Research On The Demand Prediction Of Emergency Materials And Optimal Scheduling Of Enterprise Supply After The Earthquake

As one of the most harmful sudden natural disasters,earthquake is a serious threat to human survival and social development.The Earthquake disaster is inevitable.When an earthquake occurs,the timely supply of emergency supplies can reduce losses in the disaster-stricken areas and acts as an important part of post-earthquake emergency managements,which has become a major issue for countries all over the world to confront.The key to the earthquake relief depends on the demand predictions of emergency supplies and also the supply optimizations of emergency supplies for the organizations participating in the rescues.In recent years,scholars at home and abroad have carried out a lot of researches on the demand prediction and supply optimization of post-earthquake emergency materials.However,they are not performed well in practical applications,mainly reflected in:(1)the demand prediction of emergency materials is closely related to the supply optimization.However,existing researches lack the consideration of integrated system,which is not conducive to the overall system optimization;(2)the existing research focuses more on the optimization of sub-systems such as optimization of the distribution path and the production of disaster relief materials in the emergency logistics process,but lack the whole chain optimization whole chain optimization covering the production,scheduling and distribution of emergency materials,as well as the consideration of practical scenario constraints such as the number of distribution vehicles and time windows.(3)the existing research has room for further improvements in the prediction of effects on material demands and the solution efficiency to material supply optimization problems.In view of the characteristics of the post-earthquake emergency start-up period,emergency rescue period and transitional recovery period,combined with the existing problems of emergency material demand forecasting and supply optimization related research,this paper analyzes and solves the problems encountered by enterprises in the process of participation under the guidance of system thinking emergency material demand forecasting and supply optimization problems.Based on related theories of emergency material demand forecasting,emergency material distribution and transfer point locations,the paper focuses on the construction of the model of emergency material demand prediction,integrated model of enterprise emergency material supply and distribution,and integrated model of the enterprise emergency material transfer point location distribution.The application results have revealed that the method proposed in this paper can effectively improve the dynamic demand forecast and supply optimization effect of emergency supplies after the earthquake,and has certain theoretical and practical significance.The main research work of this paper is as follows:(1)Research on dynamic demand prediction for post-earthquake emergency supplies based on machine learning.In the early stage of the earthquake,according to the demand for emergency supplies in different situations after the earthquake,considering that the emergency materials are directly determined by the number of casualties,a hybrid regression algorithm based on sub box was designed based on the idea of predicting the demand for emergency supplies based on the number of casualties after the earthquake.And the prediction model of overall earthquake casualties is constructed by using the related data available in the early stage of the earthquake;Given that the dynamic change of casualties with time,based on the empirical function obtained by fitting historical earthquake data,and In the view of time series prediction,the dynamic earthquake casualties are predicted by using short-term and long-term memory time series prediction algorithm;The material demand model of warehouse concept is introduced to satisfy the scientific prediction of material need.(2)Research on production and distribution integration of enterprise emergency materials in post earthquake emergency rescue period.On the basis of(1),for the integrated production and distribution planning problem oriented by the prediction results of emergency material demand,we explore the problem of multi-vehicle route planning with time windows and uncertain transportation time;design an integrated production and distribution model considering the actual circumstances after the earthquake,concentrate on the indeterminacy of the emergency transportation time,and use the elastic p-robust method to avoid the difficult-to-estimate probability distribution of random parameters.To solve the problem,an improved cultural genetic algorithm is proposed,substituting local search for mutation operators to construct high-performance heuristic algorithms,and adopt advanced population diversity schemes to improve the quality of the algorithm’s solution.(3)Research on location and distribution integration of enterprise emergency material transfer point in post earthquake emergency recovery period.In view of the establishment of emergency distribution transfer points and the distribution of emergency materials for participating rescue companies,especially in the transitional recovery period after the earthquake,integrated optimization management is carried out from the perspective of overall system optimization,and the timely delivery of rescue materials and transportation costs are fully considered.The benefit is optimal.According to the demand time window of each disaster site,a comprehensive dual-objective warehouse location and vehicle routing model are constructed;an efficient hybrid heuristic algorithm based on ant colony algorithm is proposed to effectively solve the dual-objective optimization problem,which improves the calculation efficiency and the quality of the solution,increases the choice space of decision makers,and thereby helps decision makers make rapid response to post-earthquake rescue work and provide stable decision support.This research is aimed at the emergency logistics needs in different rescue stages after the earthquake,and effectively uses the latest technologies in the research fields of systems engineering,supply chain management,emergency management,data mining,etc.,and aims to solve a series of core issues involved in the emergency logistics after the earthquake,such as emergency material demand forecasting,emergency material production and distribution,and location of emergency material transfer points.In the specific research work,the dynamic forecast of emergency material demand is based on the machine learning method.And on the basis of the emergency material demand forecast and from the perspective of enterprise participation in earthquake relief,a set of theoretical methods for the integration and optimization of disaster material distribution systems will be established,considering the transportation time,product value loss in the production,requirements of timeliness during the transportation of emergency materials,and building integrated models of emergency material production and distribution.The research results will provide valuable references for enhancing the rescue capability of earthquake disaster emergency,improving post-earthquake emergency rescue efficiency,reducing personnel and property losses in disaster-affected areas,reducing economic operating costs,and improving the scientificity and feasibility of enterprises participating in earthquake disaster relief in China.