As the core protective component within molded case circuit breakers (MCCBs), the performance consistency of the thermal trip unit is critical to grid safety and equipment lifespan. To overcome limitations of traditional methods in multi-objective optimization and tolerance control, this study constructs a framework integrating an enhanced Multi-Objective Particle Swarm Optimization (MOPSO) algorithm with Monte Carlo simulation. A coupled thermo-electro-structural multi physics model was established on ANSYS Workbench. The enhanced MOPSO, featuring dynamic inertia weights and adaptive learning factors, successfully co- optimized thermal response time and tripping force, yielding Pareto-optimal solutions with optimized bimetal dimensions (l = 40mm,δ= 0.4mm ). Validation showed the thermal response time at 1.3In reduced to 640s, a 22.89% improvement. A quantitative protection margin model (η) was proposed to evaluate tolerance and current fluctuation impacts. Monte Carlo simulation (5000 samples) guided a tolerance-current control strategy, achieving a 95% satisfaction rate for the robustness criterion (η >=1.2 ). This work provides a solid foundation for efficient design and scalable manufacturing of low-voltage appliances.