Research On Multi-Project Steel Procurement Strategy Ased On RFIMA Model Price Forecast

With the strong support of the government,the process of urbanization in our country is getting faster and faster,the construction industry has also been developing rapidly,and the competition among construction companies is also increasing.In housing construction projects,the project budget shows that construction companies need to spend a lot of money in the purchase of building materials.Therefore,it is of great significance to construction companies to purchase the right batch of raw materials at the right time.Compared with the extensive decentralized procurement management model,the multi-project centralized procurement management model can reduce the company’s material procurement costs and improve the company’s comprehensive competitiveness.The three main materials of housing construction projects are steel,concrete,and wood.Among them,the price of concrete and wood materials fluctuates slightly,while the price of construction steel materials fluctuates relatively large and the cost is relatively high.Therefore,this article takes construction steel as the research object,first selects the daily price of rebar in Qingdao from January 2014 to December 2018 as a sample,and introduces long-memory indicators into the time series forecast of steel prices.After processing,stationarity test,long memory test and other analysis and processing,the ARFIMA(2,0.335,1)steel price prediction model is obtained,and then the weekly forecast price of rebar in Qingdao city in 2019 is obtained by using this model.Monthly forecast price.Finally,using the predicted monthly steel price as a sample,a multi-project steel dynamic procurement is established that minimizes the total cost of raw material procurement costs,one-time procurement costs(one-time fixed procurement costs,transportation fees,and handling costs)and raw material inventory costs.By applying the PSO algorithm to solve the established procurement model,the optimal steel procurement strategy in the time domain of the procurement plan is finally obtained.This procurement strategy can effectively reduce the company’s material procurement costs and enhance the company’s market competitiveness.In order to verify the validity of the multi-project steel procurement model based on price prediction established in this article,this article uses multiple simultaneous production projects of a construction company in Qingdao as a case,and uses actual production data of the project to verify ARFIMA(2,0.335,1)The validity of the forecast model and the multi-project steel procurement model.The research results show that,compared with the decentralized steel procurement strategy before optimization,the multi-item steel procurement model based on price prediction established in this paper has stronger cost optimization capabilities and is more conducive to cost saving for enterprises.It can be seen that the combination of forecasting technology and procurement technology can well solve the problem of multi-project raw material procurement under price fluctuations,and provide a reference for companies to formulate optimal material procurement strategies.