Effects of mental workload on manufacturing systems employees: A mediation causal model

Abstract

BACKGROUND:

Although some research has been done in the Mexican manufacturing industry regarding mental workload, none has explored its association with physical fatigue, body weight gain, and human error simultaneously.

OBJECTIVE:

This research examines the association between mental workload and physical fatigue, body weight gain, and human error in employees from the Mexican manufacturing systems through a mediation analysis approach.

METHODS:

A survey named Mental Workload Questionnaire was developed by merging the NASA-TLX with a questionnaire containing the mental workload variables mentioned above. The Mental Workload Questionnaire was applied to 167 participants in 63 manufacturing companies. In addition, the mental workload was used as an independent variable, while physical fatigue and body weight gain were mediator variables, and human error was a dependent variable. Six hypotheses were used to measure the relationships among variables and tested using the ordinary least squares regression algorithm.

RESULTS:

Findings indicated that mental workload significantly correlates with physical fatigue and human error. Also, the mental workload had a significant total association with human error. The highest direct association with body weight gain was provided by physical fatigue, and body weight gain had an insignificant direct association with human error. Finally, all indirect associations were insignificant.

CONCLUSION:

Mental workload directly affects human error, which physical fatigue does not; however, it does affect body weight gain. Managers should reduce their employees’ mental workload and physical fatigue to avoid further problems associated with their health.

Keywords

Body weight gain human error manufacturing systems mental workload physical fatigue

1 Introduction

Currently, scientific and technical breakthroughs are assisting in improving the quality of life in the workplace by reducing physical risk factors such as manual materials handling, vibration, repetitive movements, force application, and awkward body postures [1]. However, mental tasks have taken more relevance in today’s industrial sectors. According to several authors, the mental workload is defined as the level of effort involved in executing a specific activity [2, 3].

Some research points out that mental workload can cause negative effects on employees’ performances, such as degradation in decision-making quality in plane pilots [4 –6], communication conflicts and accidents in aviation [7, 8], errors in the use of agricultural vehicles [9], low workability and bad quality of life in school teachers [10, 11], to mention a few.

Currently, the manufacturing industry is demanding high levels of the mental workload from employees, which in turn affects operators and managers by reducing their attention to the task, resulting in distractions and human errors that are manifested in injuries to employees and economic costs for companies [12]. Sometimes, employees are involved in different projects, and to ensure their success, they must analyze, process, and use too much information at a specific time, which may lead employees to suffer mental workload. Furthermore, employees are frequently under pressure to deliver results at a specific time, which increases the mental workload [4 , 13–16].

Some research regarding mental workload has been reported and its effects on employees in the manufacturing industry, as shown by Barajas-Bustillos et al. [12], that report a literature review on mental workload in the manufacturing industry and found that only 105 researches have been published from 1981 to 2017 (i.e., less than three researches by year) on this industry sector. In addition, Zoaktafi et al. [17] note that assessing mental workload is an important part of developing human-machine interfaces (HMIs) for the manufacturing industry.

Although several authors have performed research on mental workload in manufacturing companies in Mexico [18 –20], most of the operators, managers, supervisors, engineers and other employees unknow the methods and techniques to assess and evaluate the mental workload level and their effects on them.

Particularly, the mental workload can cause physical fatigue [21], body weight gain [22], and human error [23] in manufacturing employees. In the case of physical fatigue represents a challenging ergonomic/safety “issue” since it negatively impacts employees’ health and safety and decreases productivity [24]. Also, some authors mention that obesity is linked to absenteeism and reduced performance [25], while human error facilitates the appearance of quality defects and accidents, negatively affecting system reliability [26]. Therefore, managers should pay attention to these variables.

Mexico’s manufacturing sector is significant because it supports over 2,280,504 direct jobs [27] and is home to 604,250 maquiladora businesses [28]. A maquiladora industry is an entity that assembles, manufactures, processes, or repairs electronic materials that are imported by the hosting country for eventual re-exportation or shipment to the country of origin [29].

The Mexican maquiladora industry includes manufacturing food, clothing, wood products, chemical substances production, metallic products, machinery, and equipment, to mention a few [30]. But incorporating a variety of machinery and equipment in the maquiladora industry has partially replaced some activities carried out manually. However, most of the products made in this industrial sector require an intensive manual labor force [31].

An employee in the manufacturing sector can expect to put in an average of 521,065,000 hours per month, or 11 hours per day on average [32], which gives some idea of the mental and physical strain that this can entail. In Mexico, obesity is a current problem that has been present for a long time (since 2002 or even before) [33]. Additionally, the nation is ranked number 1 in the world for overweight and obesity [34], and nationwide, in 2018, 75.2% of adults over 20 were overweight (39.1%) or obese (36.1%) [35].

Although some research has been done in the manufacturing industry in the country and worldwide regarding mental workload, none of this research has explored the association of mental workload with physical fatigue, body weight gain, and human error simultaneously. This research represents an exploratory study on the effects of mental workload on other variables, which may act as dependent variables or mediating variables between mental workload and some independent variable(s). In this experiment, only the effects of mental workload on physical fatigue, body weight gain (as mediating variables), and human error (as the independent variable) are analyzed. However, future research, such as overwork, memory loss, depression, or social problems, and our results aim to generate new hypothetical models and investigate their associations.

This paper focuses on measuring mental workload level and its association with the variables mentioned above in senior and middle management from manufacturing companies located in Tijuana, Mexico, where the National Constitution regulates occupational safety and health, the Organic Law of the Federal Public Administration, the Federal Labour Law, the Federal Law on Metrology and Standardization, the Federal Regulations on Occupational Safety and Health, and other regulations [36]. According to article 123, section “A,” section XV of the Supreme Law, an employer must follow the legal precepts on hygiene and safety in the whole establishment and take appropriate measures to prevent accidents and guarantee the health of its employees [36].

In addition, the Mexican Ministry of Health establishes the general conditions for occupational health, declaring that employers and Unions must integrate the Safety and Hygiene Commission to issue guidelines to reduce and prevent occupational hazards and determine the required clothing and equipment for safety [37].

The next subsections present the hypotheses proposed in this research. Findings from this research will allow managers to know and have empirical and statistical evidence that will allow them to justify investment for their workers’ benefit and comply with Mexican regulations and standards that regulate them since, in a study reported by Reyes et al. [38], 62.66% of human errors in maquiladora industry are affecting employee’s health as accidents.

1.1 Association of mental workload with physical fatigue, body weight and human error

The mental workload can affect human performance and is considered critical in designing and evaluating complex human-machine systems [39]. In addition, the mental workload is constantly presented in various industrial sectors, resulting in many obvious negative effects on workers. These effects may include physical fatigue, body weight gain, and human error, among others.

Physical fatigue can be defined as a decrease in performance due to physical, mental, or emotional exertion. It can affect many physical abilities, from strength and speed to reaction time and decision-making [40, 41]. As a result of putting in too much physical work for too long, workers may start to feel physically exhausted and lose productivity on the job. [40, 42]. Sarsangi et al. [43] mentioned that mental workload is a factor that often causes physical fatigue. Realyvásquez-Vargas et al. [44] performed a descriptive study among 167 manufacturing employees to know the effects of mental workload, and findings indicate that 74, 15, and 5 declare that sometimes, usually, and always, respectively, they felt physical fatigue due to mental workload.

Mehta and Prasuraman [45] also studied the relationship between mental and physical exhaustion in an experiment with 12 participants. The analysis was performed by measuring the activation of the prefrontal cortex during submaximal fatigue handgrip exercises. The participants carried two types of exhaustion: physical (control) and mental (concurrent). To the point of exhaustion, at 30% of their maximum handgrip strength, both conditions exhibited good behavior. Their results indicated that mental workload contributes to physical fatigue. Based on this result, hypothesis H₁ is proposed:

H₁: Mental workload has a direct and positive association with physical fatigue in employees from the Mexican maquiladora industry.

Moreover, regarding body weight gain, several authors have found a relationship between mental workload and body weight gain. For instance, Overgaard et al. [46] studied for six years to know the relationship between mental workload and changes in body weight. Their research included 6704 Danish nurses, from 45 to 65 years old, who were both employed in 1993 and 1999. Consequently, nurses who said to be never busy or always busy resulted in significantly more body weight gain (3.5 and 3.1 kg, respectively) than those who said to be occupied only sometimes, who gained 2.5 kg in weight. They mention that labor requires mental work, promoting overeating and obesity in a modern context.

Similarly, Overgaard et al. [22] performed a literature review to know the association of psychological workload with body weight gain in women and men. Their study showed little evidence for the relationship between these two variables and a weak positive relationship between psychological workload and body weight. Specifically, two studies showed a positive association between body fat distribution in men and women, and there was no evidence that this association was consistent. Also, Booth et al. [47] examined the relationship between psychological workload and body weight gain. To do this, these authors performed a literature review of cross-sectional studies and found that these studies reported a weak positive relationship. Based on this background, the hypothesis H₂ is proposed:

H₂: Mental workload has a direct and positive association with body weight gain in employees from the Mexican maquiladora industry.

In the case of human error, Sheridan [23] defines it as an action that fails to meet some explicit or implicit arbitrary criterion. This author also mentions that mental workload indeed contributes to human error. Roll et al. [48] mention that work-related stress caused by a high mental workload is a significant contributor to human error. Similarly, Qeshmy et al. [49] researched the main causes of human error in assembly production lines, and they discovered that the amount of thinking, deciding, and searching for information were the leading causes of human error, i.e., the leading cause of the human error was mental workload. Other authors mention that high or low physical or cognitive demanding tasks cause human error [50, 51]. In another research, Yan et al. [52] analyzed the correlation between mental workload and the number of mistakes done by a driver. To do this, the authors developed a warning model that helped predict drivers’ mental workload. Their results indicated a positive correlation. Based on these backgrounds, hypothesis H₃ is proposed:

H₃: Mental workload has a direct and positive association with human error in employees from the Mexican maquiladora industry.

1.2 Association of physical fatigue with human error and body weight gain

Several authors have stated that physical fatigue can lead to employees making work errors. For instance, Yu and Wu [53] suggested that physical fatigue can lead to a lack of coordination and flexibility in the employees’ tasks, facilitating human error and negatively impacting product quality and production efficiency.

Similarly, Jamshidi [54] claims that physical fatigue can decrease the human workforce, facilitating human error. Also, Frone and Blais [55] mention that military researchers have focused on physical fatigue results in a Military Non-Deployed Setting, and one of these outcomes is performance errors.

Furthermore, Myszewski [56] showed that human error increases as physical fatigue rises over time. Similarly, Michalos et al. [57] developed a tool to evaluate physical fatigue and another to know the corresponding error rate. Also, Cumber and Greig [58] performed a study in a medical company to determine the effect of physical fatigue caused by changes in shifts on the human error done by doctors. As a result, nearly half of all modifications analyzed in the data demonstrated an increased risk of fatigue-related errors. Sobhani et al. [59] found that human error resulting from insufficient training and other factors inherent to workplace design negatively impacted the efficiency of production and inventory systems. They note that workers in a “in-pain state,” as opposed to a “healthy state,” are more likely to make mistakes, which in turn increases the rate at which defective products are produced.

Therefore, physical fatigue increases the frequency of human error in production-inventory systems. Dantan et al. [60] simulated a manufacturing process that includes the employee’s physical fatigue index and rejected parts due to human error, where a high correlation was found. According to the last paragraph, the hypothesis H₄ is proposed:

H₄: Physical fatigue is directly and positively associated with human error in employees from the Mexican maquiladora industry.

Few studies evaluate the relationship between physical fatigue on body weight gain. However, some authors have performed research in which these two variables are analyzed. For instance, Yamada et al. [61] tested the impact of implementing a 12-hour shift instead of a traditional 8-hour shift in an electronic part production factory. According to that study, a shift to a 12-hour workday dramatically increased workers’ weariness and led to some of them gaining weight. Workers who stayed on the 8-hour shift did not gain any noticeable weight. Based on these backgrounds, the hypothesis H₅ is proposed:

H₅: Physical fatigue is directly and positively associated with body weight gain in employees from the Mexican maquiladora industry.

1.3 Association of body weight gain with human error

There is a physiological explanation of body weight gain on executive function performance (human error). According to several authors, physiological differences between employees with overweight/obesity usually show worse values in insulin resistance, cholesterol levels, blood pressure [62, 63] and levels of glycolic metabolism activation [64]. Because of these underlying physiological differences, people who are overweight may have difficulty with executive functioning, which can lead to ineffectiveness (human error) in task performance.

Some research has linked obesity to cognitive impairment because it alters cerebrovascular blood flow [65, 66], triggers inflammatory processes due to excess body fat [67], and reduces insulin sensitivity [68]. The regions of the brain responsible for executive functioning may undergo either functional [69] (e.g., reduced functional connectivity of executive networks) or structural changes as a result of these disruptions (e.g., reduced orbitofrontal cortex) [70].

Some studies analyze the relationship between body weight and human error, suggesting weight gain may be associated with executive function errors [71]. One of these studies was performed by Sabia [72], studying these problems in the academic performance of adolescents. With information from the National Longitudinal Study of Adolescent Health, Sabia [72] used ordinary least squares, instrumental variables, and individual fixed-effects models. The results showed a statistically significant inverse association between student weight and academic performance.

Anderson and Good [73] studied the correlation between university students’ body weight and academic performance. These authors provided an optional class survey to 452 students for two consecutive academic years to perform the study. The authors compared the overall final grade for the class to the calculated body mass index (BMI), applying linear regression with unpaired t-tests and Pearson’s R correlation. They found a significantly negative correlation between BMI with students’ final grades. As performance improves, errors decrease. Therefore, it can be stated that these studies found a positive relationship between body weight gain and human error.

In a meta-analysis, Yang et al. [71] compared the executive functions of overweight and obese people to those of normal-weight controls. Overweight and obese participants made more mistakes in executive function tasks, such as those that required inhibition, planning, verbal fluency, cognitive flexibility, working memory, and decision-making, according to a meta-analysis of 72 studies involving 4904 people. In contrast, the overweight participants saw substantial errors in working memory and inhibition.

H₆: Body weight gain is directly and positively associated with human error in employees from the Mexican maquiladora industry.

The hypothetical model with the hypotheses mentioned above is shown in Fig. 1.

Fig. 1

Relationships of the hypothetical causal model.

2 Methodology

2.1 Stage 1. Mental workload questionnaire (MWLQ) development

Information from the maquiladora industries is required to validate the hypotheses in Fig. 1. According to Stanton et al. [74], the mental workload can be assessed using methods such as Subjective Workload Assessment Technique (SWAT), Modified Cooper Scales (MCH), DRA Workload Scales (DRAWS), Subjective Workload Dominance Technique (SWORD), NASA Task Load Index (NASA-TLX), among others; however, the NASA Task Load Index (NASA-TLX) is the most widely used [75 –78]. The NASA-TLX method was developed as a heuristic tool for asses the mental effort required for tasks like air traffic management, maintaining vigilance, and practicing flights. It has been more popular as a subjective verified workload assessment scale used in various fields, including transportation, cognitive psychology, human-computer interaction, janitors [78 –80], nuclear energy, transportation, healthcare, and manufacturing [76 , 81–85].

The NASA-TLX questionnaire continues being applied in so many recent studies from 2021 and 2022, as those performed by Sharifi et al. [86], Abareshi et al. [87], da Silva et al. [88], Ebrahimi et al. [89], Han et al. [90], and Moertl et al. [91]. NASA-TLX is associated with accuracy [92 –98], completion time [92 , 99–101], and reaction time [92 , 102]. Furthermore, according to a recent study, the overall mental workload scores in the NASA-TLX gradually increase with increasing levels of mental workload, and the overall scores of low mental workload (LMW) were significantly lower than those of medium mental workload (MWL) (p 0.001) and high mental workload (HMW) (p 0.001), and the overall scores of MWL were significantly lower than those of HMW (p 0.001) [103].

Pang et al. [103] discuss the use of electroencephalography (EEG) to measure the level of mental workload, noting that it takes a long time to calculate and train, that cumulative errors can occur, that the algorithms used are difficult to converge, and that it is relatively easy to tend to a local minimum. Furthermore, these authors mention recent studies on assessing mental workload with EEG, but their accuracy was insufficient.

To collect information from industries, the mental workload questionnaire (MWLQ) was applied and uploaded to the Google forms platform, and it comprises three sections. The MWLQ was developed by merging the questions in the NASA-TLX method with questions about descriptive data (age, relationship status, number of children), the number of employees working in the company, job position, years of experience, the industry sector in the company; and the effects of mental load that the participants have experienced. The first section of the MWLQ addresses the demographic data as the number of employees in the company, job position, years of experience, and industry sector.

The second section of the MWLQ aims to assess employees’ mental workload using the NASA-TLX [75 –78], and it is divided into two subsections; the first section evaluates each dimension individually (mental demand, physical demand, time demand, performance, effort, and frustration). Employees must select a value between 1 = Low and 20 = High (only for the performance dimension, 1 = Good and 20 = Poor). The second subsection of NASA-TLX contains the paired comparison, in which participants must choose the dimension that contributes the most to increasing mental workload during a specific task for each paired comparison.

Once employees answer the second section of the MWLQ (i.e., the questions corresponding to the NASA-TLX), the individual mental workload is calculated by applying equation (1). Table 1 shows how mental workload levels are converted to an ordinal scale.

Table 1
Ordinal scale for mental workload levels

Ordinal scale

Very Low (1) Low (2) Medium (3) High (4) Very high (5)

Mental workload level 1–20 21–40 41–60 61–80 81–100

	Ordinal scale
Mental workload level	1–20	21–40	41–60	61–80	81–100

$NASA - TLX = \frac{1}{15} \sum_{1}^{6} d_{i} w_{i}$ (1)

Specifically, the overall index is obtained by applying equation (1), which contains the sum of relative contributions of each dimension, representing the mental workload accumulated by a person during a specific event, situation, or time [77, 81]. Equation (1) provides a weighted average representing the final mental workload index based on the subjective rating of each attribute di and the corresponding weights w_i for the six dimensions. The weights w_i represent the number of preferences (i.e., how many times each dimension was chosen) in the 15 paired comparisons and range from 0 (not significant) to 5 (significantly more significant than any other attribute) [78].

The third section of the MWLQ addresses the association of mental workload with some observed variables on employees and their occurrence. Among the observed variables are physical fatigue, body weight gain, human error, stress, headache, mental fatigue, irritability, depression, memory loss, and social problems. The MWLQ then includes 11 observed variables as well as mental workload. However, the present research only estimates the association of mental workload with the first three observed variables. The reasons to include only these three variables are the following:

In Mexico, obesity is a current problem that has been present for a long time (since 2002 or even before) [33]. Moreover, the country ranks first globally in overweight and obesity [34], and nationally, in 2018, the percentage of adults over 20 years old with overweight and obese was 75.2% (39.1% overweight and 36.1% obese) [35]. These statistics show that in Mexico, most employees tend to gain weight easily.

Physical fatigue is supposed to reduce physical activity, which in turn is supposed to promote body weight gain.

Physical fatigue is supposed to promote human error.

Although some research has been done in the Mexican manufacturing industry and worldwide regarding mental workload, none has explored the association of mental workload with physical fatigue, body weight gain, and human error simultaneously.

The research aims to discover new things in different contexts (countries, variables (both in quantity and in significance), and work sectors, to name a few). This research represents an exploratory study on the effects of mental workload on other variables, which may act as dependent variables or mediating variables between mental workload and some independent variable(s).

The frequency of occurrence of the observed variables is measured using a Likert-type scale of 5 points (Never = 1, Rarely = 2, Sometimes = 3, Often = 4, Always = 5) since it has been successfully used in other similar studies [104]. In this task, the employee must select a frequency level for each association of mental workload with the observed variables, and all questions are mandatory. Table 2 shows the observed variables and the scale used to measure their association with the mental workload, and a digital version of the MWLQ is available at https://forms.gle/AwxwGzvxz3teVVwb7.

Table 2

Likert scale for the association of observed variables with mental workload

Variable	Likert scale
	Never (1)	Rarely (2)	Sometimes (3)	Often (4)	Always (5)
Physical fatigue
Body weight gain
Human error

In this research, we use this subjective scale because it has been reported by Bakker et al. [105] to know the subjective validity measures (they focused only on sedentary behavior) with self-reported measurement tools (including questionnaires). They found that the reliability of measures was moderate-to-good, with the quality of these studies being mostly fair-to-good. It justifies its use to ensure reliability ranges from moderate to good in this study.

2.2 Stage 2. MWLQ application

The MWLQ was applied using a convenience sample method aimed at middle and senior managers because they have a high cultural level and knowledge about mental workload and its association with the observed variables. Managers were distributed into 63 maquiladora companies established in the north of Mexico. The survey was applied according to the following two steps:

The first step is to contact the managers of companies by using the Liaison Office directory of a higher education institution and the National Institute of Statistics, Geography, and Informatics (INEGI). A manager is contacted by telephone or email in each company. Managers know about the MWLQ survey, items, and scales at this step. Therefore, an interview was arranged to respond to it, inviting them to collaborate with the project and informing the research objectives and benefits they may obtain.

The identified managers contact their colleagues to inform them about the project and the date, time, and place to respond to the survey. The MWLQ administration was performed for six months, according to the managers’ availability in each maquiladora. All surveyed participants were called to collaborate in the study, willingly respond to the survey, and sign an agreement letter. The total sample includes 167 surveyed employees distributed in the 63 manufacturing companies, as is shown in Table 3.

Table 3
Employees surveyed by company

Company NES Company NES Company NES

1 3 22 2 43 1

2 1 23 3 44 1

3 1 24 5 45 1

4 1 25 1 46 1

5 1 26 1 47 1

6 1 27 1 48 3

7 1 28 1 49 5

8 10 29 1 50 2

9 1 30 1 51 1

10 1 31 1 52 2

11 1 32 1 53 4

12 13 33 3 54 20

13 1 34 1 55 1

14 1 35 1 56 1

15 1 36 4 57 2

16 1 37 2 58 1

17 1 38 2 59 2

18 1 39 4 60 1

19 15 40 1 61 1

20 4 41 15 62 2

21 1 42 1 63 1

Total 61 Total 52 Total 54

Company	NES	Company	NES	Company	NES
1	3	22	2	43	1
2	1	23	3	44	1
3	1	24	5	45	1
4	1	25	1	46	1
5	1	26	1	47	1
6	1	27	1	48	3
7	1	28	1	49	5
8	10	29	1	50	2
9	1	30	1	51	1
10	1	31	1	52	2
11	1	32	1	53	4
12	13	33	3	54	20
13	1	34	1	55	1
14	1	35	1	56	1
15	1	36	4	57	2
16	1	37	2	58	1
17	1	38	2	59	2
18	1	39	4	60	1
19	15	40	1	61	1
20	4	41	15	62	2
21	1	42	1	63	1
Total	61	Total	52	Total	54

*NES=Number of employees surveyed.

The approval for this research was obtained from the Ethical Committee of Instituto Tecnológico de Tijuana (Technological Institute of Tijuana at Tijuana, Mexico). From all participants, electronic informed consent was obtained following the Helsinki Declaration.

2.3 Stage 3. Data analysis

2.3.1 MWLQ statistical debugging and validation

Data analysis comprised two steps. The first step refers to MWLQ statistical validation, whereas in the second step, all hypotheses in Fig. 1 were statistically tested.

Data from questionnaires was entered into a database created with the SPSS® software [106]. Since the data were collected on a Likert scale (ordinal scale), the missing values and outliers were substituted by the median to improve their quality [107 –109]. It is important to indicate that all estimations were analyzed with a confidence level of 95%.

The MWLQ is then statistically validated for each variable using Cronbach’s alpha, which occurs when it exceeds 0.7 [107, 110]. According to Tavakol and Dennick [111], Cronbach’s alpha measures the internal consistency of a test or scale. Internal consistency describes the extent to which all items in a test measure the same concept or construct and is thus related to the test’s interrelationship. These authors mention that to avoid this, homogeneity or unidimensionality can help improve alpha use. A measure is said to be unidimensional if its items measure a single latent trait or construct [111]. All observed variables with a corrected total-item correlation value lower than 0.3 are removed from the analysis in this step.

2.3.2 Hypotheses testing

The relationships between the observed variables, as shown in Fig. 1, are tested using mediation analysis because observed variables can act as both antecedent and consequent variables [112]. The PROCESS® macro (model 6) in the SPSS® software is used for the analysis because it analyzes the data of observed variables with sophisticated algorithms based on ordinary least squares (OLS) regression [112, 113] and allows the use of nonlinear models [110]. Several authors widely recommend this software for statistical analysis with small samples, assessments on an ordinal scale, and non-normal distribution [114]. In a regression analysis, the validity of statistical inferences was substantially affected only when the most severe violations of normality occurred or when the sample size was quite small [112, 115].

Some model fit indexes [112] were estimated before model interpretation [112]: R² and F ratio. R² is a scale-free metric defined as the proportion of variance in a subsequent variable Y explained by the model [112], with values greater than 0.02 accepted. Finally, the F ratio is used to determine whether the variance explained by the model differed from the variance not explained by the model significantly. It is considered appropriate when a model is statistically significant or when the variance ratio differs significantly from zero [116]. The intraclass correlation coefficient (ICC) is used to analyze data concordance, accepting ICC >0.71 [117]. The analysis is performed in SPSS®, using the two-way mixed model and the absolute agreement type.

The serial multiple mediation model was applied to measure the associations. Hypotheses in Fig. 1 were validated by using direct associations. A generic interpretation of the direct association was when two cases differed by one unit on an antecedent variable X and were estimated to differ by c’ (the value of the direct association) units on a subsequent unit or subsequent variable Y [112].

Three connections are examined in detail. First, a standardized value was obtained as a dependence measure to statistically test the hypothesis that there is a relationship between two observed variables [107]. The null hypothesis is H₀: β₁ =0 versus the alternative H₁: β₁ ≠ 0. Evidence of a correlation between the two variables can be inferred statistically if the value is non-zero.

Second, indirect connections manifest themselves throughout more than one segment (i.e., more than one antecedent and more than one subsequent variables) when one or more mediator factors exist between the antecedent and subsequent variables [107]. When an antecedent variable X changes by one unit because of its association with a mediator variable M, which impacts Y, these values were read as the expected change value of the consequent variable Y [118].

Finally, the sum of direct and indirect associations is called the total association. For the validity test of each association, the bootstrap method was used [119]. Several authors have found that the bootstrap technique yield provides the most reliable findings for determining whether or not an indirect relationship is statistically significant. [119, 120]. Finally, the significance of direct, indirect, and total associations was tested with a 95% confidence level and ten thousand bootstrap samples. If zero is not in the 95% confidence interval, all associations are significant at p 0.05 [119, 121].

3 Results and discussion

This section shows findings derived from the statistical model analysis. The section is structured following the steps of the methodology described above.

3.1 Sample description

Of the 167 employees surveyed, 100 were men, while 67 were women, ranging from 19 to 56 years old. Regarding the industrial sector in which they worked, 44 respondents mentioned that they belonged to the medical sector, 42 to the electricity sector, and 43 to the automotive sector. To a lesser extent, other respondents indicated that they belonged to the food, information technology, wood, plastic and rubber, metallic, or textile sector. Finally, regarding the work schedule, results indicated that 139 respondents work more than 35 hours per week, while 28 participants report they work 35 hours or less per week. Table 4 shows the descriptive results of the sample.

Table 4
Descriptive results of the sample

Job position Industrial sector Gender Age range

Managers 27 (16.17%) Medical 44(26.35%) Men 100 (59.88%) 19–25 63 (37.72%)

Engineers 66 (39.52%) Automotive 43 (25.75%) Women 67 (40.12%) 26–30 29 (17.37%)

Supervisors 61 (36.53%) Electricity 42 (25.15%) 31–35 35 (20.96%)

Group leaders 13 (7.78%) Other sectors 38 (22.75%) 36–40 12 (7.19%)

41–45 8 (4.79%)

46–50 9 (5.39%)

>50 11 (6.59%)

Job position	Industrial sector	Gender	Age range
Managers	27 (16.17%)	Medical	44(26.35%)	Men	100 (59.88%)	19–25	63 (37.72%)
Engineers	66 (39.52%)	Automotive	43 (25.75%)	Women	67 (40.12%)	26–30	29 (17.37%)
Supervisors	61 (36.53%)	Electricity	42 (25.15%)			31–35	35 (20.96%)
Group leaders	13 (7.78%)	Other sectors	38 (22.75%)			36–40	12 (7.19%)
						41–45	8 (4.79%)
						46–50	9 (5.39%)
						>50	11 (6.59%)

3.2 MWLQ validation

As all variables integrated into the model were observed variables, no items were used to measure them (there were no Cronbach’s alphas for individual scales for each observed variable), but only one scale for the MWLQ. The Cronbach’s alpha value from the MWLQ was 0.759, and no items were removed because Cronbach’s alpha value decreased. This relatively low value is due to the items in the MWLQ measuring different observed variables (physical fatigue, body weight gain, and human error) and not only construct, i.e. but there also is no unidimensionality. But because the test length affects Cronbach’s alpha, a high value does not always mean a high degree of internal consistency [111]. If the test length is too short, the alpha value is reduced [111 , 123]. So, to raise alpha, the test should have more related questions that test the same idea [111].

The MWLQ can be reliable based on these results, and the correlation matrix is shown in Table 5.

Table 5
Correlation matrix

Mental Physical Body weight Human

workload fatigue gain error

Mental workload 1.000 0.179 0.330 –0.104

Physical fatigue 0.179 1.000 0.213 –0.057

Body weight gain 0.330 0.213 1.000 –0.016

Human error –0.104 –0.057 –0.016 1.000

	Mental	Physical	Body weight	Human
Mental workload	1.000	0.179	0.330	–0.104
Physical fatigue	0.179	1.000	0.213	–0.057
Body weight gain	0.330	0.213	1.000	–0.016
Human error	–0.104	–0.057	–0.016	1.000

Regarding the ICC, a value of 0.82 was obtained at a confidence interval of 95%, then concordance among data was good.

3.3 Analysis of the hypothetical causal model: direct associations among variables

In Fig. 2, the β values represent the direct associations between an independent and dependent variable. Similarly, the p-values represent the significance of the associations used to test the hypothesis. Observe that three p-values are over 0.05 (red arrows). Then, the relationship between mental workload and body weight gain can be rejected. This result is similar to that obtained by Overgaard et al. [22], who conducted a literature review on the association between mental workload and body weight, finding little evidence of a general association between these two variables. The relationship between physical fatigue and human error can also be rejected. The relationship between physical fatigue and human error can also be rejected. This result differs from that mentioned by Kolus et al. [124], who developed a model in which fatigue states represent a mismatch between the technical system and the human operator, resulting in manufacturing errors.

Fig. 2

Direct associations.

This difference may be because the Kolus et al. [124] model focuses on operators, not middle and senior management, as in this research. Finally, the relationship between body weight gain and human error can be rejected. This result differs from Yang et al. [71], who state that overweight workers make errors in executive functions. This difference can be because these authors performed a literature review and synthesized it by conducting a meta-analysis of studies that included employees from different countries worldwide, not only from Mexico, as we do in this research.

The p-value for the other three relationships (black arrows) was less than 0.05, so they were significant. Only significant direct associations are shown in Fig. 3. The relationship between mental workload and physical fatigue is similar to the one obtained by Wu et al. [125], who found that a high mental workload increases fatigue in nurses. Also, the result obtained on the relationship between mental workload and human error is similar to the one obtained by Yeow [126], who found that mental workload has a significant relationship with human error in manufacturing employees in Malaysia. Finally, the relation between physical fatigue and body weight gain agrees with that obtained by Yamada et al. [61], who reported that when employees work a 12-hour shift instead of an 8-hour shift, it significantly increases fatigue and body weight gain.

Fig.3

Significant direct associations.

The β values can be seen as measures of how dependent something is. For example, the β values in Fig. 3 show that if the standard deviation of physical fatigue went up by one unit, the standard deviation of body weight gain went up by 0.304 units. Figure 2 shows a serial model with multiple mediators. Table 6 shows the regression coefficients (Coeff.), standard errors (SE), constants, and a summary of the model.

Table 6

Regression coefficients, standard errors, constant and summary information for the serial multiple mediator model

	Consequent
	Physical fatigue			Body weight gain			Human error
Antecedent	Coeff.	SE	p	Coeff.	SE	p	Coeff.	SE	p
Mental workload	0.374	0.13	0.0153	–0.0019	0.1783	0.99	–0.476	0.1014	0.0021
Physical fatigue				0.304	0.1050	0.0001	0.144	0.0626	0.733
Body weight gain							0.113	0.0444	0.1545
Constant	1.858	0.3284	0	1.46	0.484	0.003	2.79	0.2827	0
Model summary	R² = 0.035			R² = 0.0921			R² = 0.0843
	F(1,165)=6.0075, p = 0.0153			F(2,164)=8.3211, p = 0.0004			F(3,163)=6.0075, p = 0.0024

The R² values represent the percentage in which antecedent variables explain variance for consequent variables [112].

Take note that the p-value for the F ratios was less than 0.05, and the R² was greater than 0.02, so the model is suitable. Equations expressing dependent relationships for physical fatigue, body weight gain, and human error can be derived from the values provided in Fig. 3 in equations 2 to 4 and the constant values shown in Table 3 [112]: $Physical fatigue = 1 . 58 + 0 . 374 \times Mental workload$ (2) $Body weight gain = 1 . 46 + 0 . 304 \times Physical fatigue$ (3) $Human error = 2 . 79 - 0 . 476 \times Mental workload$ (4)

The study did not integrate other factors that may contribute to human error in manufacturing, including employees’ physical and mental characteristics, the nature of the work site, the nature of the materials used, the nature of the machinery employed, and the nature of the standards used to pre-program the tasks to be performed [56].

3.4 Analysis of the hypothetical causal model: indirect association among variables

There were three indirect associations between mental workload with human error. The first indirect association (indirect association 1) was passing through the mediator variable of physical fatigue; the second indirect association (indirect association 2) was through the body weight gain; and the third indirect association (indirect association 3) was through the two mediator variables, going first through the physical fatigue and then through the body weight gain variable. The highest indirect association was through the physical fatigue to human error variable.

Since zero fell inside the confidence interval [Lower Limit of Confidence Interval (LLCI), Upper Limit of Confidence Interval (ULCI)], neither the total indirect connection nor the individual indirect associations were statistically significant [112]. Table 7 illustrates the values and confidence intervals for indirect connections between cognitive load and human mistakes.

Table 7
Indirect associations from mental workload with human error and their confidence intervals

Indirect association Value LLCI ULCI

Indirect association 1 0.0539 –0.0103 0.1492

Indirect association 2 –0.0002 –0.0467 0.0482

Indirect association 3 0.0128 –0.0088 0.0429

Total indirect association 0.0665 –0.0144 0.1649

Indirect association	Value	LLCI	ULCI
Indirect association 1	0.0539	–0.0103	0.1492
Indirect association 2	–0.0002	–0.0467	0.0482
Indirect association 3	0.0128	–0.0088	0.0429
Total indirect association	0.0665	–0.0144	0.1649

3.5 Analysis of the hypothetical causal model: total associations among variables

The sum of direct and indirect associations is called the total association [112]. The mental workload had a significant negative total association of –0.4095 on human error, with a p-value of 0.0078. Then, this association was significant at the 99.9% confidence level. This total association indicated that when mental workload increases its standard deviation by one unit, the human error decreases by –0.4095 units.

4 Conclusions and recommendations

4.1 Conclusions related to the hypotheses

Figure 2 and Table 3 show that human error is significantly and directly associated only with the mental workload and not with any other variable. It is concluded that mental workload has a positive indirect association but a negative direct and total association with human error. Also, the mental workload has a direct association with physical fatigue. However, the mental workload has no direct association with body weight gain. Then, it can be concluded that mental workload does not cause body weight gain among manufacturing managers. Also, Fig. 2 and Table 3 show that body weight gain has no association with human error. Then, it can be concluded that body weight gain does not negatively affect the employees’ performance.

Similarly, human error is not directly affected by physical fatigue. However, physical fatigue has the highest direct association with body weight gain. It can be concluded that combining mental workload, physical fatigue, and body weight gain there will not negatively affect the efficiency of employees in production systems.

According to the results shown in section 4, the following conclusions are inferred about the hypotheses proposed in section 2:

H₁: There is enough statistical evidence to declare that the mental workload on employees of manufacturing systems has a direct positive association with their physical fatigue since when the standard deviation of mental workload increases by one unit, the standard deviation of physical fatigue increases by 0.374 units.

H₂: There is not enough statistical evidence to state that the mental workload on employees of manufacturing systems has a direct positive association with their body weight gain.

H₃: There is enough statistical evidence to declare that mental workload on employees in the manufacturing system is directly and positively associated with human error, but directly negatively, since if mental workload increases its standard deviation by one unit, the human error decreases by 0.476 units.

H₄: There is not enough statistical evidence to declare that physical fatigue in employees of manufacturing systems is positively associated with human error.

H₅: There is enough statistical evidence to declare that physical fatigue in employees of manufacturing systems is directly and positively associated with their body weight gain since if the standard deviation of physical fatigue increases by one unit, the standard deviation of body weight gain increases by 0.304 units.

H₆: There is not enough statistical evidence to declare that body weight gain in employees of manufacturing systems has a direct positive association with human error.

Meanwhile, the p-values of hypotheses H₂ and H₆ were over 0.05, and then there is not enough statistical evidence to declare that mental workload has a direct association with body weight gain, nor does this variable on the human error in the sample analyzed. Also, hypothesis H₃ was not statistically tested since the direct association was not positive but negative on the sample analyzed. This may be because most values of mental workload were under 50. Then, most surveyed employees did not feel a significant mental workload and did not make human errors.

As all results obtained for R² were higher than 0.02 (i.e., 0.035, 0.092, 0.084), the variance explained by the model is accepted. Also, for all the F ratios, the p-values were <0.05, then the variance explained by the model differs significantly from the variance not explained by the model. The methodology presented here is original since, according to the literature, no other research has combined the associations of mental workload, physical fatigue, and body weight gain with human error from employees in manufacturing systems. Regarding the MWLQ, statistical analysis shows that it is an effective and reliable new instrument to collect information about mental workload and its effects on employees from the manufacturing system.

4.2 Industrial recommendations

First, the authors recommend that companies in any labor sector, including manufacturing, consider their employees’ mental workload, physical fatigue, and body weight when designing tasks since these variables may influence their performance. More specifically, this paper recommends considering employees’ capabilities and limitations to ensure their satisfaction, motivation, and creativity and improve their performance. If tasks are designed based on these recommendations, companies can increase their competitive advantage since a correct mental and physical workload means motivation, creativity, and better performance.

4.3 Future research

For future research along the same line, it is suggested to explore mental workload levels and their effects on other industrial sectors, such as supermarkets, construction, education, health, and commerce. Moreover, it is important to determine employees’ mental workload, physical fatigue, and body weight gain and its associations. In addition, it is relevant to apply ergonomic tools to decrease these variables to improve employees’ well-being and performance. Also, the authors recommend improving the MWLQ survey by adding items that help explain the different types of variables, as it is done with mental workload through its six dimensions. Then, as future research, it is suggested to convert the observed variables into latent variables and measure Cronbach’s alpha individually for each latent variable.

Also, the authors suggest researching new variables, such as overwork, loss of memory, depression and social problems, to mention a few. This would allow for the generation of new hypothetical models and research regarding their associations, combining mental workload with overwork to know their associations.

In this line, it is important to mention that the fatigue analyzed in this research is treated as general physical fatigue. However, a specific type of fatigue has recently been relevant: Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS). ME/CFS is a long-term disabling, poorly understood illness accompanied by rest-induced fatigue. Currently, there is no universally accepted treatment and pathophysiology for ME/CFS [127]. Then, for future research, the authors suggest analyzing the associations of ME/CFS in manufacturing employees by applying the methods used by Josev et al. [128].

Finally, it is recommend to use reliable self-reported measurement tools to ensure more reliable results of the model. Such tools may include Swedish Occupational Fatigue Inventory (SOFI) [129] and the Multi-dimensional Fatigue Index (MFI) [130] to measure fatigue instances.

Conflict of interest

None to report.

Ethical approval

The study was approved by the Ethical Committee of Instituto Tecnológico de Tijuana (Technological Institute of Tijuana at Tijuana, Mexico) (no. ERN-2020-0001, date 8/4/2020).

Informed consent

Before answering the survey, each participant read the following statement to give her/his consent to use the data she/he provided: “I have read the information and purpose of this research in the academic sector. I understand that my participation is voluntary and I give my consent to take the proposed survey”.

Footnotes

Acknowledgments

The authors want to thank the Instituto Tecnológico de Tijuana (Technological Institute of Tijuana) and the Universidad Autónoma de Ciudad Juárez (Autonomous University of Ciudad Juarez) for the support provided by means of the use of their installations and technology. Also, the authors want to thank to all employees who were surveyed during this research, and the manufacturing companies where they work.

Funding

None to report.

References

Zare

, Black

, Sagot

J-C

, Hunault

, Roquelaure

. Ergonomics interventions to reduce musculoskeletal risk factors in a truck manufacturing plant. Int J Ind Ergon. 2020;75:1–12.

Boet

, Sharma

, Pigford

A-A

, Hladkowicz

, Rittenhouse

, Grantcharov

. Debriefing decreases mental workload in surgicalcrisis: A randomized controlled trial. Surgery [Internet]. 2017 May 1 [cited 2018 Sep 25];161(5):1215–20. Available from: https://www-sciencedirect-com-443.web.bisu.edu.cn/science/article/pii/S0039606016308005

Leung

GTC

, Yucel

, Duffy

. The effects of virtual industrial training on mental workload during task performance. Hum Factors Ergon Manuf Serv Ind. 2010;20(6):567–78.

Salas

, Maurino

. Human Factors in Aviation [Internet]. 2nd ed. San Diego, CA, USA: Gulf Professional Publishing; 1988. 684 p. Available from: https://books.google.com.mx/books?hl=es&lr=&id=Fi2Bqh_6fW4C&oi=fnd&pg=PT1&ots=wMOh3QytxD&sig=x3dUbOAfTX75N0IMXgX1faEwdxo&redir_esc=y#v=onepage&q&f=false

Endsley

, Garland

. Situation Awareness Analysis and Measurement [Internet]. Florida: CRC Press; 2000. 408 p. Available from: https://books.google.com.mx/books?id=WrJGDsjJakcC&lr=&hl=es&source=gbs_navlinks_s

Lee

, Liu

. Inflight workload assessment: Comparison of subjectiveand physiological measurements. Aviat Space Environ Med [Internet]. 2003 Nov 18 [cited 2019 Feb 1];74(10):1074–84. Available from: https://www.ingentaconnect.com/content/asma/asem/2003/00000074/00000010/art00011

Mosier

, Rettenmaier

, McDearmid

, Wilson

, Mak

, Raj

, et al. Pilot–ATC Communication Conflicts: Implications for NextGen. Int J Aviat Psychol [Internet]. 2013 Jul [cited 2019 Feb 1];23(3):213–26. Available from: http://www.tandfonline.com/doi/abs/10.1080/10508414.2013.799350

Plane Crash Info. Accident statistics [Internet]. Causes of Fatal Accidents by Decade. 2013 [cited 2019 Feb 1]. Available from: http://planecrashinfo.com/cause.htm

Bashiri

, Mann

. Impact of Automation on Drivers’ Performance inAgricultural Semi-Autonomous Vehicles. J Agric Saf Health [Internet]. 2015 [cited 2019 Feb 1];21(2):129–39. Available from: https://mspace.lib.umanitoba.ca/bitstream/handle/1993/32115/mann_danny.pdf?sequence=1

10.

Xiao

, Li

, Ren

, Wang

, et al. Effects of mental workload on work ability in primary and secondary school teachers. Chinese J Ind Hyg Occup Dis [Internet]. 2015 Feb [cited 2019 Feb 6];33(2):93–6. Available from: http://www.ncbi.nlm.nih.gov/pubmed/25916354

11.

Mohammadian

, Malekpour

, Zoghipour

, Malalekpour

. Study on Mental Workload of Teachers and Its Correlation with Their Quality of Life. J Educ Manag Stud [Internet]. 2015 [cited 2019 Feb 6];5(2):144–9. Available from: www.science-line.com

12.

Barajas-Bustillos

, Maldonado-Macías

, García-Alcaraz

, Hernández-Arellano

, Avelar-Sosa

. Considerations of the Mental Workload in Socio-Technical Systems in the Manufacturing Industry: A Literature Review. In: Realyvásquez A, Maldonado-Macías AA, Arredondo Karina C, editors. Advanced Macroergonomics and Sociotechnical Approaches for Optimal Organizational Performance. 1st ed. Hershey, Pensilvania; 2019. p. 99–117.

13.

Jin

, Zheng

, Pei

, Li

. A Method to Estimate Operator’s Mental Workload in Multiple Information Presentation Environment of Agricultural Vehicles. In: Harris D, editor. Engineering Psychology and Cognitive Ergonomics: Performance, Emotion and Situation Awareness. Vancouver: Springer; 2017. p. 3–20.

14.

Young

, Brookhuis

, Wickens

, Hancock

. State of science:mental workload in ergonomics. Ergonomics [Internet]. 2015 Jan 2 [cited 2019 Feb 1];58(1):1–17. Available from: http://www.tandfonline.com/doi/abs/10.1080/00140139.2014.956151

15.

Moriarty

, Jarvis

. A systems perspective on the unstable approachin commercial aviation. Reliab Eng Syst Saf [Internet]. 2014 Nov 1 [cited 2019 Feb 1];131:197–202. Available from: https://www-sciencedirect-com-443.web.bisu.edu.cn/science/article/pii/S0951832014001495

16.

Wiegmann

, Shappell

. A Human Error Approach to Aviation Accident Analysis. The Human Factors Analysis and Classification System [Internet]. London: Routledge; 2017 [cited 2019 Feb 1]. 184 p. Available from: https://www-taylorfrancis-com-443.web.bisu.edu.cn/books/9781351962360

17.

Zoaktafi

, Kazemi

, Choobineh

, Sabooryaraghi

, Nematolahi

, Zakerian

. Relationship between mental workload and salivary cortisol levels: A field study. Work. 2020;67(2):381–6.

18.

Hernández-Arellano

, Ibarra-Mejía

, Serratos-Pérez

, García-Alcaraz

, Brunette

. Construction of a survey toassess workload and fatigue among AMT operators in Mexico. Work. 2012;41(Supplement 1):1790–6.

19.

Hernández-Arellano

, Castillo

, Serratos-Perez

, García-Alcaraz

. Relationship between Workload and Fatigue among Mexican Assembly Operators. Int J Phys Med Rehabil. 2015;3(6).

20.

Hernández-Arellano

, Serratos-Perez

, Garcia-Alcaraz

, Maldonado-Macias

. Assessment of Workload, Fatigue, and Musculoskeletal Discomfort Among Computerized Numerical Control Lathe Operators in Mexico. IISE Trans Occup Ergon Hum Factors. 2017;5(2):65–81.

21.

Mehta

, Agnew

. Influence of mental workload on muscle endurance, fatigue, and recovery during intermittent static work. Eur J Appl Physiol. 2012;112(8):2891–902.

22.

Overgaard

, Gyntelberg

, Heitmann

. Psychological workload and body weight: is there an association? A review of the literature. Occup Med (Chic Ill) [Internet]. 2004 Jan 1 [cited 2019 Nov 8];54(1):35–41. Available from: https://academic-oup-com-s.web.bisu.edu.cn/occmed/article-lookup/doi/10.1093/occmed/kqg135

23.

Sheridan

. Risk, human error, and system resilience: fundamental ideas. Hum Factors [Internet]. 2008 Jun [cited 2020 Jan 9];50(3):426. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18689048

24.

Sedighi-Maman

, Alamdar-Yazdia

, Cavuoto

, Megahed

. A data-driven approach to modeling physical fatigue in the workplace using wearable sensors. Appl Ergon. 2017;65:515–29.

25.

, Phillips

, Fitzpatrick

. The relationship of body mass index (BMI) to job performance, absenteeism and risk of eating disorder among hospital-based nurses. Appl Nurs Res. 2019;49:77–9.

26.

Di Pasquale

, Miranda

, Neumann

, Setayesh

. Human Reliability in Manual Assembly Systems: a Systematic Literature Review. In: IFAC (International Federation of Automatic Control). 2018. p. 675–80.

27.

Realyvásquez

, García-Alcaraz

, Blanco-Fernández

. Development and validation of a macroergonomic compatibility questionnaire [in Spanish]. Contaduría y Adm. 2016;61(3):478–98.

28.

Instituto Nacional de Estadística Geografía eInformática (INEGI). Directorio Nacional de Unidades Económicas (DENUE) [Internet]. 2022 [cited 2020 Jul 5]. Available from: http://www3.inegi.org.mx/sistemas/mapa/denue/

29.

Gómez-Vega

. El desarrollo de la industria de la maquila enméxico. Probl del Desarro Rev Latinoam Econ [Internet]. 2004;35(138):57–83. Available from: https://www.redalyc.org/articulo.oa?id=11825948004

30.

Instituto Nacional de Estadística Geografía eInformática (INEGI). Clasificación de Actividades Económicas de la Encuesta Nacional de Empleo (CAE-ENE). 1st ed. Aguascalientes: INEGI; 2005. 179 p.

31.

Barajas

, Almaraz

, Carrillo

, Hualde

, Rodríguez

, Contreras

. Industria Maquiladora en México: Perspectivas del Aprendizaje Tecnológico-Organizacional y Escalamiento Industrial. Tijuana; 2003.

32.

Instituto Nacional de Estadística Geografía eInformática (INEGI). Manufacturing Statistics and Export Maquiladora [in Spanish] [Internet]. Manufacturing Statistics and Export Maquiladora [in Spanish]. Available from: 2020. https://www.inegi.org.mx/temas/manufacturasexp/

33.

Ramírez

, Grijalva-Haro

, Ponce

, Valencia

. Prevalenciade sobrepeso y obesidad en el noroeste de México por tres referencias de índice de masa corporal: diferencias en laclasificación. Arch Latinoam Nutr [Internet]. 2006 [cited 2021 Sep 9];56(3). Available from: http://ve.scielo.org/scielo.php?pid=S0004-06222006000300007&script=sci_arttext

34.

Manzano

, López-Hernández

. Relación del sobrepeso yobesidad con el rendimiento laboral en trabajadores de una empresa metalmecánica en México. Rev Colomb Salud Ocup. 2016;6(4):103–8.

35.

Instituto Nacional de Estadística Geografía eInformática (INEGI). Encuesta Nacional de Salud y Nutrición 2018. Presentación de resultados [Internet]. 2018. Available from: https://www.inegi.org.mx/contenidos/programas/ensanut/2018/doc/ensanut_2018_presentacion_resultados.pdf

36.

Secretaría del Trabajo y Previsión Social. Marco normativo de seguridad y salud en el trabajo. Marco normativo de seguridad y salud en el trabajo. 2012.

37.

Secretaría de Salud. Condiciones generales de trabajo de la Secretaría de Salud [Internet]. Condiciones generales de trabajo de la Secretaría de Salud. Mexico City; 2016. p. 83. Available from: https://webcache.googleusercontent.com/search?q=cache:uXhO3DkxHVYJ:https://www.inr.gob.mx/Descargas/acercade/CGT_de_la_SSA%25202016.pdf+&cdl=es&ct=clnk&gl=mx=3l=es&ct=clnk&gl=mx&h

38.

Reyes

, de la Riva

, Maldonado

, Woocay

, de La

. Association between Human Error and Occupational Accidents’ Contributing Factors for Hand Injuries in the Automotive Manufacturing Industry. Procedia Manuf. 2015;3:6498–504.

39.

Tao

, Tan

, Wang

, Zhang

, Qu

, Zhang

. A Systematic Review of Physiological Measures of Mental Workload. Int J Environ Res Public Heal. 2019, Vol 16, Page 2716 [Internet]. 2019 Jul 30 [cited 2021 Aug 12];16(15):2716. Available from: https://www.mdpi.com/1660-4601/16/15/2716/htm

40.

Anwer

, Li

, Antwi-Afari

, Umer

. Evaluation of Physiological Metrics as Real-Time Measurement of Physical Fatigue in Construction Workers: State-of-the-Art Review. J Constr Eng Manag. 2021;147(5).

41.

International Maritime Organization. Guidance on Fatigue Mitigation and 1059 Management. Guidance on Fatigue Mitigation and 1059 Management [Internet]. 2001;1–105. Available from: https://wwwcdn.imo.org/localresources/en/OurWork/HumanElement/Documents/1014.pdf

42.

Gawron

, French

, Funke

. An Overview of Fatigue. In: Hancock PA, Desmond PA, editors. Stress, workload, and fatigue. Mahwah, NJ: Lawrence Erlbaum Associates Publishers; 2001. p. 581–95.

43.

Sarsangi

, Salehiniya

, Hannani

, Marzaleh

, Abadi

, Honarjoo

, et al. Assessment of workload effect on nursing occupational accidents in hospitals of Kashan, Iran. Biomed Res Ther. 2017;4(8):1527–40.

44.

Realyvásquez-Vargas

, Z-Flores

, Morales

L-C

, García-Alcaraz

. Mental Workload Assessment and Its Effects on Middle and Senior Managers in Manufacturing Companies. In: Realyvásquez-Vargas A, Arredondo-Soto KC, Hernández-Escobedo G, González-Reséndiz J, editors. Evaluating Mental Workload for Improved Workplace Performance [Internet]. 1st ed. IGI Global; 2020 [cited 2020 Jan 16]. p. 109–37. Available from: http://services.igi-global.com/resolvedoi/resolve.aspx?doi=10.4018/978-1-7998-1052-0.ch006

45.

Mehta

, Parasuraman

. Effects of Mental Fatigue on the Development of Physical Fatigue: A Neuroergonomic Approach. Hum Factors J Hum Factors Ergon Soc. 2014;56(4):645–56.

46.

Overgaard

, Gamborg

, Gyntelberg

, Heitmann

. Psychological workload is associated with weight gain between and Analyses based on the Danish Nurse Cohort Study. Int J Obes. 2004;28(8):1072–81.

47.

Booth

, Wang

, Turner

, Nowson

, Torres

. Diet-Induced Weight Loss Has No Effect on Psychological Stress in Overweight and Obese Adults: A Meta-Analysis of Randomized Controlled Trials. Nutrients [Internet]. 2018 May 14 [cited 2019 Nov 8];10(5):1–12. Available from: http://www.mdpi.com/2072-6643/10/5/613

48.

Roll

, Siu

, Li

SYW

, Witte

H De

. Human Error: The Impact of Job Insecurity on Attention-Related Cognitive Errors and Error Detection. Int J Environ Res Public Heal [Internet]. 2019 Jul 8 [cited 2021 Aug 12];16(13):2427. Available from: https://www.mdpi.com/1660-4601/16/13/2427/htm

49.

Qeshmy

, Makdisi

, Da Silva

EHDR

, Angelisa

. Managing human errors: Augmented reality systems as a tool in the quality journey In: Procedia Manufacturing. Elsevier B.V.; 2019. p. 24–30.

50.

Dehais

, Hodgetts

, Causse

, Behrend

, Durantin

, Tremblay

. Momentary lapse of control: A cognitive continuum approach to understanding and mitigating perseveration in human error. Vol. 100, Neuroscience and Biobehavioral Reviews. Elsevier Ltd; 2019. p. 252–62.

51.

Memarian

, Mitropoulos

. Production practices affecting worker task demands in concrete operations: A case study. Work. 2016;53(3):535–50.

52.

Yan

, Tran

, Wei

, Habiyaremye

. Driver’s mental workload prediction model based on physiological indices. Int J Occup SafErgon [Internet]. 2019 Jul 3 [cited 2020 Jan 9];25(3):476–84. Available from: https://www-tandfonline-com-s.web.bisu.edu.cn/doi/full/10.1080/10803548.2017.1368951

53.

, Wu

. Effect of Physical Fatigue on Cognitive Ability of Workers in Furniture CNC Operation. In: Long S, Dhillon BS, editors. Man-Machine- Environment System Engineering Proceedings of the 18th International Conference on MMESE. Singapore: Springer; 2019. p. 49–55.

54.

Jamshidi

. Stochastic human fatigue modeling in production systems. J Ind Syst Eng. 2019;12(1):270–83.

55.

Frone

, Blais

A-R

. Work Fatigue in a Non-Deployed Military Setting: Assessment, Prevalence, Predictors, and Outcomes. Int J Environ Res Public Health [Internet]. 2019 Aug 13 [cited 2019 Nov 7];16(16):1–26. Available from: https://www.mdpi.com/1660-4601/16/16/2892

56.

Myszewski

. Mathematical model of the occurrence of human error in manufacturing processes. Qual Reliab Eng Int. 2010;26(8):845–51.

57.

Michalos

, Makris

, Chryssolouris

. The effect of job rotation during assembly on the quality of final product. CIRP J Manuf Sci Technol. 2013;6(3):187–97.

58.

Cumber

, Greig

. Can a tool developed for industry be used to assess fatigue risk in medical rotas? A pilot study of foundation doctors’ rotas in a tertiary centre. BMJ Open. 2019;9(2):1–6.

59.

Sobhani

, Wahab

MIM

, Jaber

. The effect of working environment aspects on a vendor–buyer inventory model. Int J Prod Econ. 2019;208:171–83.

60.

Dantan

, El Mouayni

, Sadeghi

, Siadat

, Etienne

. Human factors integration in manufacturing systems design using function–behavior–structure framework and behaviour simulations. CIRP Ann. 2019;68(1):125–8.

61.

Yamada

, Kameda

, Noborisaka

, Suzuki

, Honda

, Yamada

. Excessive fatigue and weight gain among cleanroom workers after changing from an 8-hour to a 12-hour shift. Scand J Work Environ Health [Internet]. 2001;27(5):318–26. Available from: www.jstor.org/stable/40958857

62.

Maayan

, Hoogendoorn

, Sweat

, Convit

. Disinhibited Eating in Obese Adolescents Is Associated With Orbitofrontal Volume Reductions and Executive Dysfunction. Obesity [Internet]. 2011 Jul 1 [cited 2021 Sep 13];19(7):1382–7. Available from: https://onlinelibrary-wiley-com-443.web.bisu.edu.cn/doi/full/10.1038/oby.2011.15

63.

Perpiñá

, Segura

, Sánchez-Reales

. Cognitive flexibility and decision-making in eating disorders and obesity. Eat Weight Disord – Stud Anorexia, Bulim Obes [Internet]. 2017 Oct 15 [cited 2021 Sep 13];22(3):435–44. Available from: https://link-springer-com-443.web.bisu.edu.cn/article/10.1007/s40519-016-0331-3

64.

Heymsfield

, Wadden

. Mechanisms, Pathophysiology, and Management of Obesity. N Engl J Med [Internet]. 2017 Jan 18 [cited 2021 Sep 13];376(3):254–66. Available from: https://www.nejm.org/doi/full/10.1056/nejmra1514009

65.

Verdejo-García

, Pérez-Expósito

, Schmidt-Río-Valle

, Fernández-Serrano

, Cruz

, Pérez-García

, et al. Selective Alterations Within Executive Functions in Adolescents With Excess Weight. Obesity [Internet]. 2010 Aug 1 [cited 2021 Sep 13];18(8):1572–8. Available from: https://onlinelibrary-wiley-com-443.web.bisu.edu.cn/doi/full/10.1038/oby.2009.475

66.

Qavam

, Anisan

, Fathi

, Pourabbasi

. Study of relationship between obesity and executive functions among high school students in Bushehr, Iran. J Diabetes Metab Disord [Internet]. 2015 Oct 15 [cited 2021 Sep 13];14(1):1–5. Available from: https://link-springer-com-443.web.bisu.edu.cn/article/10.1186/s40200-015-0211-9

67.

Lasselin

, Magne

, Beau

, Aubert

, Dexpert

, Carrez

, et al. Low-grade inflammation is a major contributor of impaired attentional set shifting in obese subjects. Brain Behav Immun. 2016;58:63–8.

68.

Gonzales

, Tarumi

, Miles

, Tanaka

, Shah

, Haley

. Insulin Sensitivity as a Mediator of the Relationship Between BMI and Working Memory-Related Brain Activation. Obesity [Internet]. 2010 Nov 1 [cited 2021 Sep 13];18(11):2131–7. Available from: https://onlinelibrary-wiley-com-443.web.bisu.edu.cn/doi/full/10.1038/oby.2010.183

69.

Tsai

C-L

, Chen

F-C

, Pan

C-Y

, Tseng

Y-T

. The Neurocognitive Performance of Visuospatial Attention in Children with Obesity. Front Psychol. 2016;0(JUL):1033.

70.

Cohen

, Yates

, Duong

, Convit

. Obesity, orbitofrontal structure and function are associated with food choice: a cross-sectional study. BMJ Open [Internet]. 2011 Jan 1 [cited 2021 Sep 13];1(2):e000175. Available from: https://bmjopen.bmj.com/content/1/2/e000175

71.

Yang

, Shields

, Guoa

, Liua

. Executive function performance in obesity and overweight individuals: A meta-analysis and review. Neurosci Biobehav Rev. 2018;84:225–44.

72.

Sabia

. The Effect of Body Weight on Adolescent Academic Performance. South Econ J. 2007;73(4):871–900.

73.

Anderson

, Good

. Increased body weight affects academic performance in university students. Prev Med Reports. 2017;5:220–323.

74.

Stanton

, Salmon

, Walker

, Baber

, Jenkins

. Human Factors Methods. A Practical Guide for Engineering and Design. New York, New York, USA: Routledge; 2018. 571 p.

75.

Galy

, Paxion

, Berthelon

. Measuring mental workload with the NASA-TLX needs to examine each dimension rather than relying on the global score: an example with driving. Ergonomics [Internet]. 2018 Apr 3 [cited 2019 Feb 25];61(4):517–27. Available from: https://www-tandfonline-com-s.web.bisu.edu.cn/doi/full/10.1080/00140139.2017.1369583

76.

Ruiz-Rabelo

, Navarro-Rodriguez

, Di-Stasi

, Diaz-Jimenez

, Cabrera-Bermon

, Diaz-Iglesias

, et al. Validation of the NASA-TLX Score in Ongoing Assessment of Mental Workload During a Laparoscopic Learning Curve in Bariatric Surgery. Obes Surg [Internet]. 2015 Dec 12 [cited 2019 Feb 25];25(12):2451–6. Available from: http://link.springer.com/10.1007/s11695-015-1922-1

77.

Tubbs-Cooley

, Mara

, Carle

, Gurses

. The NASA Task Load Index as a measure of overall workload among neonatal, paediatric and adult intensive care nurses. Intensive Crit Care Nurs[Internet]. 2018 Jun 1 [cited 2019 Feb 25];46:64–9. Available from: https://www-sciencedirect-com-443.web.bisu.edu.cn/science/article/pii/S0964339717303403

78.

Rizzo

, Longo

. Representing and Inferring Mental Workload via Defeasible Reasoning: a Comparison with the NASA Task Load Index and the Workload Profile. In: 1st Workshop on Advances In Argumentation In Artificial Intelligence [Internet]. Bari, Italy; 2017 [cited 2019 Feb 25]. p. 126–40. Available from: https://arrow.dit.ie/scschcomcon/216

79.

Longo

. Formalising Human Mental Workload as Non-monotonic Concept for Adaptive and Personalised Web-Design. In: Masthoff J, Mobasher B, Desmarais M, Nkambou R, editors. User Modeling, Adaptation, and Personalization [Internet]. Springer, Berlin, Heidelberg; 2012 [cited 2019 Feb 25]. p. 369–73. Available from: http://link.springer.com/10.1007/978-3-642-31454-4_38

80.

Schwartz

, Gerberich

, Albin

, Kim

, Ryan

, Church

, et al. The association between janitor physical workload, mental workload, and stress: The SWEEP study. Work. 2020;65(4):837–46.

81.

Hart

. Nasa-Task Load Index (NASA-TLX); 20 Years Later. Proc Hum Factors Ergon Soc Annu Meet [Internet]. 2006 Oct 5 [cited 2019 Feb 25];50(9):904–8. Available from: http://journals.sagepub.com/doi/10.1177/154193120605000909

82.

Hart

, Staveland

. Development of NASA-TLX (Task Load Index): Results of Empirical and Theoretical Research. Adv Psychol [Internet]. 1988 Jan 1 [cited 2019 Feb 25];52:139–83. Available from: https://www-sciencedirect-com-443.web.bisu.edu.cn/science/article/pii/S0166411508623869

83.

Hwang

S-L

, Yau

Y-J

, Lin

Y-T

, Chen

J-H

, Huang

T-H

, Yenn

T-C

, et al. Predicting work performance in nuclear power plants. Saf Sci[Internet]. 2008 Aug 1 [cited 2019 Feb 25];46(7):1115–24. Available from: https://www-sciencedirect-com-443.web.bisu.edu.cn/science/article/pii/S0925753507000872?via%3Dihub

84.

Jacobson

, Bolon

, Elder

, Schroer

, Matthews

, Szaflarski

, et al. Temporal and subjective work demands in office-based patient care: an exploration of the dimensions of physician work intensity. Med Care [Internet]. 2011 Jan 1 [cited 2019 Feb 25];49(1):52–8. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21164325

85.

Mazur

, Mosaly

, Jackson

, Chang

, Burkhardt

, Adams

, et al. Quantitative Assessment of Workload and Stressors in Clinical Radiation Oncology. Int J Radiat Oncol [Internet]. 2012 Aug 1 [cited 2019 Feb 25];83(5):e571–6. Available from: https://www-sciencedirect-com-443.web.bisu.edu.cn/science/article/pii/S0360301612001368?via%3Dihub

86.

Sharifi

, Danesh

, Gholamnia

. Improvements in musculoskeletal symptoms, mental workload and mental fatigue: Effects of a multicomponent ergonomic intervention among call center workers. Work. 2022;72(2):765–74.

87.

Abareshi

, Salimi

, Farnia

, Fallahi

, Rastaghi

. The impact of mental workload, work-related and socio-demographic factors on job burnout among emergency medical staff. Work. 2022;72(4):1269–77.

88.

da Silva

, da Silva

, Silva

ANA

. Relationship between mental workload and hospital infection in the ICU. Work. 2022;Preprint(Preprint):1–11.

89.

Ebrahimi

, Jafarjalal

, Lotfolahzadeh

, Kharghani Moghadam

. The effect of workload on nurses’ quality of life with moderating perceived social support during the COVID-19 pandemic. Work. 2021;70(2):347–54.

90.

Han

, Zhang

, Xiang

, Shang

, Anjani

, Song

, et al. Desktop lighting for comfortable use of a computer screen. Work. 2021;68(s1):S209–21.

91.

Moertl

, Hoefler

, Ewerz

. Incorporating driver activity in holistic comfort models. Work. 2021;68(s1):S119–28.

92.

Frazier

, Pitts

, McComb

. Measuring cognitive workload in automated knowledge work environments: a systematic literature review. Cogn Technol Work [Internet]. 2022 Aug 17 [cited 2022 Oct 20];24(4):557–87. Available from https://link-springer-com-443.web.bisu.edu.cn/article/10.1007/s10111-022-00708-0

93.

Chen

, Visser

TAW

, Huf

, Loft

. Optimizing the balance between task automation and human manual control in simulated submarine track management. J Exp Psychol Appl [Internet]. 2022 Sep 1 [cited 2022 Oct 20];23(3):240–62. Available from: /doiLanding?doi=10.1037%2Fxap0000126

94.

Jou

, Yenn

, Lin

, Yang

, Chiang

. Evaluation of operators’ mental workload of human–system interface automation in the advanced nuclear power plants. Nucl Eng Des. 2009;239(11):2537–42.

95.

Kaber

, Endsley

. The effects of level of automation and adaptive automation on human performance, situation awareness and workload in a dynamic control task. Theor Issues Ergon Sci[Internet]. 2007 [cited 2022 Oct 20];5(2):113–53. Available from: https://www-tandfonline-com-s.web.bisu.edu.cn/doi/abs/10.1080/1463922021000054335

96.

, Wickens

, Sarter

, Sebok

. Stages and Levels of Automationin Support of Space Teleoperations. Hum Factors J Hum Factors Ergon Soc [Internet]. 2014 Feb 18 [cited 2022 Oct 20];56(6):1050–61. Available from: https://journals.sagepub.com/doi/10.1177/0018720814522830

97.

Manzey

, Reichenbach

, Onnasch

. Human Performance Consequences of Automated Decision Aids: The Impact of Degree of Automation and System Experience. J Cogn Eng Decis Mak [Internet]. 2012 Jan 13 [cited 2022 Oct 20];6(1):57–87. Available from: https://journals.sagepub.com/doi/10.1177/1555343411433844

98.

Wilson

, Russell

. Performance Enhancement in an Uninhabited Air Vehicle Task Using Psychophysiologically Determined Adaptive Aiding. Hum Factors J Hum Factors Ergon Soc [Internet]. 2016 Jul 2 [cited 2022 Oct 20];49(6):1005–18. Available from: https://journals.sagepub.com/doi/10.1518/001872007X249875

99.

Evans

, Fendley

. A multi-measure approach for connecting cognitive workload and automation. Int J Hum Comput Stud. 2017;97:182–9.

100.

Kaber

, Onal

, Endsley

. Design of automation for telerobots and the effect on performance, operator situation awareness, and subjective workload. Hum Factors Ergon Manuf Serv Ind. 2000;10(4):409–30.

101.

Kaber

, Wright

, Sheik-Nainar

. Investigation of multi-modal interface features for adaptive automation of a human–robot system. Int J Hum Comput Stud. 2006;64(6):527–40.

102.

Stapel

, Mullakkal-Babu

, Happee

. Automated driving reduces perceived workload, but monitoring causes higher cognitive load than manual driving. Transp Res Part F Traffic Psychol Behav. 2019;60:590–605.

103.

Pang

, Guo

, Zhang

, Wanyan

, Qu

, Wang

. Subject-specific mental workload classification using EEG and stochastic configuration network (SCN). Biomed Signal Process Control. 2021;68:102711.

104.

Lichtenstein

, Malkenes

, Sibbersen

, Hinze

. Work addiction is associated with increased stress and reduced quality of life: Validation of the Bergen Work Addiction Scale in Danish. Scand J Psychol [Internet]. 2019 Apr [cited 2020 Jan 9];60(2):145–51. Available from: http://doi.wiley.com/10.1111/sjo12506

105.

Bakker

, Hartman

YAW

, Hopman

MTE

, Hopkins

, Graves

LEF

, Dunstan

, et al. Validity and reliability of subjective methods to assess sedentary behaviour in adults: a systematic review and meta-analysis. Int J Behav Nutr Phys Act. 2020;17(75):1–31.

106.

Coakes

, Steed

. SPSS: Analysis Without Anguish Using Spss Version 14.0 for Windows [Internet]. New York: John Wiley & Sons, Inc.; 2009 [cited 2016 Jan 6]. Available from: http://dl.acm.org/citation.cfm?id=1804538

107.

García-Alcaraz

, Maldonado-Macías

, Alvarado

, Rivera

. Human critical success factors for kaizen and its impacts in industrial performance. Int J Adv Manuf Technol [Internet]. 2014 Nov 15 [cited 2015 Oct 29];70(9-12):2187–98. Available from: http://link.springer.com/10.1007/s00170-013-5445-4

108.

Hair

, Black

, Babin

, Anderson

, Tatham

. Multivariate data analysis. 6th Editio. Upper Saddle River, NJ: Pearson Prentice Hall; 2006.

109.

Avelar-Sosa

, García-Alcaraz

, Cedillo-Campos

, Adarme-Jaimes

. Effects of regional infrastructure and offeredservices in the supply chains performance: Case Ciudad Juarez. DYNA [Internet]. 2014 Aug 26 [cited 2016 Feb 5];81(186):208–17. Available from: http://www.scielo.org.co/scielo.php?script=sci_arttext&pid=S0012-73532014000400028&lng=en&nrm=iso&tlng=pt

110.

Cronbach

. Coefficient alpha and the internal structure of tests. Psychometrika [Internet]. 1951 Sep [cited 2014 Oct 27];16(3):297–334. Available from: http://link.springer.com/10.1007/BF02310555

111.

Tavakol

, Dennick

. Making sense of Cronbach’s alpha. Int J Med Educ [Internet]. 2011 Jun 27 [cited 2022 Oct 21];2:55. Available from: /pmc/articles/PMC4205511/

112.

Hayes

. Introduction to Mediation, Moderation and Conditional Process Analysis. A Regression-Based Approach.2nd ed. New York: The Guilford Press; 2018. 691 p.

113.

Hayes

. PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modeling [White paper]. 2012;1–39. Available from: Www.afhayes.com/%0APublic/process2012.pdf

114.

Lee

, Shin

H-S

, Park

, Kim

, Lee

, et al. Depression as a Mediator of Chronic Fatigue and Post-Traumatic Stress Symptoms in Middle East Respiratory Syndrome Survivors. Psychiatry Investig. 2019;16(1):59–64.

115.

Hayes

. Permutation test is not distribution-free: Testing H0: ρ = 0. Psychol Methods. 1996;1(2):184–98.

116.

Ato-García

, Vallejo-Seco

. Diseños de investigaciónen psicología. 1st ed. Madrid: Ediciones Pirámide; 2016. 568 p.

117.

Esquivel-Molina

, Velasco-Rodríguez

, Martínez-Rubinde Celis

, Barbachano-Rodríguez

, González-Ávila

, Castillo-Ruiz

. Coeficiente de correlación intraclase vs correlación de Pearson de la glucemia capilar por reflectometría y glucemia plasmática. Med Int Mex [Internet]. 2006;22(3): 165–71. Available from: https://www.cmim.org/Revista/2006/200603_may_jun.pdf#page=7

118.

Hayes

, Preacher

. Quantifying and Testing Indirect Effects in Simple Mediation Models When the Constituent Paths Are Nonlinear. Multivariate Behav Res. 2010;45(4):627660.

119.

Lee

, Park

, Cho

. Maladaptive Perfectionism and College Adjustment of International Students in Korea: A Moderated Mediation Model of Social Support. Sustainability. 2020;12(11):4729.

120.

MacKinnon

, Lockwood

, Williams

. Confidence limits for the indirect effect: Distribution of the product and resampling methods. Multivariate Behav Res [Internet]. 2004 [cited 2020 Jul 5];39(1):99–128. Available from: https://www-tandfonline-com-s.web.bisu.edu.cn/doi/abs/10.1207/s15327906mbr3901_4

121.

Villotti

, Corbiere

, Dewa

, Fraccaroli

, Sultan-Taïeb

H el ene

, Zaniboni

, et al. A serial mediation model of workplace social support on work productivity: the role of self-stigma and job tenure self-efficacy in people with severe mental disorders. Disabil Rehabil. 2018;40(26):3113–9.

122.

Nunnally

, Bernstein

. Psychometric theory. 3rd ed. New York: McGraw-Hill Higher, INC; 1994. 752 p.

123.

Streiner

. Starting at the Beginning: An Introduction to Coefficient Alpha and Internal Consistency. J Pers Assess [Internet]. 2003 [cited 2022 Oct 21];80(1):99–103. Available from: https://www-tandfonline-com-s.web.bisu.edu.cn/doi/abs/10.1207/S15327752JPA8001_18

124.

Kolus

, Wells

, Neumann

. Production quality and human factors engineering: A systematic review and theoretical framework. Appl Ergon. 2018;73:55–89.

125.

, Li

, Geng

, Wang

, Zhang

. Subtypes of nurses’ mental workload and interaction patterns with fatigue and work engagement during coronavirus disease 2019 (COVID-19) outbreak: A latent class analysis. BMC Nurs [Internet]. 2021 Dec 1 [cited 2022 Oct 22];20(1):1–9. Available from: https://link-springer-com-443.web.bisu.edu.cn/articles/10.1186/s12912-021-00726-9

126.

Yeow

, Cheah

, Tan

, Chow

, Tai

. Exploring the Root Causes of Human Error in Small, Medium Enterprise (SMEs) Manufacturing Industries in Malaysia. In: International Symposium on Applied Structural Equation Modeling and Methodological Matters Sasem [Internet]. Melaka, Malaysia; 2019 [cited 2022 Oct 22]. p. 23–4. Available from: http://shdl.mmu.edu.my/9376/

127.

Lim

E-J

, Son

C-G

. Prevalence of Chronic Fatigue Syndrome (CFS) in Korea and Japan: A Meta-Analysis. J Clin Med. 2021, Vol 10, Page 3204 [Internet]. 2021 Jul 21 [cited 2021 Sep 18];10(15):3204. Available from: https://www.mdpi.com/2077-0383/10/15/3204/htm

128.

Josev

, Cole

, Scheinberg

, Rowe

, Lubitz

, Knight

. Health, Wellbeing, and Prognosis of Australian Adolescents with Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS): A Case-Controlled Follow-Up Study. J Clin Med 2021, Vol 10, Page 3603 [Internet]. 2021 Aug 16 [cited 2021 Sep 18];10(16):3603. Available from: https://www.mdpi.com/2077-0383/10/16/3603/htm

129.

Dopolani

, Arefi

, Akhlaghi

, Ghalichi

, Salehi

, Khajehnasiri

, et al. Investigation of occupational fatigue and safety climate among nurses using the structural equation model. Work. 2022;Pre-press(Pre-press):1–11.

130.

Rochester

, Jones

, Hetherington

, Nieuwboer

, Willems

A-M

, Kwakkel

, et al. Gait and gait-related activities and fatigue in Parkinson’s disease: What is the relationship? Disabil Rehabil (2006;28(22):1365–71.

	Ordinal scale
	Very Low (1)	Low (2)	Medium (3)	High (4)	Very high (5)
Mental workload level	1–20	21–40	41–60	61–80	81–100

Effects of mental workload on manufacturing systems employees: A mediation causal model

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1 Introduction

1.1 Association of mental workload with physical fatigue, body weight and human error

1.2 Association of physical fatigue with human error and body weight gain

1.3 Association of body weight gain with human error

2.1 Stage 1. Mental workload questionnaire (MWLQ) development

Table 1 Ordinal scale for mental workload levels Ordinal scale Very Low (1) Low (2) Medium (3) High (4) Very high (5) Mental workload level 1–20 21–40 41–60 61–80 81–100

2.3.1 MWLQ statistical debugging and validation

2.3.2 Hypotheses testing

3 Results and discussion

3.1 Sample description

Table 5 Correlation matrix Mental Physical Body weight Human workload fatigue gain error Mental workload 1.000 0.179 0.330 –0.104 Physical fatigue 0.179 1.000 0.213 –0.057 Body weight gain 0.330 0.213 1.000 –0.016 Human error –0.104 –0.057 –0.016 1.000

4 Conclusions and recommendations

4.1 Conclusions related to the hypotheses

4.2 Industrial recommendations

4.3 Future research

Conflict of interest

Ethical approval

Informed consent

Footnotes

Acknowledgments

Funding

References

Table 1
Ordinal scale for mental workload levels

Ordinal scale

Very Low (1) Low (2) Medium (3) High (4) Very high (5)

Mental workload level 1–20 21–40 41–60 61–80 81–100

Table 5
Correlation matrix

Mental Physical Body weight Human

workload fatigue gain error

Mental workload 1.000 0.179 0.330 –0.104

Physical fatigue 0.179 1.000 0.213 –0.057

Body weight gain 0.330 0.213 1.000 –0.016

Human error –0.104 –0.057 –0.016 1.000