The Research Of Construction Safety Management Based On The Individual Characteristic Difference Of Construction Workers

With the continuous expansion of industrial scale and the vigorous development of economic benefits,the construction industry has become a pillar industry of the national economy and also causes serious security problems.The statistics of the MOHURD showed that the number of safety production accidents and deathsin the national municipal engineeringhave increased considerably in 2016 than in 2015.Although the country has enacted a series of relevant laws,regulations and policy provisions,they don't fully consider the key element of the management object,which greatly reduces the matching of management measures with the management objects.The management process lacks scientific mode guidance,the system is in the form,and the workers are exhausted,which seriously reduces the management efficiency,and at the same time,causes the sharp increase of cost.The situation of safety production is becoming more and more severe.The urgent situation has prompted domestic and foreign scholars to carry out a lot of research in the field of construction safety management.However,the construction workers have been considered as a whole in the past research,which circumvented the problem that the different worker groups under the action of individual characteristics have an effect on the management effect.These studies have some limitations on the setting of management objects.Therefore,in order to make up the defects of the existing construction safety management,and adapt to the development trend of the personality of the construction workers in the new era,this paper is based on individual characteristics of construction workers,constructs individual characteristic evaluation index system of construction workers,and uses latent class model to classify groups.In this paper,the corresponding management measures are put forward for different groups,and security class management measures of construction workers are formulated.The paper hopes to solve the problem that management measures and management objects cann't match effectively,so as to provide new ideas for construction safety management.First of all,theliteraturereview,which consists of basic theories of unsafe behavior,influencing factors of unsafe behavior,management failure theory,safety management mode,construction safety management,personality trait theory,group theory and informal organization theory,provides a scientific basis for the choice of perspective and the construction ofresearchideas.Secondly,this study uses the method of literature statistics,concept carding and Delphi method to construct individual characteristic evaluation index system of construction workers,and designs the corresponding questionnaire for reference to the maturity scale of each index.231 valid questionnaires were issued and recovered,and the survey data were preprocessed and entered into SPSS statistical analysis software.These data provide data support for subsequentpotential class analysis.Again,the preprocessed data are entered into the Latent GOLD statistical analysis software.A latent class model is established in this paper.Then,the model outputs results,and gets four potential categories.This paper is named for worker groups according to the potential characteristics of each category,followed by "knowingly and willfully type","knowledge deficiencytype","mental tension type" and "quality type".Finally,this paper analyzes the characteristics of workers' groups in great detail,proposes corresponding safety management measures on the basis of related theories such as psychology,management and safety behavior,and sets up the construction worker safety group management measure table.The effective matching of management objects and management measures is realized.This research has important practical significance for controlling construction workers' unsafe behavior and improving construction safety management efficiency,and also provides reference for safety management innovation research.