Research On Vehicle-Cargos Matching Method Based On Freight Big Data

The road freight market is the most important freight market in my country,and its freight volume accounts for much more than other freight transportation methods.However,since more than 90% of truck drivers in my country are self-employed,it is impossible to optimize the integrated capacity.In addition,my country’s vast territory and uneven distribution of resources,logistics demand also show temporal and spatial differences and fragmentation.In this context of supply and demand,China’s road freight market lacks effective vehicle-cargos matching,presenting a situation of " Car to find hard goods,goods hard to find a car " and a market pattern of "small,scattered,scattered,and chaotic".In addition,trucks are seriously overloaded,high-risk and low-efficiency phenomena such as severe empty driving also frequently occur.The emergence of the Online freight platform has enabled the sharing of information resources between the trucks and cargos sources to a certain extent,strengthened the integration of freight resources,and has accumulated a large amount of freight platform data.However,the existing Online freight platform still does not effectively mine these historical data,summarizes the characteristics of vehicle-cargos matching,and develops an effective vehicle-cargos matching model.Based on the above background,this paper proposes a two-stage matching method for trucks and cargoes based on the freight big data of the Online freight platform.The first stage divides the matching pool by static characteristics of trucks and cargos,and the second stage considers other matching indicators to complete further matching methods.First of all,this paper analyzes the current stage of the Online freight platform,the research on the problem of vehicle-cargos matching of supply and demand,and the current national and road freight industry standards for trucks and cargos classification.At the same time,the research data of this article is introduced and preprocessed,combined with the actual situation of the industry,the classification of trucks and cargos in this paper is determined,and the static characteristics of trucks and cargos of the platform freight data are preliminary analyzed.Secondly,based on the classification of trucks and cargos,the cargo category of different vehicle models is analyzed,and the vehicle-cargo matching pool of vehicle-cargo category is preliminarily divided.Then,thepaper further analyzed the combined distribution of the length and load of various trucks,and divides the matching pool of vehicle type-cargo-type-length-load based on this.Combining the matching pool division process,establish rules for dividing trucks and cargos into different matching pools based on static characteristics.Finally,for the divided matching pool,a vehicle-to-cargo bilateral matching model that considers other matching indicators such as probability of receiving orders for trucks,freight rate,route familiarity,time matching degree,etc.,is constructed.Through the analysis of historical freight data,the definition and estimation of the key indicator "truck order probability" in this article are given.The genetic algorithm is used to solve the calculation example,and the effectiveness of the matching method proposed in this paper is compared and analyzed,which provides more ideas for the optimization of the Online freight platform for the platform.