Forecast Of Highway Spare Parts Demand Based On Analysis Of Influencing Factors

With the increase in the density of highway network in China,the influence of highway electromechanical systems in production management is increasingly significant.The failure of the electromechanical system will lead to the abnormal operation and cause a series problems in production security,which will affect the entire road network and endanger the personnel safety of the passengers.Therefore,the completeness of the highway electromechanical system is the precondition of the safe driving.Forecasting the inventory demand accurately is crucial to improving the management level of spare parts.Due to the limited service data,the highway maintenance company in city X is used as an example to carry out the research on the spare parts prediction under the small sample.Through the practical investigation,the inventory data of the highway Maintenance Company in city X is obtained,and the status quo is analyzed.The investigation found that there are blind purchases,random data entry and traditional operation in the current spare parts management.After the research on the theory of inventory demand forecasting,the spare parts are classified into continuous spare parts and intermittent spare parts according to the demand is continuous or not,and the demand forecasting methods are summarized separately.The segmentation failure model of the highway electromechanical equipment is constructed based on the relationship between the demand and the equipment failure by analyzing the research status of the equipment reliability,and estimates the time frame of Infant Mortality and the Wearout of the equipment.Based on the reliability model of the equipment,the running duration is transformed into the failure rate as the main factor,and with the other relevant factors,the sequence of influencing factors for the spare parts demand of the electromechanical system is constructed.Finally,taking the lane railing head equipment as an example,the traditional method based on random forest and gradient boosting regression are used to construct the prediction model.Secondly,the improved random forest prediction algorithm is used to optimize the model,which verifies the validity and rationality of the proposed prediction method.Besides,the results show that the equipment failure model has a higher weight in the whole training process,which plays a major role in the forecast of equipment demand.In order to prove the validity,the same predict method is used one more time without the failure model,and it is found that the prediction model combined with the failure model improved the prediction accuracy significantly.In addition,in view of the defects of the spare parts management method of the highway maintenance company in city X,the inventory optimization management method is proposed based on the research content,which has good application value in the spare parts inventory management of highway electromechanical system.