A Dynamic Risk Analysis Model Based on Workplace Ergonomics and Demographic-Cognitive Characteristics of Workers
1
Department of Occupational Health and Safety, Istanbul Commerce University, Istanbul 34854, Türkiye
2
Department of Mechatronics Engineering, Istanbul Commerce University, Istanbul 34854, Türkiye
*
Author to whom correspondence should be addressed.
Sustainability 2023, 15(5), 4553; https://doi.org/10.3390/su15054553
Submission received: 21 January 2023
/
Revised: 25 February 2023
/
Accepted: 27 February 2023
/
Published: 3 March 2023
(This article belongs to the Section Sustainable Engineering and Science)
Abstract
:Background: This study aimed to examine the effect of perception, attention, and sleep levels on the number of occupational accidents and near-misses in the mining and metal sectors. Methods: The data were collected from 53 employees in the mining (n = 30) and metal (n = 23) sectors in 2021 from a mining and metal company. The study collected the following data from the sample: demographic information forms, previous accident and previous near-miss histories, Pittsburgh Sleep Quality (PSQI) scale, pulse, sleep levels, and attention tests. Results: Having an education at primary school and below (B = 0.235; p < 0.05), and having an education at the high school level (B = 0.710; p < 0.01), being single (B = −0.291; p < 0.01), time working in the department (B = 0.027; p < 0.05), time working in the company (B = −0.034; p < 0.05), and the number of near-misses (B = 0.354; p < 0.01), had a significant impact on accidents. Having an education in primary school or below (B = −1.532; p < 0.01), not having had an accident (B = −3.654; p < 0.01), age (B = 0.074; p < 0.01), correct score (B = 0.014; p < 0.01), incorrect time (B = 0.228; p < 0.01) and unanswered score averages (B = −0.029; p < 0.01) had a significant impact on near-misses. Conclusion: Education, the working year, and working time had significant effects on workplace accidents.
1. Introduction
Occupational accidents are a crucial problem that affects both the working levels of individuals and the use of human capital in working environments. They reduce the quality of people’s lives [1,2]. Although many studies have been conducted on occupational accidents from the past to the present, occupational accidents continue in a way that can give undesirably severe results, especially in labor-intensive sectors such as mining and metal [3].
According to the literature, occupational accidents occur from successive causes. The next one does not take place unless one of these occurs, and there is no accident or injury until the sequence is completed. An unplanned and unknown potentially damaging event should also occur [4,5]. In this case, an accident must be present for injury or damage to occur, that is, to occur with all elements of the accident [6,7].
According to Article 3 of the Occupational Health and Safety Law No. 6331, which is an occupational safety regulation that the Turkish Government adopted on 20 June 2020, an occupational accident is an event that occurs in the workplace or, due to the performance of the work, causes death or disrupts mental or physical integrity. However, according to Article 13 of Law No. 5510 of the Ministry of Labour and Social Security of the Republic of Turkiye, the concept of the occupational accident is the accident that the insured person suffers while in the workplace if he/she works independently on his/her account [8].
Conceptually, accident prevention activities mean controlling the performance of the workforce, the performance of tools, devices, and machines, and the physical environment. The purpose of the controlled study included both the prevention and correction of unsafe situations and events. Preventing accidents is vital for any industrial organization [9,10]. If the accident is not prevented properly, it will cause injury, death, and/or material losses and adversely affect the business [7].
While there are diverging views on the occurrence of occupational accidents based on different studies on the weight relationships of the causes of occupational accidents, it would be correct to accept that 80% of occupational accidents are human-originated, 18% are physical and mechanical, and 2% are environmental and unexpected events. This generalization reveals that preventive measures can be taken in approximately 98% of occupational accidents [11].
Occupational accidents affect not only workers but also employers, workers’ families, society, and the country and constitute considerable costs. They also create a lot of costs, such as absenteeism, loss of income, job loss, and loss of time. According to the International Labor Organization, the economic burden of poor occupational health and safety practices is approximately 4% of the world’s gross domestic product every year [12,13,14]. For these reasons, it is crucial to spread the occupational safety culture and ensure occupational safety management to prevent occupational accidents in the workplace, reduce their effects, and create healthier and safer working environments [8].
Mining and metal industries are among the production areas where workplace accidents are most common [15,16,17]. The fact that the metal and mining sectors are interconnected in terms of raw materials and production also leads to companies operating in both sectors. Therefore, these companies have more accident factors than other companies in terms of occupational health and safety.
Although some studies have been conducted on the effect of perception, attention, and sleep levels on occupational accidents, there are not enough studies on dynamic risk modeling and the examination of these variables together in both the mining and metal sectors. For this reason, this study aims to examine the effect of perception, attention, and sleep levels on the number of occupational accidents and near-misses in the mining and metal sectors [18,19,20].
2. Method
The research was designed in a descriptive screening model, and the digital measurement process took biometric measurements through the survey method as the data collection method.
2.1. Population Sample
The research population consists of mining and metal sector employees, and the sample consists of 53 employees from the mining (n = 30) and metal (n = 23) sectors working in 2021 in a mining and metal company which is one of the companies having two working places in two cities: in Kastamonu and Ankara at Turkiye.
There has not been enough work done on a company operating in two different and related sectors. Therefore, a power analysis was conducted for sample size calculation. Based on Silva et al. [21], the effect size was calculated from Gpower 3.1.9.2 as 0.4444444, and the minimum sample size was calculated as 35 with a 1.3069516 critical T value.
In Turkiye and in many regions of the world, the mining and metal sector are close to each other, and the same company could operate both in the metal sector with different firms. In fact, in this study, the aim was to reveal the effect of the sector by examining the same company operating in two different sectors. Concentrating on only one of the sectors that the company operates in would not fully express the effects of job safety risk factors for that company.
There were no accidents or near-misses in the places where the study was carried out during the research process. Findings about accidents and near-misses were obtained based on the past experiences of the employees participating in the research.
The collection of research data was done by the researcher himself. After obtaining the necessary permissions, the research data were collected from the employees at regular intervals and transferred to the computer by using the cluster sampling method and the whole count sampling method together by creating a data collection chart.
2.2. Data Collection Tools
The data collection tools were the demographic information form, accident and near-miss histories, the Pittsburgh Sleep Quality (PSQI) scale, pulse, sleep duration and quality, and perception tests for cognition.
The parameters such as pulse, sleep, REM, and active sleep were collected from the employees by constant observation through a measurement tool worn by the employee. While the measurement values were taken, all measurements were made and recorded by the researcher in specific areas.
The Pittsburgh Sleep Quality (PSQI) scale is a validated and reliable scale measuring sleep quality with 19 items developed by Buysse et al. [22] and validated in Turkish form by Agargun et al. [23]. PSQI analyzes seven components: subjective sleep quality, sleep onset latency, sleep duration, sleep efficiency, sleep disturbance, use of hypnotic medication, and daytime dysfunction. It has been revealed that the sleep quality level taken from the scale is related to the deterioration in psychomotor and neuropsychological functions. Both clinical and psychological findings reveal that the scale adequately represents sleep quality in psychometric terms [23]. High scores indicate low sleep quality. An accident indicates any occupational accident history related to the current job, whereas a near-miss indicates an accident that could have happened soon but did not happen.
2.3. Research Hypothesis
The study had the following alternative hypotheses:
H1.
Risk factors for the working and workplace where production is processed environment in the mining and metal sectors, and accident rates have a statistically significant effect on dynamic risk analysis.
H2.
Risk factors for the working and working environment in the mining and metal sectors and near-miss rates have a statistically significant effect on dynamic risk analysis.
2.4. Statistical Method
Fischer’s exact test was used in the difference analysis of nominal and ordinal data in the study. The Kolmogorov–Smirnov test checked whether the data fit a normal distribution. The Mann–Whitney U test determined the difference between the two groups. Spearman’s rho correlation and generalized linear model (GLM) analysis was performed in relational screening analysis. All analyses were in SPSS 17.0 for Windows, with a 95% confidence interval and p = 0.05 as the threshold for statistical significance.
Since dependent variables, near-misses, and accident parameters were nonparametric categorical parameters, Spearman’s rho correlation analysis was used for a univariate relationship, and generalized linear model (GLM) analysis was used for a multivariate relationship in order to evaluate the effects of factors together. The behavior of dependent variables was dichotomous, or dummy variables and the statistical strategy was selected as a generalized logit model [24].
3. Results
The distribution of some demographic and professional characteristics of the participants according to their companies is given in Table 1.
Age, years of employment in the sector, years of employment in the company, and mean mild sleep were statistically significantly higher in the metal sector company (p < 0.05). Pulse, sleep, REM, and sleep parameters were measured by researchers using diagnostic tools. The year working in the department is higher in the company in the mining sector (p < 0.05). The education variable between the two companies also differs statistically significantly (p < 0.05).
The mean values of the perception, attention and PSQI parameters of the participants and the results of the difference analysis are in Table 2.
The average of correct questions, time, and scores, incorrect time, unanswered time, and the total score is higher in the company in the metal sector, while the average of unanswered questions and scores and PSQI is statistically significantly higher in the company in the mining sector (p < 0.05).
Spearman’s rho correlation analysis results for checking the relationship between the number of accidents and demographic characteristics, job characteristics, and research parameters are in Table 3.
According to the results of Spearman’s rho correlation analysis, a statistically significant and positive relationship was present between the number of accidents and the following parameters: education level (r = 0.060; p < 0.05), years working in the department (r = 0.139; p < 0.01), years working in the company (r = 0.109; p < 0.01), being married (r = 0.280; p < 0.01), having a near-miss (r = 0.477; p < 0.01), the number of near-misses (r = 0.600; p < 0.01), and the time spent on unanswered questions in the attention test (r = 0.077; p < 0.01). Again, a statistically significant and negative relationship was present between the number of accidents and the number of incorrect questions (r = −0.078; p < 0.01) and the incorrect score (r = −0.078; p < 0.01).
The results of the logit model (generalized linear model) analysis conducted to examine the effect of the parameters with significant correlation with the number of previous accidents on the number of previous near-misses are in Table 4.
The logit model results showed that the following conditions were significantly effective on the number of accidents: having an education at primary school and below (B = 0.235; p < 0.05), having an education at high school level (B = 0.710; p < 0.01), being single (B = −0.291; p < 0.01), time working in the department (B = 0.027; p < 0.05), time working in the company (B = −0.034; p < 0.05), and the number of near-misses (B = 0.354; p < 0.01). While being single and increasing the number of working years in the company has a decreasing effect on the number of accidents, graduating from primary school and below and high school, increasing the number of years working in the department and increasing the number of near-misses increase the number of accidents. These results show that high school graduates are more effective in the field of work and cause fewer accidents; the number of accidents increases as the years working in the department increases, and the number of accidents decreases as the time working in the company increases.
Spearman’s rho correlation analysis results for the relationship between the number of previous near-misses and demographic characteristics, professional characteristics, and research parameters are given in Table 5.
According to Table 5, a statistically significant and positive relationship was present between the number of near-misses and the following conditions: being male (r = 0.078; p < 0.01), education (r = 0.174; p < 0.01), being married (r = 0.186; p < 0.01), having had an accident before (r = 0.587; p < 0.01), the number of correct questions (r = 0.131; p < 0.01), the correct number score (r = 0.132; p < 0.01), and the incorrect time (r = −0.113; p < 0.01). Again, a statistically significant and negative relationship was present between the number of near-misses and the following parameters: age (r = −0.070; p < 0.05), correct time (r = −0.073; p < 0.05), number of unanswered questions (r = −0.123; p < 0.01), and unanswered score (r = −0.122; p < 0.01).
The results of the logit model (generalized linear model) analysis conducted to examine the effect of the parameters with significant correlation with the number of near-misses on the number of accidents are in Table 6.
The following parameters have a statistically significant effect on the number of near-misses: having an education in primary school or below (B = −1.532; p < 0.01), not having had an accident (B = −3.654; p < 0.01), age (B = 0.074; p < 0.01), correct score (B = 0.014; p < 0.01), incorrect time (B = 0.228; p < 0.01), and unanswered score means (B = −0.029; p < 0.01). When the beta coefficients are examined, being a woman, having an education in primary school and below, not having an accident, being single, and having a high mean score reduce the number of near-misses.
4. Discussion
In this study, the extent to which the accident and near-accident cases of the employees in the mining and metal sectors are suitable for dynamic risk modeling methods and the effects of these parameters were analyzed with a single and multivariate method approach.
Studies on occupational health and safety from the past to the present report that the effect of demographic characteristics on occupational accidents varies according to the department, sector, job status, and company structure [25,26,27]. In these studies, although the mining and metal sectors are not sufficiently compared together, both sectors have been subject to research and examination because of their labor-intensive structure [11,28,29,30]. According to the results of this study, the following parameters were significantly statistically higher in the metal sector: the average age, years of work in the sector, years of work in the company, and mild sleep (p < 0.05). The number of working years in the department is higher in the mining sector (p < 0.05). Sex and education level variables between the two companies in two different sectors also significantly differ statistically (p < 0.05).
Studies in the mining and metal sectors in the literature reveal that demographic characteristics in the dimension of univariate analysis are statistically significantly related to accidents in both sectors [31,32,33,34]. In this study, a statistically significant and positive relationship was present between the number of accidents and the following demographic characteristics: education (r = 0.060; p < 0.05), years working in the department (r = 0.139), years working in the company (r = 0.109), being married (r = 0.280), having a near-miss (r = 0.477), the number of near-misses (r = 0.600), and the time spent on the unanswered questions in the attention test (r = 0.077). A statistically significant and negative relationship was present between the number of accidents and the number of incorrect questions (r = −0.078), and the incorrect score (r = −0.078). Correlation coefficients indicate that the most effective parameters are the number of near-misses and near-misses, followed by being married.
Multivariate analysis methods give more effective results in modeling and explaining the real situation than univariate analysis [35]. The results of the logit model in this study showed that the following parameters have statistically significant effects on the number of accidents: having an education at primary school and below, having an education at the high school level, being single, time working in the department, time working in the company, and the number of near-misses. There are fewer studies conducted on near-misses than on accident cases. Additionally, the number of studies evaluating the mining and metal industries together and discussing the accident rates together with the near-misses is quite low. A statistically significant and positive relationship was present between the number of near-misses and the following characteristics: being male (r = 0.078), education (r = 0.174), being married (r = 0.186), having had an accident before (r = 0.587), the number of correct questions (r = 0.131), the correct number score (r = 0.132), and the incorrect time (r = −0.113). A statistically significant and negative relationship was present between the number of near-misses and the following characteristics: age (r = −0.070), correct time (r = −0.073), number of unanswered questions (r = −0.123), and unanswered score (r = −0.122). According to the correlation coefficients, having an accident has the highest impact, followed by being married, education, the number and score of corrects, and unanswered and incorrect numbers.
According to the results of the multivariate analysis, the following characteristics have a statistically significant effect on near-misses: having an education in primary school and below, not having had an accident, age, mean correct score, mean incorrect time, and mean unanswered score. This finding shows that the effect of the tests performed is significant in multivariate analyses. Therefore, unlike accident statistics, attention test results in near-miss statistics gain more meaning.
In the selection phase of the employees, definitions and limits, such as recruitment criteria, education level, and experience, are sometimes not enough to prevent accidents from occurring. In addition, the creation of a culture of occupational safety and the social life and health of employees directly affects occupational safety. Therefore, in order to increase the safety of employees, it is very important to observe them and to detect situations where they are not suitable for work conditions. In this way, it is important to allow them to focus on their day-to-day work and avoid dangerous behavior.
The mining and metal sectors are critical and high-risk sectors due to both the working environment and the heavy workload. The training, warnings, and necessary precautions are sometimes insufficient. For this reason, the use of developing technological opportunities is an important proactive approach to preventing accidents. In this way, the employee’s suitability for daily work can be measured, and situations in which he is not even aware of himself can be determined.
Limitations, Benefits and Future Directions
The conducted research has twofold benefits: (1) İn two related sectors (metal and mining) operated by one company, occupational health and safety studies are conducted. (2) The research gives detailed information on risk factors that are important health and safety problems in these sectors.
Occupational burnout in accidents at work is an important limitation of the study. Since occupation takes an important time duration in daily life and having an accident in the workplace causes a decrease in the performance of workers, it also causes burnout, and this result may reflect results. Further research on the effects of occupational burnout will help to increase health and safety measurements.
Allocating longer time for data collection will reduce errors and increase consistency. The study can be improved further by collecting more diverse data via utilizing Internet of Things (IoT) devices and by comparing these data through accurate assessments. In addition, adding other sectors with other high accident rates and conducting them with larger participants will enable more sensitive and large-scale assessments to be made. This will allow seeing overlooked points. Utilizing artificial intelligence (AI) may help to make faster proactive decisions and reduce the occupational safety risk factor, even before near-misses. In this way, important results will be obtained for the detection of occupational accidents before they occur.
5. Conclusions
According to the results of the research, high school graduates are more skillful in preventing accidents in the workplace and cause fewer accidents, and the number of accidents increases as the years working in the department increase. On the other hand, it shows that as the time working in the company increases, accidents decrease. This situation needs to be examined in terms of whether the positive relationship between the increase in time working in the mining and metal sectors and the number of accidents is related to occupational burnout and fatigue or whether the existence of more working years means the possibility of more accidents. Therefore, work can be developed in this area with multicentered and larger samples. In this respect, the study constitutes a reference for further studies.
The following characteristics reduce the number of near-misses: being a woman, having an education in primary school and below, not having an accident, being single, and having a high mean score. These results show that the effect of attention and employees’ perception is more crucial in the near-miss part of the accident. In addition, the relationship between accidents and near-misses reveals that if near-miss cases are not prevented, they are more likely to turn into accidents.
As a result, when the accidents and near-miss cases in the mining and metal sectors are considered in terms of dynamic risk modeling, it is seen that the effect of the variables related to the working environment and the employee is more critical in the cases with near-misses. This finding shows that occupational accidents in the mining and metal sectors can be prevented at the near-miss stage, and attention test results can be used effectively.
Author Contributions
Conceptualization, A.M.Ö.; Methodology, validation, formal analysis, A.T. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
The study was conducted in accordance with the Declaration of Helsinki and approved by the Istanbul Commerce University Ethics Committee (protocol code E-65836846-044-269359, 8 March 2021).
Informed Consent Statement
Informed consent was obtained from all subjects involved in the study.
Data Availability Statement
Data are available only for this research, according to ethical approval.
Conflicts of Interest
The authors declare no conflict of interest.
References
- Rikhotso, O.; Morodi, T.J.; Masekameni, D.M. The Extent of Occupational Health Hazard Impact on Workers: Documentary Evidence from National Occupational Disease Statistics and Selected South African Companies’ Voluntary Corporate Social Responsibility Disclosures. Sustainability 2022, 14, 10464. [Google Scholar] [CrossRef]
- Ji, Z.; Pons, D.; Su, Z.; Lyu, Z.; Pearse, J. Integrating Occupational Health and Safety Risk and Production Economics for Sustainable SME Growth. Sustainability 2022, 14, 14565. [Google Scholar] [CrossRef]
- Ergör, O.A.; Demiral, Y.; Piyal, Y.B. A Significant Outcome of Work Life: Occupational Accidents in a Developing Country, Turkey. J. Occup. Health 2003, 45, 74–80. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Oldenburg, M.; Baur, X.; Schlaich, C. Occupational Risks and Challenges of Seafaring. J. Occup. Health 2010, 52, 249–256. [Google Scholar] [CrossRef] [Green Version]
- Swat, K. Monitoring of Accidents and Risk Events in Industrial Plants. J. Occup. Health 2006, 39, 100–104. [Google Scholar] [CrossRef]
- Kortum, E. Perceptions of Psychosocial Hazards, Work-related Stress and Workplace Priority Risks in Developing Countries. J. Occup. Health 2011, 53, 144–155. [Google Scholar] [CrossRef] [Green Version]
- Arık, B.; Akçın, N.A. Iş Kazalarının Önlenmesi Ve Iş Güvenligi Analiz Tekniginin Uygulanması. In Proceedings of the Türkiye 13 Kömür Kongresi Bildiriler Kitabı, Zonguldak, Türkiye, 29–31 May 2002. [Google Scholar]
- Erginel, N.; Toptancı, Ş. Iş Kazası Verilerinin Olasılık Dagılımları ile Modelenmesi. Mühendislik Bilim. Tasarım Derg. 2016, 5, 201–212. [Google Scholar] [CrossRef]
- Kahraman, E.; Akay, Ö.; Kılıç, A.M. Investigation into the relationship between fatal work accidents, national income, and employment rate in developed and developing countries. J. Occup. Health 2019, 61, 213–218. [Google Scholar] [CrossRef]
- Thepaksorn, P.; Thongjerm, S.; Incaroen, S.; Siriwong, W.; Harada, K.; Koizumi, A. Job safety analysis and hazard identification for work accident prevention in para rubber wood sawmills in southern Thailand. J. Occup. Health 2017, 59, 542–551. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Camkurt, M.Z. Çalışanların Kişisel Özelliklerinin Iş Kazalarının Meydana Gelmesi Üzerindeki Etkisi. TÜHIS Hukuku Iktisat Derg. 2013, 24, 70–101. [Google Scholar]
- Kajiki, S.; Mori, K.; Kobayashi, Y.; Hiraoka, K.; Fukai, N.; Uehara, M.; Adi, N.P.; Nakanishi, S. Developing a global occupational health and safety management system model for Japanese companies. J. Occup. Health 2019, 62, e12081. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Nakata, A.; Ikeda, T.; Takahashi, M.; Haratani, T.; Hojou, M.; Swanson, N.G.; Fujioka, Y.; Araki, S. The Prevalence and Correlates of Occupational Injuries in Small-Scale Manufacturing Enterprises. J. Occup. Health 2006, 48, 366–376. [Google Scholar] [CrossRef] [Green Version]
- Okubo, T. Recent State and Future Scope of Occupational Health in Japan. J. Occup. Health 2006, 40, 161–167. [Google Scholar] [CrossRef]
- Sanmiquel, L.; Bascompta, M.; Rossell, J.; Anticoi, H.; Guash, E. Analysis of Occupational Accidents in Underground and Surface Mining in Spain Using Data-Mining Techniques. Int. J. Environ. Res. Public Health 2018, 15, 462. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Zhou, L.; Cao, Q.; Yu, K.; Wang, L.; Wang, H. Research on Occupational Safety, Health Management and Risk Control Technology in Coal Mines. Int. J. Environ. Res. Public Health 2018, 15, 868. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Muthelo, L.; Mothiba, T.M.; Malema, N.R.; Mbombi, M.O.; Mphekgwana, P.M. Exploring Occupational Health and Safety Standards Compliance in the South African Mining Industry, Limpopo Province, Using Principal Component Analysis. Int. J. Environ. Res. Public Health 2022, 19, 10241. [Google Scholar] [CrossRef]
- Niu, L.; Zhao, R. The Effect of Safety Attitudes on Coal Miners’ Human Errors: A Moderated Mediation Model. Sustainability 2022, 14, 9917. [Google Scholar] [CrossRef]
- Małysa, T.; Gajdzik, B. Predictive Models of Accidents at Work in the Steel Sector as a Framework for Sustainable Safety. Energies 2020, 14, 129. [Google Scholar] [CrossRef]
- Zhang, S.; Hua, X.; Huang, G.; Shi, X.; Li, D. What Influences Miners’ Safety Risk Perception? Int. J. Environ. Res. Public Health 2022, 19, 3817. [Google Scholar] [CrossRef]
- Silva, A.F.; Dalri, R.C.M.B.; Eckeli, A.L.; Sousa-Uva, A.; Mendes, A.C.; Robazzi, M.L.C.C. Sleep quality, personal and work variables and life habits of hospital nurses. Rev. Latino-Am. Enferm. 2022, 30, e3538. [Google Scholar] [CrossRef]
- Buysse, D.J.; Reynolds, C.F.; Monk, T.H. The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research. Psychiatry Res. 1989, 28, 193–213. [Google Scholar] [CrossRef]
- Agargun, M.Y.; Kara, H.; Anlar, O. Pittsburgh Uyku Kalitesi İndeksi’nin Geçerligi ve Güvenirligi. Turk. Psikiyatr. Derg. 1996, 7, 107–111. [Google Scholar]
- Arias-Castro, E. Principles of Statistical Analysis: Learning from Randomized Experiments (Institute of Mathematical Statistics Textbooks) Kindle Edition; Cambridge Press: Cambridge, UK, 2022. [Google Scholar]
- Alves, A.M.S.; Gonçalves Filho, C.; Santos, N.M.; Souki, G.Q. Factors influencing occupational accidents: A multidimensional analysis in the electricity sector. Gestão Produção 2020, 27, e4609. [Google Scholar] [CrossRef]
- Baradan, S.; Akboga, Ö.; Çetinkaya, U.; Usmen, M.A. Ege Bölgesindeki Inşaat Iş Kazalarının Sıklık ve Çapraz Tablolama Analizleri. Tek. Dergi 2016, 27, 7345–7370. [Google Scholar]
- Ilsoon, S.; Jun-Byoung, O.H.; Kwan Hyung, Y.I. Workers’ Compensation Insurance and Occupational Injuries. Saf. Health Work. 2011, 2, 148–157. [Google Scholar]
- Hoła, B.; Nowobilski, T. Analysis of the influence of socioeconomic factors on occupational safety in the construction industry. Sustainability 2019, 111, 4469. [Google Scholar] [CrossRef] [Green Version]
- Amponsah-Tawiah, K.; Mensah, J. Occupational health and safety and organizational commitment: Evidence from the Ghanaian mining industry. Saf. Health Work. 2016, 7, 225–230. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Sadullah, Ö. İş kazası istatistikleri ve olay oranı. I.Ü. Işletme Fakültesi Derg. 1997, 26, 109–113. [Google Scholar]
- Eskandari, D.; Jafari, M.J.; Mehrabi, Y.; Kian, M.P.; Charkhand, H.; Mirghotbi, M. A qualitative study on organizational factors affecting occupational accidents. Iran. J. Public Health 2017, 46, 380–388. [Google Scholar]
- Yi, K.H.; Lee, S.S. A policy intervention study to identify high-risk groups to prevent industrial accidents in Republic of Korea. Saf. Health Work. 2016, 7, 213–217. [Google Scholar] [CrossRef] [Green Version]
- Khanzode, V.V.; Maiti, J.; Ray, P.K. A methodology for evaluation and monitoring of recurring hazards in underground coal mining. Saf. Sci. 2011, 49, 1172–1179. [Google Scholar] [CrossRef]
- Bernackie, E.J. Factors influencing the costs of workers’ compensation. Clin. Occup. Environ. Med. 2004, 4, 249–257. [Google Scholar] [CrossRef] [PubMed]
- Liu, H.-T.; Tsai, Y.-L. A fuzzy risk assessment approach for occupational hazards in the construction industry. Saf. Sci. 2012, 50, 1067–1078. [Google Scholar] [CrossRef]
Table 1.
Distribution of some demographic and professional characteristics of the participants according to their companies.
Table 1.
Distribution of some demographic and professional characteristics of the participants according to their companies.
| Company | Mine (N = 30) | Metal (N = 23) | Total (N = 53) | Test Value | p-Value |
|---|---|---|---|---|---|
| Age (Mean ± SD) | 31.37 ± 4.85 | 38.61 ± 7.48 | 34.51 ± 7.11 | 69,457.500 | 0.000 *a |
| Education, n (%) | |||||
| Primary school or below | 1 (3.3) | 9 (39.1) | 10 (18.9) | ||
| High school | 6 (20.0) | 4 (17.4) | 10 (18.9) | 12.099 | 0.002 c |
| University and above | 23 (76.7) | 10 (43.5) | 33 (62.3) | ||
| Sector year, mean ± SD | 6.02 ± 4.57 | 14.61 ± 7.39 | 9.75 ± 7.32 | 47,407.500 | 0.000 *a |
| Department year, mean ± SD | 5.18 ± 4.01 | 5.17 ± 4.40 | 5.18 ± 4.18 | 132,961.500 | 0.000 *a |
| Company year, mean ± SD | 3.48 ± 2.18 | 5.17 ± 4.40 | 4.22 ± 0.3 | 141,120.000 | 0.035 a |
| Marital status, n (%) | |||||
| Single | 21 (70.0) | 9 (39.1) | 30 (56.6) | (50.51) | 0.024 b |
| Married | 9 (30.0) | 14 (60.9) | 23 (43.4) | ||
| Previous accident, n (%) | |||||
| No | 25 (83.3) | 20 (87.0) | 45 (84.9) | 0.133 | 0.514 b |
| Yes | 5 (16.7) | 3 (13.0) | 8 (15.1) | ||
| Number of previous accident experiences | 0.27 ± 0.77 | 0.57 ± 2.06 | 0.40 ± 1.48 | 148,176.000 | 0.230 a |
| Previous near-miss, n (%) | |||||
| No | 23 (76.7) | 16 (69.6) | 39 (73.6) | 0.338 | 0.393 b |
| Yes | 7 (23.3) | 7 (30.4) | 14 (26.4) | ||
| Previous near-misses, mean ± SD | 0.77 ± 1.57 | 1.83 ± 3.32 | 1.23 ± 2.54 | 145,089.000 | 0.087 a |
| Instantaneous pulse, mean ± SD | 89.90 ± 17.90 | 91.19 ± 18.27 | 90.46 ± 18.07 | 145,821.000 | 0.234 a |
| Daily pulse, mean ± SD | 89.34 ± 17.08 | 91.07 ± 17.75 | 90.09 ± 17.39 | 143,452.000 | 0.102 a |
| Deep sleep, mean ± SD | 1.54 ± 0.87 | 1.47 ± 0.86 | 1.51 ± 0.87 | 145,277.500 | 0.196 a |
| REM, mean ± SD | 0.92 ± 0.54 | 0.92 ± 0.51 | 0.92 ± 0.53 | 151,962.500 | 0.973 a |
| Mild sleep, mean ± SD | 2.74 ± 1.48 | 3.01 ± 1.61 | 2.86 ± 1.54 | 138,274.500 | 0.009 a |
| Active time, mean ± SD | 0.75 ± 0.44 | 0.74 ± 0.43 | 0.74 ± 0.44 | 151,206.000 | 0.860 a |
a Mann–Whitney U Test, b Fischer’s Exact Test, c Chi-square similarity ratio, SD: Standard Deviation, * bold values are statistically significant.
Table 2.
The mean values of the perception, attention, and PSQI parameters of the participants and the results of the difference analysis.
Table 2.
The mean values of the perception, attention, and PSQI parameters of the participants and the results of the difference analysis.
| Mean ± SD | Mining (N = 30) | Metal (N= 23) | Total (N = 53) | Value of the Test | p-Value a |
|---|---|---|---|---|---|
| Correct answer | 10.75 ± 2.92 | 12.52 ± 3.04 | 11.52 ± 3.10 | 104,253.000 | 0.000 |
| Correct time | 2.88 ± 0.66 | 2.99 ± 0.42 | 2.93 ± 0.57 | 127,825.000 | 0.000 |
| Correct score | 51.21 ± 13.63 | 59.63 ± 14.47 | 54.86 ± 14.60 | 103,747.000 | 0.000 |
| Incorrect answer | 5.79 ± 3.33 | 5.78 ± 2.48 | 5.79 ± 2.99 | 149,581.500 | 0.628 |
| Incorrect time | 2.93 ± 1.21 | 3.41 ± 0.68 | 3.14 ± 1.04 | 108,080.500 | 0.000 |
| Incorrect score | 27.59 ± 15.87 | 27.52 ± 11.83 | 27.56 ± 14.25 | 149,581.500 | 0.628 |
| Unanswered question | 4.46 ± 3.32 | 2.69 ± 2.07 | 3.69 ± 2.98 | 105,635.000 | 0.000 |
| Unanswered time | 4.54 ± 1.45 | 4.91 ± 0.68 | 4.70 ± 1.19 | 140,973.000 | 0.000 |
| Unanswered score | 21.22 ± 15.82 | 12.80 ± 9.88 | 17.56 ± 14.19 | 105,957.500 | 0.000 |
| Total score | 0.49 ± 0.10 | 0.54 ± 0.05 | 0.51 ± 0.08 | 103,534.500 | 0.000 |
| ST | 9.13 ± 2.95 | 9.18 ± 1.87 | 9.16 ± 2.54 | 150,892.000 | 0.812 |
| PSQI | 13.78 ± 2.18 | 13.18 ± 2.86 | 13.52 ± 2.51 | 130,301.500 | 0.000 |
a Mann–Whitney U Test, SD: Standard Deviation.
Table 3.
Spearman’s rho correlation analysis results for the relationship between the number of accidents and demographic characteristics, job characteristics, and research parameters.
Table 3.
Spearman’s rho correlation analysis results for the relationship between the number of accidents and demographic characteristics, job characteristics, and research parameters.
| Number of Previous Accidents | r | p |
|---|---|---|
| Education level | 0.060 * | 0.045 |
| Age | −0.017 | 0.572 |
| Sector years | −0.052 | 0.084 |
| Department years | 0.139 ** | 0.000 |
| Company years | 0.109 ** | 0.000 |
| Marital status | 0.280 ** | 0.000 |
| Previous near-miss | 0.477 ** | 0.000 |
| Number of previous near-miss incidents | 0.600 ** | 0.000 |
| Instantaneous pulse | −0.018 | 0.540 |
| Daily pulse | 0.036 | 0.233 |
| Deep sleep | −0.020 | 0.507 |
| REM | −0.019 | 0.527 |
| Mild sleep | −0.009 | 0.773 |
| Active time | −0.046 | 0.122 |
| Company | −0.036 | 0.230 |
| Measurement | 0.000 | 10.000 |
| Correct answers | 0.001 | 0.963 |
| Correct time | −0.054 | 0.071 |
| Correct score | 0.002 | 0.940 |
| Incorrect answers | −0.078 ** | 0.009 |
| Incorrect time | −0.009 | 0.766 |
| Incorrect score | −0.078 ** | 0.009 |
| Unanswered question | 0.037 | 0.220 |
| Unanswered time | 0.077 | 0.011 |
| Unanswered score | 0.037 | 0.212 |
| Total score | 0.004 | 0.896 |
| ST | 0.038 | 0.211 |
| PSQI | −0.036 | 0.225 |
* p < 0.05; ** p < 0.01.
Table 4.
The results of the logit model (generalized linear model) analysis conducted to examine the effect of the parameters with a significant correlation with the number of previous accidents on the number of accidents at the multivariate level.
Table 4.
The results of the logit model (generalized linear model) analysis conducted to examine the effect of the parameters with a significant correlation with the number of previous accidents on the number of accidents at the multivariate level.
| Parameter Name | B | Std. Error | 95% Wald Confidence Interval | Hypothesis Tests | ||
|---|---|---|---|---|---|---|
| Minimum | Maximum | Wald x2 | p | |||
| (Intercept) | −0.042 | 0.1201 | −0.277 | 0.193 | 0.122 | 0.727 |
| [Education = primary school and below] | 0.235 | 0.0938 | 0.051 | 0.419 | 6.272 | 0.012 |
| [Education = high school] | 0.710 | 0.0935 | 0.527 | 0.893 | 57.632 | 0.000 |
| [Education = university] | 0 | - | - | - | - | - |
| Marital status: single | −0.291 | 0.0734 | −0.435 | −0.147 | 15.730 | 0.000 |
| Marital status: married | 0 | - | - | - | - | - |
| Department years | 0.027 | 0.0115 | 0.004 | 0.049 | 5.389 | 0.020 |
| Company years | −0.034 | 0.0139 | −0.062 | −0.007 | 6.141 | 0.013 |
| Number of near-miss incidents | 0.354 | 0.0140 | 0.326 | 0.381 | 636.794 | 0.000 |
| Incorrect score | −5.769 × 10−5 | 0.0024 | −0.005 | 0.005 | 0.001 | 0.981 |
| Scale | 1.226 | −0.0520 | 1.129 | 1.333 | ||
Log-likelihood: −1692.844; Akaike Information Criterion (AIC): 3403.688.
Table 5.
Spearman’s rho correlation analysis results for the relationship between the number of previous near-misses and demographic characteristics, professional characteristics, and research parameters.
Table 5.
Spearman’s rho correlation analysis results for the relationship between the number of previous near-misses and demographic characteristics, professional characteristics, and research parameters.
| Number of Previous Near-Misses | r | p |
|---|---|---|
| Education level | 0.174 ** | 0.000 |
| Age | −0.070 * | 0.020 |
| Sector years | 0.050 | 0.094 |
| Department years | −0.059 | 0.050 |
| Company years | −0.025 | 0.403 |
| Marital status | 0.186 ** | 0.000 |
| Previous accident | 0.587 ** | 0.000 |
| Instantaneous pulse | −0.004 | 0.897 |
| Daily pulse | 0.017 | 0.561 |
| Deep sleep | 0.015 | 0.624 |
| REM | −0.001 | 0.980 |
| Mild sleep | 0.015 | 0.627 |
| Active time | −0.040 | 0.187 |
| Company | 0.051 | 0.087 |
| Measurement | 0.000 | 1.000 |
| Correct question. | 0.131 ** | 0.000 |
| Correct time | −0.073 * | 0.014 |
| Correct score | 0.132 | 0.000 |
| Incorrect question | −0.033 | 0.274 |
| Incorrect time | 0.113 ** | 0.000 |
| Incorrect score | −0.033 | 0.274 |
| Unanswered question | −0.123 ** | 0.000 |
| Unanswered time | 0.022 | 0.464 |
| Unanswered Score | −0.122 ** | 0.000 |
| Total Score | 0.050 | 0.093 |
| Week | 0.000 | 1.000 |
| ST | 0.047 | 0.116 |
| PSQI | 0.047 | 0.113 |
* p < 0.05; ** p < 0.01.
Table 6.
The results of the logit model (generalized linear model) analysis to examine the effect of the parameters with significant correlation with the number of previous near-misses on the number of previous accidents at the multivariate level.
Table 6.
The results of the logit model (generalized linear model) analysis to examine the effect of the parameters with significant correlation with the number of previous near-misses on the number of previous accidents at the multivariate level.
| Parameters | B | Std. Error | 95% Wald Confidence Interval | Hypothesis Tests | ||
|---|---|---|---|---|---|---|
| Minimum | Maximum | Wald x2 | p | |||
| Intercept | 1.199 | 0.5098 | 0.200 | 2.199 | 5.536 | 0.019 |
| [Education = primary school and below] | −1.533 | 0.1692 | −1.864 | −1.201 | 82.000 | 0.000 |
| [Education = high school] | −0.062 | 0.1607 | −0.377 | 0.254 | 0.146 | 0.702 |
| [Education = university and above] | 0 a | - | - | - | - | - |
| [Accident = no accident] | −3.654 | 0.1758 | −3.999 | −3.310 | 431.951 | 0.000 |
| [Accident = accident] | 0 a | - | - | - | - | - |
| [Marital status = single] | −0.134 | 0.1310 | −0.391 | 0.123 | 1.048 | 0.306 |
| [Marital status = married] | 0 a | - | - | - | - | - |
| Age | 0.074 | 0.0092 | 0.056 | 0.092 | 63.629 | 0.000 |
| Correct score | 0.014 | 0.0047 | 0.005 | 0.023 | 8.529 | 0.003 |
| Incorrect time | 0.228 | 0.0610 | 0.109 | 0.348 | 14.016 | 0.000 |
| Unanswered score | −0.029 | 0.0052 | −0.039 | −0.019 | 31.088 | 0.000 |
| Scale | 3.754 b | 0.1591 | 3.455 | 4.079 | ||
a Reference category, b Likelihood estimation.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
MDPI and ACS Style
Tasdelen, A.; Özpinar, A.M. A Dynamic Risk Analysis Model Based on Workplace Ergonomics and Demographic-Cognitive Characteristics of Workers. Sustainability 2023, 15, 4553. https://doi.org/10.3390/su15054553
AMA Style
Tasdelen A, Özpinar AM. A Dynamic Risk Analysis Model Based on Workplace Ergonomics and Demographic-Cognitive Characteristics of Workers. Sustainability. 2023; 15(5):4553. https://doi.org/10.3390/su15054553
Chicago/Turabian StyleTasdelen, Ahmet, and Alper M. Özpinar. 2023. "A Dynamic Risk Analysis Model Based on Workplace Ergonomics and Demographic-Cognitive Characteristics of Workers" Sustainability 15, no. 5: 4553. https://doi.org/10.3390/su15054553
Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.
