| At present,shipping is an essential component of global logistics.Meanwhile maritime accidents of all kinds inevitably occur and cause grave harm to humanity.Research shows that human factors are the significant factors causing maritime accidents.However,when exploring human factors,the studies focus on risk assessment for the majority,while those on intervention strategy rarely exist.The relevant research is fragmented and lack of systematicness.Therefore,this paper focuses on the selection of intervention strategies and system construction,in order to reduce maritime accidents caused by human factors and improve navigation safety.The main research works of this paper are as follows:(1)Based on the literature review,the maritime accident causation framework model,causation correlation model and accident intervention theory were systematically sorted out,and a conceptual model of maritime accident intervention system was constructed by combining it with Behavior-based Safety(BBS)framework.In this model,intervention targets were confirmed in the first place,and human factors usually used in accident reports and related academic studies were selected as intervention targets.Then,text mining and relevance testing were used to screen and deconstruct intervention targets.(2)Based on the characteristics and correlation features of intervention targets,8 types of intervention means(4 are direct intervention means,and the rest indirect intervention means)were selected as the research priority by combining with PSF human factor model and referring to accident reports and cases in other fields.Results with statistical meaning were gained by comparing the effecti ve test of 8 kinds of intervention means and considering the comparison situations of sailors with different experiences in the experiment.(3)Based on the correlation analysis of maritime accidents and the test results of intervention means,the system dynamics model of human factors intervention in maritime accidents was constructed.Vensim-PLE software was used to determine the boundary of the system,the initial dynamic hypothesis and the equation relationship among the factors,so as to simulate and verify the effect of the comprehensive intervention system proposed in this paper.(4)Based on the research results of this paper and“Investigation of Causes of Maritime Accidents”issued by IMO,comprehensive intervention strategies for human factors in maritime accidents were formulated.Conclusions from the above research are as follows:(1)Based on the BBS framework,this paper fully considers the characteristics and causes of maritime accidents and establishes a conceptual model for the selection,verification and combination of maritime accident intervention means.On the basis of HFACS and the Swiss cheese model,the screening and deconstruction of intervention targets were finished through the text mining and correlation test of maritime accident reports and literature about human factors in maritime accidents.According to the calculation results,14 causes of 5 types of maritime accidents are determined as the intervention targets,and 17 pairs of significant correlation between the intervention targets are obtained.(2)Based on the characteristics and correlation features of intervention targets,8 types of intervention means(4 are direct intervention means,and the rest indirect intervention means)were selected as the research priority by combining with PSF human factor model and referring to accident reports and cases in other fields.The four types of direct intervention means are place check lists,provide accurate aerial information,add role rotation and immediately remind.The four indirect intervention methods are enhance team tacit understanding,promote scene familiarity,strength safety training and improve overconfidence.(3)Based on the comparison of simulation experiment design and statistical test,the effectiveness of eight types of intervention means was tested one by one,and the experience and characteristics of sailors with different identities were considered in the test.The results show that all the eight types of intervention means are effective for the overall sample.Some intervention means were less effective for the captains but more effective for the first mates,pilots and students;some intervention means were less effective for the captains,mates and pilots,but more effective for the students;some intervention means were less effective for the pilots and students but more effective for the captains and first mates.(4)To test the effectiveness of the maritime accident comprehensive intervention system,the paper included the effectiveness parameters obtained by single intervention means into the system dynamics model of the maritime accident comprehensive intervention system.The results show that the maritime accident comprehensive intervention system is effective.When eight kinds of intervention means are applied simultaneously,the intervention effect can reach 32.14%.Different kinds of intervention means hold different effects on the intervention system,among which enhancing team tacit understanding has the best effect on the intervention system,while adding role rotation has the worst effect on the intervention system.The intervention system has different overall intervention effect on sailors with different identities.The direct intervention system has the best intervention effect for the students but the worst intervention effect for the captains.Most direct intervention means showed significant improvement for students without sailing experience,but for sailors with some sailing experience,the indirect intervention was more advantageous.In the end,based on the optimized comprehensive intervention system for maritime accidents and the“Guidelines for the Investigation of the Cause of Maritime Accidents" issued by the IMO,a set of comprehensive intervention strategies for the human factors in maritime accidents are obtained.The innovation of this paper is mainly embodied in:(1)This paper selected human factors in maritime accidents as the research content,focusing on the study of intervention strategies to solve the key problems related to maritime accidents.This paper put forward a conceptual model of intervention system suitable for maritime accidents,aiming to raise effective intervention strategies through the identification of intervention targets,correlation analysis of intervention targets,presentation of intervention means,and intervention means test.The conceptual model integrates methodologies including text mining,experimental simulation,statistical test and system dynamics verification.(2)Unlike the accident report analysis and questionnaire survey of current mainstream applications,this paper chose simulator experiments to design the effectiveness test of various intervention means.To reduce the impact of individual heterogeneity of participants,the experimental design of this study is characterized by a long time span,a wide range of participants and diverse experimental contents.The experiment lasted for one and a half years,during which a total of 35 batches and 273 simulation experiments were completed,and more than 2000 valid experimental data were collected.The relevant data proved to be statistically significant through chi-square test,Mann-WhitneyU test and Kruskal-Wallis test.(3)This study constructed the dynamics model of maritime accident human factor intervention system by integrating the human factor transmission mechanism of maritime accidents obtained from text mining model and testing the effectiveness of single intervention means.Based on the results of system dynamics simulation,this paper proposed a set of comprehensive intervention strategies for maritime accidents,which can be used as a supplement to the Ship Safety Management System(SMS)of the International Maritime Organization(IMO).Value can be created for the prevention of unsafe behavior of seafarers during actual navigation in the future maritime practice. |
PDF Full Text Request