Application of Artificial Intelligence Algorithms for Technical Textiles Design Based on Performance Requirements

This paper presents a systematic approach to optimize technical textile selection using fuzzy logic as one of the AI techniques. Traditional methods for material design rely heavily on iterative prototyping, which is time-consuming and lacks adaptability to complex performance requirements. To address this, we propose a fuzzy inference system (FIS) that maps four critical performance parameters—tensile strength, elastic recovery, thermal conductivity, and moisture regain—to a suitability score for a predefined set of textiles. The system leverages MATLAB's Fuzzy Logic Toolbox to model linguistic variables (e.g., "high," "low") and employs a Mamdani FIS with tunable membership functions and rule bases. Case studies demonstrate the system's ability to recommend materials with high accuracy compared to expert evaluations, thereby significantly reducing design cycles. This framework is particularly valuable for industries requiring rapid, data-driven decisions, such as aerospace, healthcare, and sportswear. This framework is particularly valuable for industries requiring rapid, data-driven decisions, such as aerospace, healthcare, and sportswear.