With the rapid development of the Chinese industry, the reliability assessment of special safety-critical equipment in variable and complex working environments is becoming increasingly important. In particular, an innovative Remaining Useful Life (RUL) prediction method is proposed for the reducer, a core component of such equipment. The method, which applies the meta-action theory, establishes a transmission efficiency model of the reducer, derives the performance degradation process of the meta-action chain in long-term operation, and calculates the degradation threshold of the reducer. The meta-action transmission efficiency is then used as the degradation index to establish a nonlinear Wiener stochastic degradation process model. The likelihood function of the unknown parameter is solved using the particle swarm optimization (PSO) algorithm for parameter estimation. The proposed method is then validated with experimental data of a certain model reducer, demonstrating its robustness and effectiveness in predicting the RUL distribution of the reducer.