Study On Feature Engineering And Portrait Technology Of Safety Production Supervision In Logistics Enterprises

The optimization of production safety supervision business of logistics enterprises needs to clarify the influencing factors of supervision behavior and their relationship,which is also the basis of evaluating the attitude and ability of production safety supervision team of logistics enterprises.At present,the research on the supervision behavior of production safety of logistics enterprises is mostly qualitative and summarized according to experience.The data value hidden in the big data of production safety of logistics enterprises has not been mined and effectively utilized,and the problem of data support for supervision decision-making needs to be solved urgently.In 2018,Beijing municipal administration of emergency logistics enterprise security regulatory domain business data as the research object,the logistics enterprises regulatory supervision ability the supervisory ability of the research and quantitative characterization of vector technology research,regulators’ ability to the logistics enterprise to carry on multi-dimensional characterization and evaluation,there are three main aspects of the work content:(1)using the characteristics of engineering and thinking and methods of data analysis,from the human input,inspection process,punishment and accident frequency point of trajectory and behavior analysis of logistics enterprise supervision team,in-depth discussion and inherent law in all kinds of data,puts forward optimized for regulatory business direction;(2)build portrait model of the supervisory ability of the logistics enterprises to realize quantitative describe government departments and regulators at the grass-roots level the content of the work for all kinds of logistics enterprise supervision full degree,the degree of reasonable arrangement of work,work quality from three aspects,information,provide for logistics enterprise safety regulators and business personnel is operational and comprehensive data support;(3)The cartoon face cluster visualization map and supervision display platform of the supervision ability of the supervisors of various logistics enterprises in Beijing are constructed,which can most intuitively judge the index level,similarities and differences and characteristics of the specific indicators of the supervision ability of various logistics enterprises,strengthen the comparison and accelerate the analysis speed of decision makers.In this paper,by means of information technology,the safety supervision of logistics enterprises is transformed from experience decision-making to data-driven,so as to improve the safety supervision of logistics enterprises and improve the efficiency of supervision work.According to the latest data from the Emergency Management Agency,the accuracy rate of law enforcement inspection has increased from 73 percent in 2015 to 91 percent now,and the rectification rate of hidden dangers has increased from 58 percent in 2015 to 95 percent now.Moreover,the results of the visualization system of logistics enterprises’ safety production supervision are connected to the Beijing leadership cockpit,providing the city leaders with special auxiliary decision-making services for urban safety.