Spare Parts Demand Forecast And Inventory Optimization Based On Preventive Maintenance

The development of agricultural modernization makes significant increase in thenumber of agricultural equipment, as well as large-scale, automation and intelligence ofequipment, which result in larger need and increasingly prominent of maintenance. But theexisting later maintenance mode makes high costs on maintenance and downtime. Inaddition, the unreasonable repair parts inventory program causes overmuch or shortage ofspare parts inventory, resulting in increased cost of shortage and inventory. Therefore, asthe main content of equipment maintenance, it is significant to do research on maintenancestrategy and spare parts inventory management.In order to determine the reasonable management strategy for maintenance and spareparts inventory, agricultural equipment manufacturers are seen as the research object in thispaper. Based on a comprehensive analysis of existing maintenance and inventorymanagement model, a set of maintenance and spare parts inventory management programfor agricultural equipment is proposed combined with relevant researches at the same time.Maintenance strategies for preventive maintenance are determined according to agriculturalcharacteristics and failure modes. Then researches on spare classification methods forpreventive maintenance are conducted. Analytic Hierarchy Process and ABC classificationare combined based on the traditional ABC classification as well as the characteristics ofagricultural machinery spare parts. Key spare parts are selected to do need forecastingresearch combined with results of spare parts classification. First, work characteristics andprediction methods of agricultural machinery are analyzed, and BP neural networkalgorithm is determined as the most appropriate predict method. Then key factors affectingspare parts demand are analyzed. By data processing, neural network building and training,the network will eventually be used for the training of qualified spare parts demand forecast.Then two kinds of inventory models of centralized ordering and regional allocation are builtby setting minimize inventory costs as the goal, which can be used for inventorymanagement among inter-service stations and regional centers, regional centers andsuppliers levels, and genetic algorithm is selected to solve the inventory model, so that theoptimal inventory control program is determined. At last, on the basis of theoreticalresearch, maintenance and spare parts inventory management systems for agriculturalequipment manufacturers are designed and validated with relevant examples.