Thermal conductivity is a key determinant of thermos-physiological comfort in apparel textiles, governing the rate ofheat transfer between the human body and the external environment. Although pure silk (silk-by-silk) fabrics are widelyappreciated for their lusture, drape, and smooth handle, previous studies report relatively higher thermal conductivitydue to the compact filament structure and reduced air entrapment, which may limit insulation performance undervariable climatic conditions. Earlier investigations have concentrated on mechanized woven systems or single-fiberconstructions, with limited systematic optimization of handloom-woven silk–cotton union fabrics using statisticaldesign approaches. Addressing this research gap, the present study employed Response Surface Methodology based ona Box–Behnken Design to optimize thermal conductivity by varying weft yarn count (20–40 Ne), twist per inch (10–20TPI), and picks per inch (55–65 PPI), while maintaining constant silk warp parameters. Thermal conductivity rangedfrom 0.0014 to 0.0020 W/m·K, with pick density showing the most significant influence; the minimum value (≈0.0014W/m·K) was achieved at moderate yarn fineness, controlled twist, and higher pick density. Compared with silk-by-silkfabrics, the optimized union fabrics exhibited lower thermal conductivity due to enhanced air entrapment from cottonweft yarns. The study demonstrates the potential of statistically engineered handloom fabrics for sustainable, climateresponsive apparel, with future prospects in smart textile integration and performance-oriented garment design