Conductive rubber electric heating technology is an innovative deicing technology, which provides a safe and environmentally friendly long-term solution. In the process of snow melting, the wavy temperature distribution generated by the conductive rubber heating system leads to discontinuous snow melting, which makes the existing snow melting performance indicators insufficient to accurately evaluate its performance. Therefore, this study introduced two new snow melting performance indicators (temperature non-uniformity coefficient, CT and temperature range, RT), and proposed a multi-dimensional comprehensive quantitative evaluation method for road snowmelt performance. Firstly, a two-dimensional finite element snow melting model of bridge deck pavement structure with conductive rubber heating system (BDPS-CRHS) was established and verified by outdoor snow melting test. The effects of design parameters such as laying width, laying spacing, buried depth and input power on snow melting performance indicators were analyzed by thermal simulation. Subsequently, the prediction models of snow melting performance indicators were established using response surface method. A new comprehensive evaluation indicator (snow melting performance coefficient, SMPC) was proposed and its calculation method was provided. Finally, the SMPC was used as the fitness function, and the design parameters of conductive rubber snow-melting bridge deck were optimized by the I-PSO algorithm. The results showed that the optimal design scheme was conductive rubber width of 0.55 m, laying spacing of 0.15 m, buried depth of 0.05 m, input power of 587 W/m2. This study provides a scientific basis for the quantitative evaluation and design parameters optimization of snow melting performance of electric heating pavement, and ensures the efficiency and reliability of snow melting effect.