The Analysis of the Effect of a Mental Workload and Burnout on Employees’ Safety Behavior in the Oil and Gas Industry Using Roster Systems ^†

Abstract

The oil and gas industry is a high-risk industry because of the work, meaning that employees in this industry to be very prone to work accidents. Employees who work in isolated locations such as offshore platforms in the long term can experience mental and emotional fatigue, resulting in burnout. Therefore, this research was conducted with the aim of analyzing and identifying the influence of mental workload and burnout on safety behavior with different roster systems of 2 weeks and 3 weeks in the oil and gas industry. The researchers used several approaches to analyzing mental workload, burnout, and safety behavior using the PLS-SEM method with SMART PLS software. Based on the results obtained from the comparative analysis of the roster systems, it was concluded that the 2-week roster system is better than the 3-week one. This research can also provide information to companies regarding the level of mental workload, burnout, and their relationship with safety behavior and recommendations programs for companies to minimize the risk of accidents by paying attention to the mental health aspects of their employees.

1. Introduction

Industries operating in the oil and gas sector are industries that have high risks, such as in the upstream sector with production and drilling activities. Employees of the offshore oil and gas industry have more mental health problems than employees working onshore. When mental health issues arise, this can reduce the level of perceived safety in work situations [1]. Employees on offshore oil and gas platforms have a very complex and demanding job, given the extreme working environment, such as unpredictable weather conditions, high ocean waves, strong winds, and isolation. In addition, 24 h operational uncertainty in the oil and gas industry is a concern because employees must be able to cope with unexpected situations and make the right decisions fast.

The oil and gas industry applies a roster system. This system is usually applied to industries that have long working hours ranging from 10 to 14 h, with an average of 12 h considered the standard. The increased duration of work can make the number of tasks that must be handled by employees increase in one period, in addition to employees having to maintain productivity for longer periods of time. The inability to concentrate properly due to continuous work, resulting in burnout, can have an impact on employee safety and health, especially in an offshore platform environment. Mental workload is considered as the complexity of tasks that must be performed for daily activities, or the interaction between an employee’s capacity and their job tasks [2]. When mental workload increases, control is lost, errors occur, and the amount of work completed by an individual tends to decrease. This can reduce the amount of attention given to the surrounding environment, including safety behavior [3].

Looking at the problems discussed above, researchers want to analyze the cognitive ergonomic factors, namely mental workload and burnout, which can affect work safety behavior. These cause work accidents that disrupt the productivity of a company and the individual employees working in the oil and gas sector in Indonesia in companies that have changed the roster system from 2 weeks on duty and 2 weeks off duty to 3 weeks on duty and 3 weeks off duty. The researchers only focus on employees’ safety behavior influenced by mental workload and burnout.

2. Literature Review

2.1. Roster System

A roster system is one of the shift systems of work that involves traveling long distances to work in isolated areas. Another term that describes this work system is fly-in fly-out (FIFO). The roster system differs from other work systems in that there is a rotating work schedule consisting of predetermined consecutive days of work (on duty), followed by periods of leave (off duty), such as 2 weeks on duty/2 weeks off duty, or 8 weeks on duty/4 weeks off duty. The roster system is characterized by extended working hours ranging from 10 to 14 h per shift [4]. This roster system is applied to work offshore and in the platform oil and gas sectors, but also in the mining and construction sectors.

Working using a roster system is usually associated with relatively high salaries and long leave periods. However, roster system work is often associated with several health and well-being issues among employees. Long working hours, the job demands, and the lack of social support can lead to high levels of stress, poor mental health, and decreased well-being [5].

2.2. Mental Workload

Mental workload is a condition of overload caused by allocating all available resources and having no spare capacity to meet a demand, which most often leads to a poor performance, and even total failure [6]. Mental workload is related to the relationship between the human mind and its limitations. If humans are overloaded, learning new things or performing tasks will be hard. When job demands are met with time pressure and task complexity, the perception of mental workload will increase [7]. Intensifying tasks can cause errors and increase the number of errors based on the response time. Incapacity regarding changes in task difficulty explains why participants show a much greater decrease in performance [8]. In this case, as the duration of roster increases, the job demands and time pressure will increase so that employees need to maintain constant mental effort in order to have a resource reserve capacity.

2.3. Burnout

Burnout is a process of mental exhaustion and depersonalization that can occur among individuals experiencing job burnout and stress. Burnout also causes a decrease in the quality and quantity of completed work [9]. There are emotional changes during the enforcement of the roster system. When the duration of work increases, employees’ resources are reduced, so they do not feel competent enough to complete the work and have to suppress their true feelings, causing burnout to occur over time [10].

Burnout is the most important concept from an employee perspective, defined as a state of emotional and mental exhaustion that involves a loss of energy, idealism, perspective, and purpose and causes continuous stress, hopelessness, helplessness, and the feeling of being stuck in situations [7]. The complexity of activities in the petroleum industry requires employees to maintain constant attention over long periods of time. Employees are also exposed to high-level risks due to being in offshore locations, such as risks of fires and explosions, bad weather, and isolation on platforms, and they have 12 h shifts and work to a roster system, which can cause burnout [11].

2.4. Safety Behavior

Safety behavior is an activity carried out by individuals in an organization related to safety. In safety behavior, there are several influencing factors, such as employees’ physical health or employees’ mental health, the safety climate, leadership, organization, and others. Few studies have investigated psychosocial safety behavior, which refers to the psychological safety in the workplace of employees [12]. Psychosocial safety behaviors include changing work habits to reduce work stress or initiating actions, such as reporting workplace incidents or accident events [3].

Safety behavior is related to work safety factors; safety climate factors have a larger influence on safety behavior than work experience [13]. In one of the oil and gas industries, there was a roster change, which led to increased emotional exhaustion, depersonalization, and impaired personal achievement, which certainly affected the people at work. Job burnout can lead to dysfunctional attitudes that affect employees’ behavior. In addition, it will affect employees’ safety behavior, which includes safety compliance and employee safety participation. Safety compliance has a direct influence on accidents as it relates to whether or not employees comply with the safety policies. Participation in safety can help reduce accidents by improving the work environment through safety training, or accident reporting to improve workplace safety. Employees who experience a lot of work pressure are less likely to use safety equipment or initiate and report accidents [14].

3. Methods

According to the literature review, we will determine the influence between mental workload variables and burnout variables on safety behavior variables. We measure mental workload using Carmen-Q to obtain a pure mental workload measurement and remove the physical aspects. In addition, to measure burnout, we use the MBI-GS related to objects in the oil and gas industry. In this study, the second-order model is analyzed using a two-stage approach. The variables used in this study are mental workload, burnout, and safety behavior; these three variables are of the second order. In the mental workload variable, the dimensions of the first order used in this study are the dimensions of cognitive demands, temporal demands, emotional demands, and performance demands; in the burnout variable, there are dimensions of exhaustion, cynicism, and professional efficacy; and in the safety behavior variable, there are dimensions of safety participation and safety compliance. Figure 1 shows the second-order model used in this study.

Based on previous research and various existing theories that have been described in relation to this model, several hypotheses can be proposed as follows:

H1: 

Mental workload has a significant effect on burnout;

H2: 

Mental workload has a significantly negative effect on safety behavior;

H3: 

Burnout has a significantly negative effect on safety behavior.

Design of Survey and Data Collection

Based on the literature review, a cross-sectional questionnaire was methodically developed tailored to the object under study. This research was conducted in the territory of Indonesia. The questionnaire was distributed to employees of related companies in a paper-based form; determining the minimum number of respondents required using G-Power Analysis. We obtained a minimum of 68 respondents using an effect size value of 0.15, an α of 0.05, and a power of 0.80. In this study, there were 90 respondents. This study used PLS-SEM, with more than 120 questionnaires distributed to employees of the oil and gas industry on offshore platforms.

The Likert scale method was used to measure information from the sample. Mental workload was measured through 29 items contained in 4 dimensions, burnout was measured through 16 items contained in 3 dimensions, and safety behavior was measured through 6 items contained in 2 dimensions. The Likert scale requires a source of information to indicate the level of agreement or disagreement with written or verbal statements, and the results of this information are processed in the form of quantitative data. The statements form a series of questions, with the respondents asked to choose on the category scale. The questionnaire in this study used a Likert scale with the following scores, 1 = strongly disagree, 2 = disagree, 3 = undecided, 4 = agree, and 5 = strongly agree [15].

4. Results

In PLS-SEM, the measurement model aims to determine the relationship between the latent variables and their dimensions. In this study, we used Smart-PLS software (Version 4.1.0.8). As shown in

Table 1. Constructs’ convergent validity for 2 weeks.

Latent Variable
	Dimension	Loading Factor	Cronbach’s Alpha	Composite Reliability	AVE
Mental Workload	CD	0.845	0.746	0.765	0.568
	TD	0.789
	ED	0.732
	PD	0.739
Burnout	E	0.860	0.867	0.899	0.790
	C	0.930
	PE	0.874
Safety Behavior	SC	0.953	0.897	0.897	0.907
	SP	0.952

and

Table 2. Constructs’ convergent validity for 3 weeks.

Latent Variable
	Dimension	Loading Factor	Cronbach’s Alpha	Composite Reliability	AVE
Mental Workload	CD	0.921	0.905	0.934	0.779
	TD	0.853
	ED	0.885
	PD	0.871
Burnout	E	0.880	0.863	0.916	0.785
	C	0.914
	PE	0.864
Safety Behavior	SC	0.940	0.855	0.932	0.873
	SP	0.929

, the loading factor for all the items ranges from 0.732 to 0.953, where the value is greater than the threshold of 0.7 [16]. Reliability testing aims to see whether the dimensions are reliable for measuring the latent variables. In Table 1 and Table 2, all the Cronbach’s alpha values reflect the lowest reliability value of a variable, composite reliability reflects the true reliability value of a variable, and AVE determines that convergent validity for all the constructs is above the threshold. Namely, for Cronbach’s alpha, the minimum value is 0.6; for composite reliability, the minimum value is 0.7; and for AVE, the minimum value is 0.5 [16], so it can be concluded that all the constructs have met the standard and are reliable.

In

Table 3. The statistical results of the measurement model for 2 weeks.

Hypothesis	Latent Variable	Original Samples	T Statistics	p-Value	Information
H1	MW -> B	0.514	5.938	0.000	Accepted
H2	MW -> SB	−0.307	2.550	0.011	Accepted
H3	B -> SB	−0.315	2.524	0.012	Accepted

and

Table 4. The statistical results of the measurement model for 3 weeks.

Hypothesis	Latent Variable	Original Samples	T Statistics	p-Value	Information
H1	MW -> B	0.629	7.604	0.000	Accepted
H2	MW -> SB	−0.344	2.410	0.016	Accepted
H3	B -> SB	−0.337	2.167	0.030	Accepted

, we present the path coefficient to show the level of significance in hypothesis testing. In week 2 of the roster system, the path coefficient of MW on SB is 0.011, and the T statistical value of 2.550 is greater than the t-table of 1.96; therefore, it can be concluded that mental workload has a significant negative effect on safety behavior, and the effect is −0.307. Furthermore, to test the other hypotheses in the 2-week and 3-week roster systems, it can also be concluded that all the hypotheses are significant because the T statistical value is greater than the t-table. The path diagram can be seen in Figure 2.

H1 is in accordance with previous research regarding mental workload, finding that it increases simultaneously with burnout; in other words, the more people feel time pressure, such as increased shift duration, and higher cognitive demands, the more they will feel burnout at work [2]. H2 states that a high mental workload in employees can cause an increase in human errors and results in work accidents due to unsafe behavior. Work that requires constant mental effort consequently leaves limited resources to manage safety behavior [3]. H3 regarding burnout in the work environment can cause employees to ignore some safety norms, thereby exhibiting behaviors that are not conducive to safety and have an impact on humans at work and weaken compliance with the safety behavior standards [17].

5. Discussion

We analyzed the hypotheses using the 2-week and 3-week roster systems to find out which roster system is better. The hypothesis results for 2 weeks and 3 weeks can be seen in

Table 5. Comparative analysis of 2-week and 3-week hypotheses.

Hypothesis	Latent Variable	2 Weeks		3 Weeks
		Original Samples	Information	Original Samples	Information
H1	MW -> B	0.514	Accepted	0.629	Accepted
H2	MW -> SB	−0.307	Accepted	−0.344	Accepted
H3	B -> SB	−0.315	Accepted	−0.337	Accepted

.

After comparative analysis, it was found that the greatest influence of mental workload (MW) and burnout (B) on safety behavior (SB) was found at 3 weeks. When the mental workload increases and burnout occurs, it has enough of an impact on safety behavior well. At 3 weeks, there is an increase in the negative impact on safety behavior, namely in the mental workload, which affects safety behavior by values ranging from −0.307 to −0.344, and burnout, which affects safety behavior by values ranging from −0.315 to −0.337. This shows that the roster system of 2 weeks is better than the roster system of 3 weeks.

There is an emotional change during the roster cycle; employees have to suppress their true feelings and portray themselves more positively, leading to increased burnout. For this, the organization must ensure optimal job autonomy and emotional demands [10]. When the period of time on the roster system increases, it causes negative reactions such as stress; in addition, job demands and an increased mental workload can cause reactions that have a negative psychological impact [18]. This is also influenced by intense time pressure; these employees have to work long shifts at a consistent pace, which does not allow much room for error, so psychological distress and the rates of mental health problems are significantly higher in roster workers compared to those of employees using other systems [19]. Organizations need to implement training programs and set up systems of support to more effectively instill wisdom and support decision making; such programs should incorporate effective management and improve employee resources, including mental health and well-being [10].

6. Conclusions

In this study, we conducted the analysis of the oil and gas industry on offshore platforms using mental workload, burnout, and safety behavior variables with two different roster systems, namely ones lasting 2 weeks and 3 weeks. It was found that the 2-week roster system was better for the mental health of the employees. This is because 3 weeks of work has a greater negative influence on safety behavior. It can be concluded that when the mental workload increases and burnout occurs in employees in one of the oil and gas industries in Indonesia, it stops employees from performing safe behavior. Improvement recommendations were obtained to improve employees’ mental health, such as mental health education programs, counseling services, physical health programs, and anti-stigma campaigns.

Furthermore, future research could add other related variables that can explain safety behavior, conduct intervention analysis to complement this research, or analyze other demographics; this will provide further insights that can indirectly affect mental health. This research only considered one object; it should be considered to carry out research on more than one object and discuss, in more detail, the differences between the employees of the company that is used as the object and the third party in the company.