Comprehensive analysis of occupational accidents contributing factors in the electrical industry

This research investigates the linkages between the contributing factors of occupational accidents in the electrical industry, i.e., the business unit factor, the time of accidents, the level of injuries, the gender, and the day when occurred. The research also analyzes the linkage between the contributing factors of accidents to the occurrence of occupational accidents. The research method employed crosstab analysis as a quantitative technique through reports of occupational accidents, especially in the Indonesia Electrical Industry. Moreover, dummy variable regression is utilized to analyze the impact of contributing factors on the likelihood of work accidents. The results show that there is no significant statistical relationship between the contributing factors of occupational accidents, namely the business unit to the level of injury, the business unit to the day, the time of the accident to the level of injury, the time of the accident to the day, and the level of injury to the day. This research also finds that the significant variable in occupational accidents is the type of generation unit. As for the variable type of injury and the days, both are not proven to correlate statistically. This study can help the electrical industry improve occupational health and safety (OHS) performance through analysis to optimize the company's resources. The Management of the electrical company can allocate the right policy to increase OHS operational performance. This research integrates the identifying significant factors of occupational accidents and analyses their linkage, which can help guide future research in this area. © 2023 by the authors. Licensee SSBFNET, Istanbul, Turkey. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).


Introduction
Approximately 2.3 million individuals lost their lives due to occupational accidents or work-related illnesses (ILO, n.d.). The Indonesia National Social Security Agency for Employment has revealed that the number of occupational accidents steadily increased over the past five years. In Indonesia, occupational accidents have risen steadily over the past five years. In 2021 the number of occupational accidents in Indonesia rise by 5.65% from 2020. Likewise in 2020, there was an increase of 21.28% compared to 2019(General of Development of Labor Inspection and Occupational Safety and Health, 2022;Mahdi, 2022).

561
According to PLN's sustainability report, the rate of occupational accidents at PLN, in general, has increased from 2016 to 2020 (PLN, 2020(PLN, , 2021. In 2016, there were 32 occupational accidents at PLN. When compared to 2017, there was an increase in occupational accidents of around 47%, namely 47 incidents when compared to 2016. For 2018, the level of occupational accidents was almost the same as the previous year, namely 45 incidents of occupational accidents. for 2019 and 2020 the level of occupational accidents at PLN is at almost the same level, namely 61 incidents and 60 incidents. The 2019 occupational accident rate showed an increase of around 35% compared to 2018. As for data for 2021, there was indeed a decrease in the number of occupational accidents at PLN, namely 35 work incidents, but it should also be noted that the decrease of around 42% can be assumed to be due to the covid pandemic -19 so that the reduction in the rate of occupational accidents must still be welcomed with consistent and strict OHS work performance so that occupational accidents can be truly controlled towards zero accidents. Occupational accidents that occur can result in serious injury and even death. This can disrupt the productivity of PLN. Work accidents can also damage the company's reputation. Negative news about work accidents can generate negative views from the public, customers, and investors. This can reduce customer trust and loyalty, as well as reduce brand value and corporate image. A work accident or accident because of work is an incident where there is contact or exposure or movement of an object, material, or other people, which causes injury or other possible consequences (Heinrich, 1941). Uncontrolled potential hazards can lead to fatalities. The company must ensure proper controls are implemented to prevent accidents. Accidents can cause significant harm to workers and their families and damage the company's reputation. Therefore, the company needs to prioritize occupational health and safety measures to prevent accidents and protect the well-being of its workers. Furthermore, the management of occupational safety risks in the electrical industry has become a top priority in the electricity business management standard. The national electricity company, PLN, which has the slogan "Nothing is more important than human life," is highly concerned with monitoring and improving workplace safety incidents to achieve zero accidents. Work accidents can be attributed to human and environmental factors (Ahad & Saldy, 2020). rules violations categorize as unsafe behavior and unsafe conditions (Mitropoulos & Cupido, 2009). Unsafe behavior refers to unsafe actions taken by humans, such as not following applicable procedures or not using appropriate personal protective equipment (PPE) while working. Unsafe conditions refer to environmental factors around the work area that are not safe, such as unsafe equipment. Work accidents can occur due to one or more causal factors (Ivascu et al., 2021). Evaluating accidents can provide opportunities to improve occupational health and safety (OHS) performance (Porcu et al., 2020). There has been no study on the temporal characteristics of work accidents and their frequency (Khalid et al., 2021). Therefore, conducting such an analysis could provide added value for companies. It is essential to explore every potential safety factor that can lead to work accidents (Khalid et al., 2021).
This research offers an in-depth analysis of work accidents in the electrical industry. Firstly, Jung et al.'s (2022) study analyzed work accidents temporally (Jung et al., 2022). In addition, this research uses time series data from 2016 to 2021. Secondly, this research also includes demographic factors, such as the gender of the victims and the business unit where the work accidents occurred, to provide a more comprehensive analysis. Thirdly, the data used in this study is based on actual work accidents in the electrical industry in Indonesia, specifically from the PLN institution, which can accurately depict the existing conditions in the field during work operations.
This research aims to analyze the relationship and linkage of factors that contribute to work accidents in the electricity industry, specifically within the state-owned enterprise namely PLN in Indonesia. The analyzed factors include the PLN business unit where work accidents occur, the level of accidental injuries, the time of work accidents, and the gender of work accident victims. Through quantitative methods with crosstab testing, chi-square analysis, and dummy variable regression analysis, this research aims to predict the magnitude of the value of the event of work accidents and provide a comprehensive understanding of the causes of work accidents in the industry. By identifying these factors and their relationships, this research aims to support the implementation of occupational accident prevention programs and promote OHS cultural awareness in PLN and other companies in the electricity sector.

Literature Review
"A work accident or accident as a result of work is an incident where there is contact or exposure or movement of an object, material, or other people which causes injury or other possible consequences (Heinrich, 1941). Based on the Law of the Republic Indonesia No 1 of 1970 concerning Occupational Safety," a work accident is an incident that is unexpected and unwanted, which disrupts the regulated process of an activity and can cause harm to both human victims and property (Ministry of Labor, 1970). From these two definitions, a work accident is an unexpected condition caused by contact/exposure/movement between a person and objects/materials/other people and results in loss of life and property.
Job design or work design refers to the content, structure, and organization of tasks and activities (Van Den Broeck et al., 2017). Job design is the content and organization of one's work tasks, activities, relationships, and responsibilities, which are also related to the organization's goal, be it safety, performance, innovation, and others. The design means that workplace safety has a significant influence on the job design of workers. Job design can have a positive impact on increasing work productivity as well as the effectiveness and efficiency of work patterns and labor costs. Job Design can be a powerful tool to learn, develop, maintain, and improve physical and mental health and simultaneously achieve control and flexibility (Parker, 2014). Job design can foster employee-customer identification and enhance service performance (Li et al., 2021). The Government must increase the productivity of an organization with implementation work detail (Rinayati et al., 2023). Without a good job design, it will affect work results at all levels, including individual, and employee involvement or the level of worker safety, and how performance targets are achieved while still prioritizing occupational health and safety (OHS) for workers. Therefore, the risk of an inaccurate job design will affect workers who become uncomfortable and insecure at work. The risks can potentially lead to an increase in occupational accidents. A work accident can occur because of one or more causal elements (Ivascu et al., 2021). Evaluating accidents that occurred would provide opportunities for improving OHS performance (Porcu et al., 2020). By focusing on safety management, when the frequency of work accidents occurs a lot, the effect of safety management would be maximized even with limited resources (Khalid et al., 2021). Occupational safety for all operational business processes is essential for companies, so they must prioritize work safety for their workers. Every company needs to control the risks that may occur in its operational business processes to ensure the quality of work for each employee.

Contributing Factors to Occupational Accidents
Work accidents can be caused by two factors, namely, humans and the environment (Ahad & Saldy, 2020). Rules violations are considered unsafe behavior and conditions (Mitropoulos & Cupido, 2009). Unsafe behavior/action is a dangerous category of action humans carry, i.e., not doing work according to applicable procedures, not using appropriate Personal Protective Equipment (PPE) when working, etc. Unsafe condition is a condition/state of the category that is unsafe for the environment due to the equipment around the work area. There had not been a study of the temporal characteristics of the occurrence of work accidents on the frequency of work accidents (Khalid et al., 2021). The findings reveal that complex combinations of risk factors often contribute to accidents (Reiman et al., 2018). Therefore, it is necessary to carry out this analysis to add value for the company.

Business Unit where the accident occurred (as X1)
The interaction within the system influences job design, including the work environment surrounding a job (Mondy, 2008). This work environment includes locations, facilities, tools, lighting, cleanliness, and calm. The work environment is an essential scope because it is the location of the center of worker activity. According to Law Number 1 of 1970, a workplace is any room or field, closed or open, moving, or fixed, where the workforce works, or which the force often enters for a business and where there are sources or resources -a source of danger(Ministry of labor, 1970). The work location or business unit where the accident occurred will affect the level of work risk faced. The more dangerous the workplace will be, the greater the potential for work accidents. The management must identify the workplace and assess the work risks that arise due to the influence of the workplace. The work analysis reveals the need to consider organizational factors in prevention practices (Filho et al., 2012). The identification of the workplace can be influenced by the type of the company's core business. Companies in the electricity sector are generally divided into three core businesses, namely the Generation, Transmission, and Distribution Business Units. The Generation Business Unit is a business unit whose main task is to ensure that the electricity supply from generation is available in an efficient, reliable, and environmentally sound manner with world-class generator governance in its implementation. The Generation Unit usually consists of various resources, both fossil and renewable fuels. The Transmission business unit is a business unit that has the main task of transmitting or flowing electricity from the generator to the main substation. The distribution business unit is a business unit that has the main charge of distributing electricity from the transmission system to load centers. The company's business processes in the electricity sector are also supported by the construction unit, which ensures the construction of networks and generation that support operational processes. Of course, each business unit carrying out its work has specific potential hazards according to its respective business activities. Possible risks that are not controlled will result in unsafe actions, unsafe conditions, and near misses and can eventually lead to fatalities.

Occupational Accident Injury Rate (as X2)
Analyzing the frequency of accidents based on injury class will contribute to the development of more guidelines to prevent accidents (Jung et al., 2022). In this study, researchers tried to classify types of work accidents into 3 (three): minor, central, and death. The explanation for the level of injury is according to Table 1. Non-hospitalization or Minor injured Occupational accidents that result in minor to medium injuries but do not require hospitalization 2 Hospitalization or seriously injured Work accidents that result in victims must get further treatment at the hospital 3 Fatality Work accident resulting in death Source: (Jung et al., 2022) According to Table 1, the level of this injury must be analyzed to gain insight and a more comprehensive understanding of the work accidents that occur so that it can provide learning in carrying out the next OHS operational strategy (Jung et al., 2022).

Time of Occurrence of Work Accident (as X3)
The time limit for implementing construction projects results in workforce stress (Muhammad Affandi Maulana et al., 2022). Implementation time is crucial in the context of risky work because it affects the workers' condition.
Recent research indicates that the risk of work accidents can be influenced by the function of time and hours in a day (Kerstin Hänecke et al., 1998). Work can be performed at any time, whether within normal working hours or outside the standard working hours, commonly known as overtime. Additionally, there are shift work arrangements where workers are divided into groups according to the designated shifts set by management. All these work time arrangements can significantly increase the potential for work accidents.
The time of occurrence of a work accident is a parameter of a work accident incident which describes a series of moments when the process or work accident situation occurs. Working time is focused on the range of employee working hours, around 7.30 am to 04.00 pm. Other timeframes will adjust and be considered as a single work accident data unit for the time the work accident occurred.

Gender of Occupational Accident Victims (as X4)
Analysis of work accidents cannot be separated from the human factor itself (Ahad & Saldy, 2020). One of these human factors is the demographic factor, namely gender. Gender can affect work productivity, especially since the job is a job that uses solid and stable physical conditions. So gender can indicate a person's level of productivity. Universally, the productivity level of men is higher than that of women. Gender is influenced by factors owned by men, such as being physically stronger. In work, they tend not to use feelings or biological factors.

Day of Occurrence of Occupational Accident (as X5)
The day factor is a factor that is a variable in work accidents (Rey-Merchán & López-Arquillos, 2022). Days of the week are assumed to affect work accidents that occur. For example, it may happen that work on weekends is not work done effectively because it is not a working day, and so on. According to (Szóstak, 2019), the day factor is one of the factors in determining the analysis of work accidents in the construction industry. Working in unfavorable conditions can certainly affect workers' safety and occupational health, with the worst-case scenario resulting in work accidents.

The Linkage Between Contributing Factors of Occupational Accidents
From the above explanation, ten association relationships can be concluded, according to Table 2, and analyzed by crosstab and chisquare test.

Analysis of Dummy Variable Regression
Dummy Variable Regression was carried out to determine the effect of contributing factors of the occupational accident on the occurrence of occupational accidents in the PLN Group, as shown in Table 3.

No
Variable Hypothesis 1 X1 to X2 H1: There is no significant statistical relationship between the type of business unit factor where work accidents occurred and the time of occupational accidents 2 X1 to X3 H2: There is no significant statistical relationship between the type of business unit factor where work accidents occurred and the level of accident injuries that occurred 3 X1 to X4 H3: There is no significant statistical relationship between the business unit factor where work accidents occurred and the gender of occupational accident victims 4 X1 to X5 H4: There is no significant statistical relationship between the business unit factor where work accidents occurred and the day when work accidents occurred 5 X2 to X3 H5: There is no significant statistical relationship between the time of occupational accidents and the level of accident injuries that occurred 6 X2 to X4 H6: There is no significant statistical relationship between the time of occupational accidents and the gender of occupational accident victims 7 X2 to X5 H7: There is no significant statistical relationship between the time of occupational accidents and the day when work accidents occurred 8 X3 to X4 H8: There is no significant statistical relationship between the level of accident injuries that occurred and the gender of occupational accident victims 9 X3 to X5 H9: There is no significant statistical relationship between the level of accident injuries that occurred and the day when work accidents occurred 10 X4 to X5 H10: There is no significant statistical relationship between the gender of occupational accident victims and the day when work accidents occurred

Source: Authors
The quantity or value of the qualitative variables can be measured and converted into quantitative using the dummy variable.

Sample and Data Collection
This research applies a quantitative approach, explicitly using crosstab analysis and chi-squared test to identify the relationships among variables contributing to work accidents. Furthermore, regression analysis of dummy variables will be conducted to determine how the variables of type of business unit, time, the severity of the injury, and gender of work accident victims and the day are related. In general, it can be explained through the following Figure 1. To carry out the analysis in this study, generally carried out two pieces of analysis. In the first stage, after obtaining secondary data according to the research model, a crosstab test was carried out with chi-squared analysis on the factors that influence occupational accidents, such in the PLN business unit, i.e., the accident occurred, the time the accident occurred, the level of injury that occurred, gender and the day when accidents occurred.
This crosstab analysis was carried out to understand and explain the data more meaningfully using the relationships between factors so that it can be followed up towards other decisions (Maulid, 2022). The following analysis is to use a dummy variable regression. The purpose of dummy variable regression is to determine the value of the dependent/dependent variable based on one or more independent/independent variables, where one or more of the independent variables used are dummy. From this analysis, a relationship will be obtained between the contributing factors of accidents to the occurrence of occupational accidents. The finding can assist PLN management in developing strategies to improve OHS performance.
The dependent variable (tied) used in this study is the number of accidents. The number of accidents that occur is measured quantitatively by using the number of frequencies. The work accident data is data from PLN starting from 2016-2021.
The type of business unit where the accident occurred is identified according to the classification carried out by PLN. The grouping of business units is carried out based on the main types/activities carried out by the PLN unit into one variable regardless of the location factor according to the informed consent determined by PT PLN (Persero), namely, in this case, the OHS Division of the Head Office as the owner of data authority. The business where the accident occurred consists of 4 elements: Generation Unit, Transmission Unit, Distribution Unit, and Construction Unit.
Especially for secondary data when work accidents occur are data only for 2021. The accident injuries consist of minor, major, and fatalities/deceased. The gender variable consists of only men, which happens because the workers involved in the activities carried out by the operational (field activities) of the PLN group are men. Data will be analyzed on work accidents that occur based on the sex of the victim. The data of the day of occupational accidents is a typical calendar day when accidents occur. Secondary data on the rate of work accident injuries, gender, business unit, and the day of occupational accidents start from 2016-2021.
The occupational accident data used is PLN exclusive data obtained from the PLN work accident data manager, namely the Health, Safety, Security & Environmental (HSSE or K3L) division at the PLN head office. Work accident data snippets can also be obtained through the PLN sustainability report.

Measure and Statistical Methods
In this research, the construct measurements are developed in a quantitative research setting based on the variable definition and the validated measure from the previous studies. The research model can be shown in Figure 2. From the crosstab analysis, as shown in Table 4, we can know the result of crosstab analysis from the data of occupational accidents in PLN. The most fatalities occurred in the distribution unit by 33 times.

3
Unit Type * Gender The gender of the victim is only men with the most occurred in the distribution unit.

Unit Type * Day
Most days are Thursdays occurred most are in Distribution Unit.

Time *Injury Type
The most injuries type is fatality with the most occurred is at 10.01 am-07.00 pm.

Time * Gender
The gender victim is only men with the most occurred is at 10.01 am -07.00 pm.

Source: Authors
This crosstab analysis describes a wealth of information about the relationship between the variables. The calculation results for the study of the relationship between factors are in Table 5 from the chi-square analysis results. the gender data is only for men, so the data does not vary.
Most days are Thursdays with the most occurred is at 10.01 am-07.00 pm. 8

Injury Type * Gender
The gender victim is only men with the most type of injuries are fatalities.

9
Injury Type * Day Most days are Thursdays with the most type of injuries being fatalities.

Gender * Day
Most days are Thursdays with the gender of the victim is only men. the gender data is only for men, so the data does not vary.

9
X3 to X5 H9: the level of accident injuries that occurred affects the day when work accidents occurred Sig. 0.559. H0 is accepted, meaning there is no influence between the Injury Type Variable (X3) and the Day of Work Accident (X5).
-10 X4 to X5 H10: the gender of occupational accident victims affects the day when work accidents occurred The analysis cannot be carried out because there is a Chi-Square requirement that the variables cannot fulfill used (variables are constant).
the gender data is only for men, so the data does not vary.

Source: Authors
Based on Table 5, it can be observed that the chi-square analysis results indicate a symmetric relationship among the defined variables, namely Business Unit, Injury Rate time of the occupational accident, gender of the victim, and the day of the occupational accident occurred. There is no evidence of variables influencing each other. This supports the subsequent regression analysis, indicating that the independent variables should not have a mutual influence. It also illustrates that the highest level of significance is found in the relationship between the level of accident injuries that occurred and the day when work accidents occurred, at 55.9%, indicating no statistical relationship between these two variables. The same holds true for the relationships between the type of business unit and the level of accident injuries, the business unit factor and the day when work accidents occurred, the time of occupational accidents and the level of accident injuries that occurred, and the time of occupational accidents and the level of accident injuries that occurred.
Because of dummy variable regression, the effect of the X variable on the Y variable is obtained according Table 6 and Formula (2).

Findings
The chi-square analysis result illustrates that due to limited data owned by researchers, several variable relationships cannot be further analyzed, as shown in Table 5 and Figure 3. This is because the variables used are constant variables so the analysis cannot be obtained in more detail. The chi-square analysis results indicate no statistical relationship between the two variables namely relationships between the level of accident injuries and the day when work accidents occurred, the type of business unit and the level of accident injuries, business unit factor and the day when work accidents occurred, the time of occupational accidents and the level of accident injuries, and the time of occupational accidents and the level of accident injuries.
The result of dummy variable regression in this research is shown by the F statistical test that describes its significant value as 0.084 or greater than 0.10. The results mean that there is no influence of the variables X1, X3, X4, X5 together on variable Y. This illustrates that the model formulated is quite good through the impact of the variables X1, X3, X4, X5 together on the variable Y. X2 cannot be analyzed because the secondary data is only 2021 data.
In addition, the t-test found that the significance value of the generating unit type variable was 0.013 or less than 0.10. so that it is considered significant to variable Y. Meanwhile, other variables are above 0.10, so they are not significant to variable Y. This illustrates the influence of the generating unit variable on variable Y.
As for the R-square test, it was found that the resulting r-square test was worth 0.253, meaning that the independent variable can explain 25% of the distribution of the dependent variable. As for the remaining 75%, it cannot be explained by independent variables or can be explained by variables outside the independent variable (error factor). Based on the values, it can be said that the dependent variable in this study is influenced by 25% by the independent variable. The condition can happen considering the characteristics of work accident data are the minimum expected numbers.
Based on Figure 3 and Formula (2), we can conclude that the generating unit was the X variable that was significant to the Y variable because the p-value is less than 10% which is 0,013. The constant for the generating unit variable is -3,840. It can be said that when the work unit is a generation, the incidence of work accidents will decrease by 3,840.

Discussions
This research found that the generation unit was the X variable that was significant to the Y variable. When the work unit is a generation, the incidence of work accidents will decrease by 3,840. There is an indication that the unit type variable is significant by adding various and representative data or sample characteristics. As for the variable type of injury and days, it is not proven to have a relationship statistically. Meanwhile, the gender variable cannot be included in the regression model because the data is uniform, so it cannot explain the dependent variable adequately. For the day variable, it can be concluded that work accidents can occur at any time on any day because they do not have a statistically significant effect on work accidents.
In research conducted by (Jung et al., 2022), a similar relationship was also found, a temporal investigation of Construction Equipment-related Accidents (CEAs) shows the frequency of accidents based on time and injury class. It was found that the frequency of CEAs as a function of time is statistically significantly different.
This research is also in line with (Camino López et al., 2008)that was shown that the time of the accident, the day of the week, and the remainder of the variables under analysis influenced the seriousness of the accident. It also has a similar idea to (Rey-Merchán & López-Arquillos, 2022) that sex, unit sector, and the day are the factors analyzed in the incidence of injury. More importantly, no previous studies have been found that have analyzed the linkage between contributing factors in occupational accidents to the occurrence of occupational accidents.

Conclusions
The objective of this research is to determine the relationship between the contributing factors of occupational accidents and the relationship between the contributing factors of occupational accidents to the occurrence of occupational accidents. The results show that there is no statistical relationship between the contributing factors of occupational accidents, namely the business unit where work accidents occurred to the level of injury sustained, the business unit where work accidents happened to the day when work accidents occurred, the time of the accident to the level of injury sustained, the time of the accident to the day when work accidents occurred, and the level of injury sustained to the day when work accidents occurred. Some correlations cannot be analyzed due to limited data.
This research also finds that the significant variable to the occupational accidents is the type of generation unit. As for the variable type of injury and the days, both are not proven to correlate statistically. Meanwhile, the gender variable cannot be included in the regression model because the data is uniform, so it cannot explain the dependent variable adequately.
The results provide managerial implications that explain the relationship between factors that contribute to workplace accidents by using dependent variables. This research also has significant policy embodiment within the company and the company ecosystem.
Research proves that there is a relationship between the independent variables. Specifically, the variable type of generation unit is a variable significant to the dependent variable, namely the incidence of work accidents.
Occupational safety for all elements involved in operational business processes is essential for companies, so they must be committed to prioritizing work safety for their workers. Every company needs to control risks that may occur in its operational business processes to ensure the quality of work for each employee.
No primary data was collected for the current research, then only secondary data was performed. Secondary data have advantages availability and access and disadvantages such as lack of specific research or incomplete information. The other factor could affect occupational accidents. Some additional factors contributing to the occupational accident can be addressed in future research to complete the research. Additionally, complementing research with qualitative interviews will also result in more comprehensive research.
Based on the results of this research, PLN as the relevant institution can follow up and pay attention to the influence of the variables used on the level of work accidents that occur. From the management perspective, PLN units need to ensure daily supervision of work, especially for high-risk tasks. From the perspective of PLN Head Office management, improving the database on work accident information would be highly significant as this information can serve as a strength in future analyses.