At present, heavy industrial competition makes more manufacturing pay attention to the service based on their products. Therefore, product service system has caused the extensive value of the academic circles. Service resource scheduling is the key step in the product service delivery, which depends on the knowledge mined from history service data and product information in the use process, namely the different fault maintenance scheme, technicians' skill, equipment state information, fault prediction information, work plan, etc. Based on this, this paper puts forward a resource scheduling method for predictive maintenance services of equipment whose location change dynamically, aiming at eliminating potential failure, minimizing service cost and outage cost, considering the technicians' ability of different maintenance task, fault prediction information, equipment operation plan and other constraints. First, a mathematical model is set up to describe this problem. Then this paper adopts a hybrid algorithm to resolve that. Last, the result that the best time of servicing, route planning and the reasonable technician, shows this method can improve the service level and reduce total cost.