To explore the influencing factors and mutual influence mechanism of the construction safety of China’s high-speed railway stations, this study takes the Hanghuang high-speed railway Fuyang station as the subject and studies 17 risk factors in 4 categories affecting construction safety based on system safety theory, and the interaction relationship and degree among the factors were analysed. Based on DEMATEL (Decision-making Trial and Evaluation Laboratory) and ISM (Interpretative Structural Modelling) methods, through a questionnaire survey, the logical relationship among the influencing factors is quantified. Then, the influencing degree, influenced degree, centrality, and causality of the influencing factors were calculated, and a multilevel hierarchical hybrid model is established to systematically analyse the influencing factors and the mechanism of high-speed railway station construction. The results show that the factors of construction safety risk are summarized as 3 main factors, 6 important factors, and 7 direct factors and personnel factors and management factors need to be controlled with emphasis. In addition, some measures are proposed. This research provides a theoretical basis and method for preventing accidents and improving the safety of high-speed railway station construction.
With the rapid development of China’s high-speed railway, its construction scale and speed of development have ranked first in the world [
In the study of safety accidents, safety accident theories such as human error theory [
There are two types of risk assessment research: qualitative and quantitative [
In the 1960s, Tavistock institute proposed the complex social technology system, referring to an industrial organization that is technology-intensive and capital-intensive and accumulates large amounts of energy [
To determine how to control a complex system, we must first study the relationship of the factors. Existing methods for complex system factor analysis include ISM [
Various factors affecting the safety of complex systems interact. The influencing factors of high-speed railway station construction systems have an extraordinarily complex relationship and interactions, showing strong nonlinearity and reverse feeding. From 2012 to 2018, the types of accidents that frequently occurred in China’s construction industry were falling from a height, structural collapse, lifting injuries, object attack, vehicle injuries, mechanical injuries, electric shock, poisoning and suffocation, fire and explosion, and other injuries. These accidents were the result of the interaction of multiple factors, including humans, machines, the environment, and management, rather than each case having a single cause.
Hamid believed that the cause of an accident is a multifactor phenomenon mainly involving worker negligence and bad attitudes towards safety, lack of safety devices, unskilled workers, etc. [
Through the analysis and summary of the relevant literature [
Construction safety risk factors and their interpretation.
Factor | Interpretation |
---|---|
Cognitive deficits R1 | Managers and workers cannot recognize the possibility of safety accidents. Safety awareness is a prerequisite for operational safety. |
Poor physical or mental health R2 | Personality traits such as impulsiveness, emotional response, and carelessness. Physiological conditions such as illness and fatigue. |
Illegal operations R3 | During the construction process, the construction is not carried out according to the construction safety operation guidelines. |
Unreasonable equipment selection and layout R4 | The equipment model does not meet the needs of the project, the site layout is not conducive to the construction, and there are potential safety hazards. |
Lack of equipment maintenance and overhaul R5 | The equipment lacks daily maintenance and professional overhaul assessment. |
Mechanical failure R6 | During the construction process, for various reasons, equipment malfunctions, stops operating, and even causes chain accidents. |
Numerous large-scale equipment cross-operations R7 | The operations of large-scale mechanical equipment such as tower cranes and construction vehicles are complex. |
Failure of construction monitoring equipment R8 | Construction monitoring such as foundation pit monitoring, mass concrete pouring temperature monitoring, steel structure stress monitoring, and equipment monitoring fails or is performed incorrectly. |
Unforeseeable natural factors R9 | Earthquakes, typhoons, floods, and so forth occur. |
Complex hydrogeological environment R10 | The natural conditions of construction and harsh geological conditions may cause foundational engineering accidents. |
Complex construction site environments R11 | The combination of ventilation, lighting, temperature, humidity, noise, dust, and harmful gases affects the physical and psychological conditions of construction workers. |
Imperfect safety management systems R12 | Relevant safety regulations are lacking or unreasonable. |
Incomplete safety disclosure R13 | The construction leader does not train the direct production workers on safety operational rules and precautions before beginning the production operation. |
Inadequate acceptance of key processes R14 | As a necessary condition for the acceptance of a branch project, key process acceptance is an effective quality control measure that highlights the key points of supervision. |
Inadequate supervision of safety behaviours R15 | Leaders and management personnel at all levels do not sufficiently inspect and supervise the workplace to correct improper behaviour and ensure the safety of production. |
Inadequate safety protection measures R16 | The arrangement of devices to prevent operators from generating risks in the production process is not in place. |
Errors in construction safety operational guidelines R17 | Operational sequences and operational instructions are not specified in the method and safety regulations for technical operations. |
In the construction safety risk of high-speed railway station, personnel factors, mechanical factors, environmental factors, and management factors interact and influence each other. The direct objects of management are human, machine, and environment; people have subjective initiative which will have an impact on things, environment, and management; mechanical equipment is affected by the environment, and the state of the equipment depends on personnel operation and reasonable and effective management; the environment has direct or indirect influence on people, things, and management. The interaction and mutual influence of human, material, environment, and management factors finally form human-material-environment-pipe coupling risk.
System Dynamics (SD) is the subject of information feedback systems. It is a combination of system theory, control theory, and information theory and provides a way to understand and solve system problems [
SD model causality diagram.
DEMATEL, a method proposed by Bottle in 1971, uses matrix operations to calculate the direct or indirect causality and degree of influence of the elements. By using the causal diagram to determine the basic nature of the complex problem, the core problem is determined and solutions are suggested. DEMATEL can be used to divide factors into causal and influencing factors. In addition, by ranking or prioritizing causal factors, core problems can be solved quickly and effectively to improve performance [
The process of using DEMATEL-ISM model as shown in Figure Determine the construction safety risk factors Generate the direct impact matrix of construction safety risk factors. According to the experience of experts and construction site personnel Obtain a new normalized impact matrix from the direct impact matrix where Calculate the comprehensive impact matrix Since Calculate the influencing degree The greater the centrality, the more important the factor. Calculate centrality and causality. Centrality Establish the causality-centrality diagram of the influence factor, the causality is the ordinate, centrality is the abscissa, and the diagram is drawn on the Cartesian coordinate system. Calculate the overall impact matrix where Calculate the reachable matrix, The key to establishing the reachable matrix is to determine the threshold Determine the reachable set Verify that the following formula holds. If so, the corresponding factor is the bottom factor and the Repeat steps 10 and 11 until all factors are crossed out. According to the order in which the factors are crossed out, combined with the causality and centrality, the multilevel hierarchical hybrid model of causality-centrality is established.
Basic flow of DEMATEL-ISM model.
This study takes the Fuyang station of Hanghuang high-speed railway as the subject. Fuyang station is located in Fuyang District, Hangzhou City, Zhejiang Province. The station house has three floors, one underground and two above ground. The underground is reinforced concrete frame structure, and the above-ground part is steel structure. The location of Fuyang station is shown in Figure
The location of Fuyang station (from Google Maps).
The excavation depth of the foundation pit is 6.8∼14.4 meters, which is the first level foundation pit. Fuyang District has abundant rainfall and high groundwater level. According to the above-determined construction safety risk factor set, we adopt the method of expert consultation to study the relevant influence relations of the 17 factors that affect construction safety. Twenty-four questionnaires were distributed to professors and associate professors of the School of Civil Engineering and Architecture of Ningbo Institute of Technology, Zhejiang University, and safety managers of China Railway Fourth Bureau, and twenty-two questionnaires were recovered (the recovery rate was 92%). The influence of a factor is represented by a five-point scale, where 0 means no influence, 1 means weak influence, 2 means general influence, 3 means strong influence, and 4 means extremely strong influence.
Arithmetic mean processing of the data in the questionnaire is carried out to eliminate expert error as much as possible, and the direct impact matrix
According to step 3, normalize the direct influence matrix to obtain the normalized influence matrix
According to step 4, calculate the comprehensive influence matrix
Calculate the influencing degree, influenced degree, centrality, and causality, as shown in Table
DEMATEL parameter table.
Factor | Influencing degree | Influenced degree | Centrality | Causality | Centrality ranking | Factor attribute |
---|---|---|---|---|---|---|
R1 | 2.435 | 0.868 | 3.303 | 1.567 | 6 | Reason factor |
R2 | 1.267 | 0.855 | 2.122 | 0.412 | 15 | Reason factor |
R3 | 1.613 | 2.317 | 3.929 | −0.704 | 1 | Result factor |
R4 | 1.512 | 1.629 | 3.141 | −0.117 | 10 | Result factor |
R5 | 1.194 | 1.669 | 2.862 | −0.475 | 14 | Result factor |
R6 | 1.140 | 2.043 | 3.183 | −0.902 | 8 | Result factor |
R7 | 1.546 | 1.580 | 3.126 | −0.034 | 11 | Result factor |
R8 | 1.516 | 1.926 | 3.442 | −0.410 | 3 | Result factor |
R9 | 1.421 | 0.191 | 1.612 | 1.230 | 17 | Reason factor |
R10 | 1.440 | 0.186 | 1.626 | 1.253 | 16 | Reason factor |
R11 | 1.798 | 0.832 | 2.630 | 0.966 | 14 | Reason factor |
R12 | 1.977 | 1.170 | 3.147 | 0.807 | 9 | Reason factor |
R13 | 1.600 | 1.729 | 3.330 | −0.129 | 5 | Result factor |
R14 | 1.061 | 2.029 | 3.090 | −0.968 | 12 | Result factor |
R15 | 1.691 | 2.032 | 3.723 | −0.341 | 2 | Result factor |
R16 | 1.047 | 2.218 | 3.265 | −1.171 | 7 | Result factor |
R17 | 0.980 | 2.419 | 3.399 | −1.438 | 4 | Result factor |
As shown in Table
Causal diagram of influencing factors of construction safety risk.
According to step 9, select multiple thresholds
Comparison of different threshold
Factor | 0.09 | 0.1 | 0.11 | 0.15 |
---|---|---|---|---|
R1 | 13 | 13 | 13 | 10 |
R2 | 5 | 3 | 2 | 0 |
R3 | 27 | 24 | 18 | 5 |
R4 | 18 | 17 | 11 | 1 |
R5 | 15 | 11 | 5 | 1 |
R6 | 18 | 16 | 13 | 1 |
R7 | 19 | 14 | 11 | 0 |
R8 | 22 | 21 | 15 | 0 |
R9 | 7 | 5 | 3 | 0 |
R10 | 9 | 5 | 1 | 0 |
R11 | 14 | 10 | 9 | 1 |
R12 | 15 | 13 | 13 | 5 |
R13 | 19 | 16 | 10 | 2 |
R14 | 15 | 11 | 10 | 3 |
R15 | 23 | 18 | 16 | 5 |
R16 | 17 | 16 | 12 | 3 |
R1 | 18 | 18 | 16 | 6 |
Finally, determine the selected threshold,
According to steps 9–11, process the reachable matrix, and, finally, obtain the factor set
Multilevel hierarchical hybrid model.
According to the DEMATEL analysis, the factors can be divided into cause factors and result factors. In the actual construction process, more attention should be paid to the causes, and effective control should be carried out to improve the safety of the construction site. As Table
The eight factors with high impact include cognitive defects R1, imperfect safety management systems R12, complex construction site environments R11, inadequate supervision of safety behaviours R15, illegal operations R3, inadequate safety disclosure R13, numerous large-scale equipment cross-operations R7, and failure of construction monitoring equipment R8. Therefore, management factors and human factors can be classified as factors that potentially affect construction safety by influencing other factors. The influenced degree indicates the comprehensive influence of other factors on the factor Ri. At the level of impact, errors in construction safety operational guidelines R17, illegal operations R3, inadequate safety protection measures R16, mechanical failure R6, inadequate supervision of safety behaviours R15, inadequate acceptance of key processes R14, failure of construction monitoring equipment R8, and inadequate safety disclosure R13 are the leaders, which shows that the security level can be improved by increasing influencers. Inadequate supervision of safety behaviours R15, inadequate safety disclosure R13, and failure of construction monitoring equipment R8 are all important influencing factors and affected factors; improving these factors will help improve the circulatory effects in the system.
Centrality reflects the importance of factors in the system. As shown in Table
In the multilevel hierarchical hybrid model, cognitive defects R1, imperfect safety management systems R12, and complex construction site environments R11 are deep factors, and unforeseeable natural factors R9, numerous large-scale equipment cross-operations R7, lack of maintenance and overhaul of equipment R5, inadequate supervision of safety behaviours R15, unreasonable equipment selection and layout R4, and poor physical or mental health R2 are middle factors. Mechanical failure R6, errors in construction safety operational guidelines R17, complex hydrogeological environments R10, inadequate acceptance of key processes R14, failure of construction monitoring equipment R8, inadequate safety disclosure R13, inadequate safety protection measures R16, and illegal operations R3 are surface factors.
Human-related factors, such as cognitive deficits R1, are deep factors; illegal operations R3 ranks second in centrality; and poor physical or mental health R2 belongs to the middle factors. These findings show that human-related factors play a pivotal role in construction safety risks and must be taken seriously. Factors related to management, such as imperfect safety management systems R12, are surface factors, and poor supervision of safety behaviours R15, inadequate safety disclosure R13, and inadequate safety protection measures R16 are ranked 2nd, 5th, and 7th in centrality, respectively. The system has an important impact.
According to Figure
As the above analysis shows, the human and management aspects need to focus on control. In addition, the following corresponding measures are proposed: Strengthen safety education. Personnel training and education can be provided through multimedia resources such as Virtual Reality (VR) equipment and on-site teaching simultaneously. On-site self-help and self-protection training should be conducted. Improve safety and protection measures. Safety protection measures should be taken for major hazards to avoid safety accidents. Scientifically arrange the construction sequence to reduce environmental uncertainty. Improve rules and regulations. Enterprises should improve the responsibility system for production safety and establish various rules and regulations for safety inspection and on-site management. Focus on the production management of key processes, master the construction status in advance, eliminate safety hazards, and formulate effective preventive measures. Develop a special safety plan. It is necessary to combine the characteristics of the project to prepare specialized, targeted, and operable special safety programmes and technical measures. Strengthen the implementation of safety supervision. Strengthen the enforcement capacity of safety supervision, and timely discover to solve safety problems in production. Master professional skills and knowledge. Ensure that workers master relevant practices. Check for related knowledge and skills, reward and update them, and retest regularly. Perform good safety technical clarification. The construction technology department and the safety risk management agency should cooperate to implement construction safety.
Construction accidents in high-speed railway station construction are the result of the joint actions of humans, machines, the environment, and management. Based on system safety theory, an index system of the influencing factors of construction safety is established with human-machine-environment management aspects, and the influencing mechanism among the influencing factors is analysed. The main conclusions are as follows: DEMATEL and ISM can be combined to analyse the relationship among various factors affecting construction safety. DEMATEL is used to analyse the centrality and causation of construction safety risk factors and to determine the key factors. According to the results, causation was divided into 6 causes and 11 results. ISM is used to divide construction safety risk factors into three hierarchical structures and to obtain a multilevel hierarchical hybrid model of influencing factors. The conclusion is that personnel factors and management factors need to be controlled with emphasis. Three main factors, six important factors, and seven direct factors are obtained by analyzing the risk factors affecting the construction safety of high-speed railway station. The factors that affect the construction safety of high-speed railway stations are extraordinarily complex and have classification modes. In security risk control, we should not only pay attention to the proximal cause but also start from the overall situation and consider all factors comprehensively.
The data used to support the findings of this study are included within the article.
The authors declare that there are no conflicts of interest regarding the publication of this paper.
The authors acknowledge the financial support of Science and Technology Project of Ningbo Transportation Bureau (Grant no. 202007) and Science and Technology Research and Development Project of China Railway Construction Group Co., Ltd. (2020-12).