Research And Implementation Of Intelligent Bargaining Decision System Based On Logistics Big Data

With the development of science and technology,the current logistics industry has generally introduced information technology to automatically collect,aggregate,and analyze information generated in the processes of warehousing,transportation,and distribution.Its freight mode has changed to the core platform economy,and the new mode " Car-free Carrier Platform " was born in China.As a freight operator,the car-free carrier platform needs to intervene in freight transactions and negotiate freight rates with the owners of the actual freight,so as to maximize profits.With the increasing number of orders in the logistics and transportation industry,there are more and more factors affecting freight prices.Using traditional linear formulas to calculate freight reference prices and contacting the carrier drivers to negotiate prices on transportation orders will result in waste of transportation resources and low operational efficiency.In order to improve the operation efficiency of the car free carrier platform and make full use of freight resources,this thesis proposes an intelligent bargaining decision system based on Logistics big data platform.The system takes the multi-data source car-free carrier platform data as the research object,and adopts the Spring Cloud Netflix microservice architecture to realize smart bargaining based on the pricing model.In order to realize the system,the transportation data of other platforms was crawled through the Scrapy framework to make up for the lack of the original data,and the transportation data was preprocessed through the rules engine to formulate cleaning rules.Then,based on the cleaned data,the gray correlation analysis method and Spearman’s rank correlation coefficient are used to comprehensively measure the degree of correlation between transportation indicators and freight rates,so as to determine the features required to build a pricing model,and evaluate the multiple linear regression pricing model and the effect of the BP neural network pricing model.In order to fully dispatch transportation resources,design and implement a route similarity algorithm to match bargaining drivers,and then formulate bargaining decision rules based on application scenarios,dynamically adjust the price of the waybill,so as to facilitate the completion of the waybill as soon as possible.Finally,realize the system visualization through the Android application.In the research of the price of transportation orders on the car-free carrier platform,most scholars are committed to the analysis and realization of the waybill pricing module,and the realization of the system engineering is less involved in the bargaining part.This thesis uses two different types of correlation analysis methods: gray correlation analysis and Spearman’s rank correlation coefficient to analyze the influencing factors of freight rates,and determines the input characteristics of the pricing model by combining the analysis results of the two methods.And through two types of test data to compare the platform’s existing guide price calculation formula,multiple linear regression pricing model,and pricing effect based on the BP neural network pricing model.On the basis of studying the pricing of the waybill,according to the actual business scenario,design the bargaining driver matching algorithm and bargaining decision rules,and implement it through the microservice architecture.The test results show that the system has realized the basic functions related to bargaining,and has good scalability and stability.By comparing the MSE and MAE of the three pricing methods,the difference between the waybill price given by the three pricing models and the final transaction price is compared,it is found that the accuracy of the waybill price obtained by the BP pricing model is better than the multiple linear regression pricing model and the platform’s existing guide price calculation formula.