven workflows across three dimensions: design feasibility, user preference alignment and acceptance, and methodological effectiveness. This study provides a reproducible methodological framework for age-friendly smart cockpit design, offering novel theoretical support and practical pathways for AIGC-assisted design development.