In the process of production scheduling, equipment often bears different loads due to processing different jobs, that is, heterogeneous loads. Due to the impact of heterogeneous loads, the degradation rate of the equipment in each operation is different. This will affect production scheduling and maintenance planning, resulting in idle resources and increased time costs. To solve this problem, a joint strategy of single machine scheduling and predictive maintenance is formulated considering the impact of heterogeneous loads. On this basis, an integrated model is established to minimize the total weighted expected completion time. A degradation model based on the Wiener process is established for equipment affected by the heterogeneous load in the single machine scheduling process. And the cumulative distribution function of the remaining useful life of the equipment is derived. Numerical experiments are conducted to compare the optimization results of the corresponding integration models of the conditions of heterogeneous load with the average load respectively, which indicates that it is necessary to consider heterogeneous load in the integration model. And the sensitivity analysis of the parameters verifies the effectiveness of the integrated model.