Research On Reserve Quota Of Equipment Spare Parts In Coal Enterprises

Equipment spare parts inventory management plays an important role in the production and operation of coal enterprises,which is the premise of continuous production.Due to the particularity of the production environment of coal enterprises,in order to improve the production efficiency and ensure the safety of personnel,in the whole production process,coal enterprises have implemented mechanized production at the largest level.There are many types of mechanical equipment in production,and most of them are large and complex in structure,which will inevitably lead to a large number of equipment spare parts demand,and the types of demand are more diversified.Therefore,how to ensure the normal production and operation of enterprises,improve the efficiency of spare parts inventory management,reduce the cost of enterprise spare parts inventory management has become the focus of coal enterprises to solve the problem.Based on the analysis of D coal enterprise's spare parts inventory management and spare parts demand,and the previous research on the classification of spare parts,spare parts inventory management,spare parts reserve quota of coal enterprise,this paper finds that(1)the classification of spare parts is single,and(2)there are two main factors affecting the spare parts reserve quota,which are spare parts demand and safety stock,and different demand Different methods should be used to predict the characteristic spare parts.Therefore,this paper first optimizes the BP neural network classification model of literature,and the system automatically identifies the changes of classification parameters.According to the changes of classification parameters,the classification model carries out self-learning dynamic classification.Secondly,the improved BP neural network prediction dynamic self-learning model is constructed by selecting important types of spare parts with more frequent delivery.In this paper,the BP neural network is established to predict the consumption of spare parts,and then the GM(1,1)prediction model is used to correct the prediction error(safety stock),and the consumption of spare parts is predicted by the corrected prediction value.In this paper,103 groups of daily data of a spare part from December 2019 to April 2020 are selected as the original data of neural network prediction.The first 85% of the data are selected for training,and the remaining 15% of the data are used for testing.A 4-9-1 BP neural network model with one hidden layer is established to predict the demand of spare parts.Then,taking the error value of spare parts demand prediction as the original sequence,the GM(1,1)prediction model is established to predict the safety stock.The prediction model adjusts the prediction model by self-learning with the change of prediction parameters.Finally,the forecast quantity of spare parts demand and the forecast residual(safety stock)are added to determine the final spare parts reserve quota.The results show that the average relative error of spare parts reserve quota based on BP neural network model and GM(1,1)model is 14.31%,and the average accuracy reaches85%,which greatly improves the accuracy of spare parts reserve quota of coal enterprises,and provides the enterprise with the minimum spare parts reserve to ensure production The decision basis of the continuity of the management level and the improvement of the spare parts inventory management level.