Analysis And Application Research Of DS Truck Operation Based On Positioning Data

With the development of the Internet,cloud computing and artificial intelligence,data mining technology has been widely used,and application data mining has brought many practical values to enterprises.In recent years,due to the popularization of GPS positioning device,China has published relevant regulations in 2016.It requires that heavy-duty trucks must install and use satellite positioning devices with driving record function.The transportation industry has accumulated a large amount of vehicle positioning data,and the promotion of data mining,artificial intelligence and cloud computing has provided the possibility to analyze and mine these positioning data.Therefore,how can we use the data mining technology to mine the positioning data generated by the truck,find the valuable information contained in it,and use it for the enterprise to improve the management level of the enterprise and the competitiveness in the industry?This is the problem which worth considering and studying.This paper takes the positioning data generated by all the trucks of DS company as the research object,and analyzes and discusses the data mining and application problems according to the company’s needs.Firstly,the general positioning data preprocessing including data coordinate transformation,data denoising,trajectory compression,map matching and trajectory completion is described,and a high cohesive low coupling positioning data preprocessing framework is proposed.The research data is preprocessed,combined with the requirements of DS company and the characteristics of the company’s positioning data,which laid the foundation for the later research.At the same time,relevant researchers can combine their own research purposes to pre-process and further study the data within the framework.Secondly,based on the results of DS company positioning data preprocessing,the truck stay is analyzed,which includes the overview of truck stay analysis,the stay analysis process based on the positioning data,and the automatic generation of the electronic fence.Also,based on the results of-data mining,application analysis and research are carried out in three aspects:electronic fence,service point distribution,residence time and duration.Aiming at the inaccuracy and overlap of the electronic fences drawn by the existing manual methods,and the problem that the DS company trucks stayed at certain business points for too long,the corresponding solutions were proposed.Thirdly,based on the travel interval generated by the data processing in the first two chapters,according to the DS company’s demand,data characteristics and its concerns,the main content of the DS company’s driving analysis is determined.The data mining and analysis are carried out in three aspects:the truck driving time and duration,the truck driving efficiency and truck running speed.The analysis shows that the company’s trucks mainly have long-term driving,fatigue-time driving and speeding,and the problem is more serious.At the same time,the company’s truck utllization rate(20%-40%)is relatively low.In response to these problems,we will propose corresponding solutions and measures from the actual situation of DS company.Through the analysis and mining of the positioning data generated by the DS trucks,the company basically meets the needs of the company to understand the operation status of the company’s trucks through the positioning data,thus helping the DS company to better manage the trucks and further improve the management efficiency.And management level.