Using Digital and Physical Simulation to Focus on Human Factors and Ergonomics in Aviation Maintainability

Abstract

Objective:

This research aimed to evaluate the differences in the assessments made by three simulation tools used in a maintainability design office to perform human factor/ergonomics (HFE) analysis: digital human modeling (DHM), virtual reality (VR), and physical mock-up (PMU).

Background:

Maintainability engineers use digital/physical simulation tools in the early design phase to analyze whether the design is well adapted for maintenance operators. Knowing the potential of these simulation tools would encourage maintainability stakeholders to integrate HFE in the design process more efficiently.

Method:

Eleven maintenance tasks were analyzed from the participation of six maintenance operators. Various HFE indicators including physical, cognitive, and organizational indicators were assessed. Each operator repeated 11 maintenance tasks on VR and PMU. Based on the anthropometric parameters, six manikins were created to analyze 11 maintenance tasks on DHM.

Results:

A significant difference was found for the organizational indicators between VR and PMU, whereas the physical and cognitive indicators are similar. DHM, VR, and PMU are compared with the common HFE indicators for the physical dimension and present a significant difference for individual tasks.

Conclusion:

To reduce the gap between simulation tools, a better physical representation is requested on the VR platform, improving the perception of work sequences in the virtual world. Concerning DHM, a new paradigm is proposed to study a few tasks per work area instead of studying each task independently.

Application:

This study will help develop a new methodology and tools specifically for non-HFE experts in the maintainability department.

Keywords

human factors ergonomics simulation tools virtual reality physical mock-up maintainability aeronautics

Introduction

Human Factors in Maintenance

Human performance has been studied from the very beginning in the aviation field, especially regarding aircraft accidents (Maurino, Reason, Johnston, & Lee, 2017; Wiener & Nagel, 1988). Initially, the reliability of machines was the primary concern, and many attempts were made to improve the technology of the devices (Maurino et al., 2017; Wiener & Nagel, 1988). During the following decades, human factors/ergonomics (HFE) raised as a discipline to design the flight deck and to investigate the interaction between human (pilot) and machine, particularly, during the training phase of the pilot in a simulator (Horeman, Akhtar, & Tuijthof, 2015; Wiener & Nagel, 1988). Safety and comfort in the cockpit and the passenger cabin were also improved by considering HFE principles during the subsequent decades (Spenser, 2008). However, 12% of all aviation accidents were due to HFE issues during maintenance activity (Hobbs, 2000). Integrating HFE in maintainability may increase the quality of maintenance activities and reduce the rate of mistakes/errors (Gruber, De Leon, George, & Thompson, 2015). The design engineers in the maintainability department interact and collaborate with other engineering departments (e.g., aerodynamic, hydraulic and electric integration, and architecture), and raising HFE culture between them could effectively affect the future maintenance activity.

However, aviation accidents are not the only problem that demonstrates the need to improve HFE for maintenance activities. The health and safety of maintenance operators is also a key contributor to maintenance errors (Hobbs, 2000). Various studies have already highlighted the fact that maintenance activities can cause health problems (musculoskeletal disorders, stress, and high mental workload) and workplace accidents (French Association of Maintenance Engineers (AFIM; 2004; European Agency for Safety and Health at Work, 2010). In a survey of 2,500 maintenance operators from various industries (automotive, train, and aeronautics), AFIM showed that 62% of respondents considered their occupation to be dangerous. Another study performed in Europe showed that 15%–20% of accidents at work occurred in the field of maintenance, suggesting that maintenance tasks are the most dangerous activities in an industry (European Agency for Safety and Health at Work, 2010).

Simulation Tools for Assessing Human Factors

The design engineers simulate the maintenance activities in the digital mock-up (DMU) and physical mock-up (PMU) to integrate HFE in the design process of maintainability in the early design phase (Stoffregen, Bardy, Smart, & Pagulayan, 2003). They proactively analyze the future maintenance activity, either in a DMU or in a PMU, by different ergonomic tools. The DMU can be categorized into two simulation tools: digital human modeling (DHM) and virtual reality (VR). DHM allows the integration of ergonomics in the preliminary phases of the design process of the product, manufacturing, and maintenance (Ahmed, Gawand, Irshad, & Demirel, 2018; Krüger & Nguyen, 2015; Perez, de Looze, Bosch, & Neumann, 2014). Various types of DHM tools have been developed. Rasmussen, Damsgaard, Christensen, and Surma (2002) presented the AnyBody digital manikin, which aims to couple the specific geometric criteria of a future product with the ergonomic requirements of a multidisciplinary optimization approach. Jack manikins (EAI-UGS) have been used in various industries to evaluate HFE indicators (disassembly, accessibility, usability) on maintenance tasks (Di Gironimo, Monacelli, & Patalano, 2004). The Human Builder is the most frequently used manikin tool in the aviation field and presents various useful characteristics (visual fields, posture analysis, anthropometric measurements, and biomechanical single action analysis) for maintainability studies (Hashim, Kamat, Halim, & Othman, 2014; Nishanth, Muthukumar, & Arivanantham, 2015). Geng, Zhou, Lv, and Wang (2013) propose a methodology to improve the use of ergonomics options for the human builder to anticipate the risks of maintenance activities on the auxiliary power unit of a plane. According to these authors, the visual field, reach envelope, and the rapid upper limb assessment (RULA) score are the existing options for studying the manikin’s interactions with its environment in maintainability. Another study attempts to assess the occupational risks of the semiautomatic pump manufacturing industry using the human builder (Nishanth et al., 2015). However, DHM cannot anticipate the full activity of the real operators, and it is difficult to make decisions and validate the design based solely on DHM. Quintero-Duran and Paul (2016) explain the limitations of using DHM during the design development stage. These authors highlight the fact that the influence of design on the human’s behavior cannot be simulated and anticipated with DHM studies. It is also not possible to observe psychosocial and environmental stressors, which presents a limitation in the study of human factors using DHM. Johnson and Fletcher (2015) have detailed the subjectivity of human factor analysis with DHM because the manikins’ degrees of freedom are not always representative.

For this reason, VR is used—as a typical DMU tool—to study the operator’s accessibility and the interaction with the architecture, particularly in maintainability (Amundarain, Borro, Matey, Alonso, & de Guipúzcoa, 2003). VR is easy to edit and cost-effective, and the aviation industry uses it to evaluate the ergonomic features of a task, whereas PMU is less predominant when simulating maintenance tasks (Shi & Liu, 2012). The virtual setting is not sufficient to consider all aspects of HFE in the design process, and PMU is utilized to design space forms and architectural aesthetic. However, this design approach is technocentric design (more focused on technology) whereas the anthropocentered design (interactions between operators and the architecture) might be more suitable, which will consider organization, mental processes (e.g., perception, reasoning), and motor responses (Bittencourt, Duarte, & Béguin, 2017; Czerniak, Brandl, & Mertens, 2017; Das & Sengupta, 1996; Meister, 2014; Sagot, 1999).

The background/experience of the engineers allowed them to focus almost entirely on the physical dimension of HFE (Bernard, Bazzaro, Paquin, & Sagot, 2017; Bernard, Zare, Sagot, & Paquin, 2019; Broberg, 2007). The maintainability engineers may underutilize the potential of DMU and PMU for performing ergonomic analyses on simulated maintenance tasks due to their lack of HFE skills and knowledge (Sharma, Singhal, & Sonia, 2018). Few studies have investigated the contribution made by DMU and PMU to the design process, and more specifically, the complementary aspect of VR and PMU. Although various DHM tools have already been compared in previous studies (Duffy, 2016; Lämkull, Hanson, & Örtengren, 2009) there is very little literature concerning the contribution of DHM, VR, and PMU during the design process (Pontonnier, Dumont, Samani, Madeleine, & Badawi, 2014). Recent studies have demonstrated the possibility of applying VR and augmented reality for ergonomic studies in the design process, but they concluded the significant difference between real and virtual settings results (Aromaa & Väänänen, 2016; Pontonnier et al., 2014).

The challenge in the maintainability department of the aeronautical industry is to integrate better HFE, using simulation tools (DHM, DMU, and PMU) during the design phases. We hypothesize that engineers use the simulation tools more efficiently for assessing ergonomic factors when they know the differences and complementary aspects of these tools. The first step in this proposed improvement is to study the existing disparity between the simulation tools. This study was, therefore, designed to investigate the difference in the assessments of HFE indicators made by DHM, VR, and PMU.

The Context of the Study

Maintainability is defined as the ability of a product to be maintained/repaired easily during maintenance operations that include the functions for which it was designed (EN 13306:2010, 2010). Maintenance tasks are expensive for the aeronautical industry, reaching between 12% and 15% of an airline’s total costs per year (Čokorilo, 2011). Furthermore, the duration of each maintenance task (referred to as the mean time to repair, MTTR) must be respected (Chang & Wang, 2010). The ratio of “hours of maintenance/flight hours” is a priority in the helicopter industry. Maintainability stakeholders must anticipate the future maintenance operator’s activities in three main ways: the item’s architecture, the operator’s tools, and the operator’s ways of working (Lee, Ma, Thimm, & Verstraeten, 2008). However, it can be challenging to predict the human activity involved in maintenance for maintainability engineering, mainly by taking customer feedback into account.

More than half of maintenance professionals from various industrial and service sectors interviewed in a survey believed that their occupation is hard (French Association of Maintenance Engineers, AFIM). Health and safety are, therefore, the significant maintenance issues that should be studied further.

To study maintenance activity, we focus our analysis on the upper deck of a helicopter, where the engine is located as illustrated in Figure 1. The maintenance tasks performed on the engine were complicated due to the reduced accessibility and the limited available workspace.

Figure 1.

Helicopter representation with the upper deck as the work area.

Method

Research Setting

The PMU used for our experiment represents the upper deck. The scaffold is created to provide access to the right/left engine for the operators. Figure 2 shows the top view of the upper deck. The maintenance operation is carried out from the corridor of the maintenance scaffolding. The maintenance tasks are carried out on both the left and right sides, depending on the arrangement of the technical elements (asymmetrical tasks). Some tasks must/can be performed through hatches requiring standing, kneeling, or lying postures.

Figure 2.

Top view of the upper deck layout with ladder and maintenance corridor on the right side (the same is installed on the left side).

Maintenance Tasks Selected

Eleven maintenance tasks were evaluated on the upper deck (Table 1). We selected these tasks because of their recurrence and representativeness among other maintenance tasks of the helicopter and plane, particularly, regarding the parameters of posture, gestures, and effort. Various studies in the aviation field have confirmed the interest of the engine part in performing HFE analysis (Amundarain et al., 2003; Dmitriev, Burlakov, Popov, & Popov, 2015; Geng et al., 2013), particularly because of the variability of tasks and work postures (Lockett & Arvanitopoulos-Darginis, 2017; Lu, Zhou, & Li, 2015; Shi & Liu, 2012). In addition, the tasks selected in our study are included as part of regular and mandatory maintenance activities—the most frequent task (Task 1), which is performed every week, and the least frequent task (Task 7), which occurs once a month. Furthermore, all the maintenance tasks have a theoretical duration estimated using the mean time measurement (MTM) method (Laring, Forsman, Kadefors, & Örtengren, 2002); this theoretical duration was taken as a reference in our experiment to compare with the real time needed to perform the tasks on VR and PMU.

Table 1

Description of Each Task and the Main Postures Associated

Task	Description	Main Posture
1	Manipulation of the ladder and upper deck to anchor them	Standing
2	Disassembly/assembly of the main generator	Standing and bending
3	Manipulation of the electric plug behind the firewall	Kneeling
4	Manipulation of engine electric plug (floor)	Kneeling/lying down
5	Disassembly/assembly of the hydraulic pump	Kneeling/standing
6	Verification of the drain collector	Kneeling
7	Disassembly/assembly of the auxiliary generator	Standing and bending
8	Disassembly/assembly of engine support	Lying down
9	Tightness check of the RPM	Kneeling
10	Tightness check of the suspension support	Kneeling
11	Tightness check of the primary rotor actuator	Kneeling

Specification of Simulation Tools

As the study was conducted in the aerospace industry, we used the “Human Builder” simulation tool, which is already employed by many maintainability engineers in various industries (Geng et al., 2013; Nishanth et al., 2015). The maintainability design engineer performs design analysis with a computer-aided design (CAD) software system (CATIA). This CAD integrates a DHM that is easily accessible for all design office stakeholders and that gives the option of carrying out ergonomic analyses and validating the design environment. The technical inputs, such as the weight of the manipulated parts and the duration of the tasks, came from other departments of the design office. The maintainability stakeholders performed the simulation with various statures (Figure 3).

Figure 3.

Interaction of the human builder manikin with the environment for the same task: on the left, 5th percentile female and on the right 95th percentile male.

The VR technology used by the helicopter manufacturer was realistic human ergonomic analysis (RHEA), and we used this technology for the experiment. The VR is based on “Virtools” software and is included in the design process, allowing a fast transfer between the CAD and VR platform. Real operators performed 11 maintenance tasks simulated in the VR platform. The physical interactions, such as contacts with the architecture of the upper deck, assembling/disassembling elements that represent real volume, the real weight of the components, and the sensors on the components to carry them virtually, were created on VR (Boy, 2018) to synchronize the virtual and real world (Figure 4). We could, therefore, keep the simulation setting as close as possible to the usual process and the current industrial context. The VR platform is composed of a VR headset (Oculus Rift) worn by the operator performing the maintenance tasks. The VR headset view was displayed on the screen, which was shared with the virtual work setting, including the operator’s avatar (reproducing the behavior of the real operator). This output ensures a full observation by maintainability stakeholders in the room to correlate the activity of the real operator and the evolution of his or her avatar in the virtual world.

Figure 4.

Physical part integrated during the virtual reality simulation.

Each operator repeated the same 11 maintenance tasks on the PMU under realistic conditions—reproducing the real weight and center of gravity. The PMU was composed of wooden parts and the real components of the upper deck. The realistic constraints were reproduced between the mock-up and the maintenance operators: the removable parts (with their real weight and center of gravity), the torque force, and small removable components (screws, nuts) were installed.

Participants and Organization of the Experiment

Six male operators were selected according to the NF EN ISO 15537 (2004) standard “Principles for selecting and using test persons for testing anthropometric aspects of industrial products and designs.” We could not include more operators for our experiment due to the constraints related to industrial maintenance planning and the availability of the simulation tools (VR and PMU). Two main characteristics were considered when selecting the subjects: anthropometric parameters (Table 2) and work experience (knowledge and expertise concerning the maintenance of helicopter engine systems). In addition, informed consent was obtained from each participant and the subjects were informed about the objectives and the methodology of the study. During the VR simulation, one operator (out of six subjects) was only able to complete four of the 11 tasks because he experienced motion sickness.

Table 2

The Operator’s Characteristics and Anthropometric Measurements Applied to Create the Manikins in the Digital Human Modeling

	Age	Experience (years)	Weight (Kg)	BMI (Kg m²)	Stature (cm)	Trochanteric Height (cm)	Shoulder Width (cm)	Pelvis Width (cm)	Hand Length (cm)	Hand Width (cm)
Operator 1	42	11	69	25.2	165.5	84	40	31.5	17.2	6.6
Operator 2	31	13	70	24.2	170	80	41.2	33.5	18.3	7.4
Operator 3	52	22	87	27.8	177	90	43	36.5	21.6	7.8
Operator 4	45	24	76	23.9	178.5	95	43	33	19.8	7.8
Operator 5	46	11	82	24.5	183	97	44.3	35.9	20.6	9.6
Operator 6	51	30	105	31.0	184	102	48.4	38.7	19.4	8.6
M (SD)	44.5 (7.6)	18.5 (8.0)	81.5 (13.4)	26.1 (2.8)	176.3 (7.3)	91.3 (8.3)	43.3 (2.9)	34.9 (2.7)	19.5 (1.6)	8.0 (1.0)

The experiment started on the PMU, with each of the six operators performing the 11 tasks individually over 2 hr. The same protocol was deployed within the VR platform 6 weeks later. In total, 12 hr of experimentation per simulation tool were analyzed based on direct observation and video recording. To perform the third experiment with DHM, a series of anthropometric parameters were measured on real operators to simulate the digital manikin in a way that reflected the operator. These measurements were carried out based on the NF EN ISO 7250 (2017) standard “Basic human body measurements for technological design.” The six main anthropometric parameters were stature, trochanteric height, shoulder width (bideltoid), pelvis width (seated subject), and the length and width of the hand (Table 2). We used the “posture editor” option in CATIA to set the manikin consistent with the anthropometric dimensions of each body segment and to adjust the angulation of each joint (Hashim et al., 2014; Neumann, 2006).

We can divide our analysis into two parts: comparing two simulation tools by considering six HFE indicators with the participation of the real operators (VR and PMU) and comparing three simulation tools (DHM, VR, and PMU) by integrating three HFE indicators. Due to the limitation of DHM, it was not possible to assess all aspects of activity without a real operator. The following section will discuss the common HFE indicators and investigation instruments that were used in this study in detail.

The Investigation Instrument

The operator’s activity was analyzed through two frequent maintenance activities that are often studied in maintainability: assembly and disassembly—defined as the capacity for an operator to disassemble and assemble a mechanical system (De Leon, Díaz, Martínez, & Marquez, 2012). We selected six HFE indicators to analyze these activities, which made it possible to cover the majority of HFE dimensions (physical, cognitive, and organizational dimensions) as defined by the International Ergonomics Association (IEA; 2000). The measurement tools used to quantify HFE indicators were as follows (Table 3):

The reach envelope: we measured the percentage of time it took the operator to reach the maintenance area on VR and PMU. For DHM, the “reach envelope” was analyzed and measured as a percentage of time by estimating the total duration of the task via the MTM method. The DHM proposed an assessment of reach envelope, simulating a virtual envelope around the manikin (Guan, Lei, & Li, 2012; Hashim et al., 2014; Neumann, 2006).

RULA assesses the upper limbs’ exposure to MSD risks. This method takes into account three main indicators: posture, force, and frequency (McAtamney & Corlett, 1993). This tool measured the postures of the subjects with a paper-based checklist on VR and PMU. The posture assessment option is available in CATIA software and allowed us to evaluate the postures of manikins on DHM (Guan et al., 2012; Vyavahare & Kallurkar, 2015).

The Borg scale (Borg, 1982) was used to measure perceived exertion during physical activity. Its recent version, CR10, ranging from 0 (no effort) to 10 (very hard), was used only for the experience on VR and PMU.

Safety work area: A checklist developed by the Association for Improvement of Working Conditions (called the APACT) is used to evaluate the safety criteria of DMU, VR, and PMU settings (Chitescu et al., 2003). This checklist possesses 22 parameters evaluating the HFE of the workplace in terms of ergonomic standards. Section nine of this checklist, which relates to “work area safety,” was used to assess the potential risks of indicators including working at height, falls, burns, or breaks, by combining the level of personal protection and the integrity of work protection, thus showing us the organizational dimension of the task. The score is ranked from 0 to 10 and considered damaging below 6.5.

NASA-TLX (Hart & Staveland, 1988) was used to evaluate mental workload. Based on six dimensions (mental demands, physical demands, time requirement, personal performance, frustration, and stress), it provides a final score rated from 0 to 100 that shows the degree of mental workload (low ≤ 30 or high ≥ 30). The operators answered the NASA-TLX questionnaire immediately after performing the simulated task, but only for the experience on VR and PMU.

Nordic questionnaire (Kuorinka et al., 1987): MSD symptoms were assessed by a body map questionnaire developed based on the Standardized Nordic questionnaire. This questionnaire was only used for the experiments on VR and PMU.

Table 3

Common HFE Indicators Between DHM, VR, and PMU Related to Maintenance Tasks, and the Measurement Tools Used to Evaluate These Indicators

Indicators	Measurement toolsVR and PMU	Measurement toolsDHM	Metrics
Posture	Reach envelope (video/photo)	Reach envelope (analysis option)	% (time)
Posture	RULA (Paper version)	RULA (analysis option)	Scale from 1 to 7
Musculoskeletal symptoms	Nordic questionnaire		Neck
			Shoulders
			Upper body
			Elbows
		/	Lumbar
			Wrists/hands
			Hips
			Knees
			Ankles/feet
Effort	Borg (perceived physical exertion)	/	Scale from 0 to 10
Workplace safety	APACT check list(section 9)	APACT check list(section 9)	Scale from 0 to 10
Time	Duration	/	Task duration
Mental workload	NASA-TLX	/	Scale from 0 to 100

Note. HFE = human factors/ergonomics; DHM = digital human modeling; VR = virtual reality; PMU = physical mock-up; RULA = rapid upper limb assessment; APACT = Association for Improvement of Working Conditions; NASA-TLX = National Aeronautics and Space Administration—Task Load Index.

Statistical Analysis

We used the nonparametric Friedman and Wilcoxon tests because of the small sample size and the dependent variables (same population in three settings). The differences in execution time for performing the 11 maintenance tasks were tested for three variables (the theoretical execution time, the execution time through VR platform, and the PMU setting) using the Friedman test. The common variables (RULA and workplace safety score) for three situations (DHM, VR, and PMU) were also analyzed using the Friedman statistical test. The nonparametric Wilcoxon test was used to compare the indicators for musculoskeletal symptoms, mental workload, and perceived physical exertion on the VR and PMU settings. Statistical analysis was performed in SPSS 24.0 (IBM), and p < .05 was considered significant.

Results

Each task performed on both simulation tools (VR and PMU) required an execution time higher than the theoretical time estimated by the MTM method, and this difference was statistically significant for nine tasks (Figure 5). Furthermore, variable execution times were found between VR and PMU. For example, the duration for executing Task 1 was 7.6 (±0.3) s on the PMU, whereas it was 164.7 (±12.3) s on the VR platform—much higher than PMU. However, it was opposite for Task 10 as the execution time was 393.2 (±68.1) s on VR and 168.4 (±11.3) s on PMU.

Figure 5.

The execution time of 11 aviation maintenance tasks on the physical mock-up (PMU) and virtual reality (VR). The theoretical execution time estimated by the mean time measurement (MTM) method provided for each task (Friedman *p ≤ .05;**p ≤ .01; ***p ≤ .001).

We compared the various HFE indicators tested on the simulation tools (VR and PMU) with the participation of real operators. The RULA score and the safety at work score were not significantly different between VR and PMU. The musculoskeletal symptoms in the lower back and knee represented the meaningful percentages felt by the participants. The lower back strain was significantly different between VR and PMU only for Task 8 (Table 4). The percentages of musculoskeletal symptoms were negligible for the other body segments on both simulation tools. Perceived physical exertion (Borg scale) and mental workload (NASA-TLX) were compared in individual tasks with both VR and PMU. The Wilcoxon test showed no significant difference between the results of the Borg scale and NASA-TLX on VR and PMU. However, the NASA-TLX score presented a few differences, and mental workload was higher for VR than for PMU in each task. Tasks 2 and 10 received the highest scores—regardless of simulation tool—due to the excessive time and difficult accessibility associated with a static effort to retain an alternator weighing 18 kg. A long static effort quickly caused pain; the operator was irritated and wanted to accomplish the task as soon as possible, which heightened the mental workload. Tasks 1 (displacement in the engine cowling) and 9 (checking and tightening torque) were the easiest, as 1 required little effort.

Table 4

Musculoskeletal Symptoms and Mental Workload (NASA-TLX) for Each Task on the VR and PMU Tools

Ergonomics Parameters	Task	N	VRM (SE)	PMUM (SE)
Musculoskeletal symptoms Lower back	1	6	0 (0)	0 (0)
	2	6	0.5 (0.1)	0.7 (0.1)
	3	5	0.4 (0.1)	0 (0)
	4	5	0.2 (0.1)	0.2 (0.1)
	5	5	0.2 (0.1)	0.2 (0.1)
	6	6	0 (0)	0 (0)
	7	5	0 (0)	0.2 (0.1)
	8	5	0 (0)––-*––––––1(0)
	9	5	0.2 (0.1)	0.4 (0.1)
	10	6	0 (0)	0.7 (0.1)
	11	5	0.2 (0.1)	0.4 (0.1)
Perceived physical exertion (Borg scale)	1	6	1.8 (0.3)	0.8 (0.2)
	2	6	5.7 (0.6)	4.5 (0.4)
	3	5	2.2 (0.3)	1.1(0.2)
	4	5	2.4 (0.3)	2 (0.3)
	5	5	3 (0.3)	3.4 (0.3)
	6	6	1.8 (0.2)	1.9 (0.3)
	7	5	2.5 (0.3)	2.4 (0.3)
	8	5	3 (0.2)	3.2 (0.4)
	9	5	3.2 (0.3)	1.3 (0.2)
	10	6	3.8 (0.3)	5.2 (0.6)
	11	5	3.4 (0.3)	3 (0.2)
Mental workload (NASA-TLX)	1	6	34.3 (6.1)	18.2 (5.9)
	2	6	61.3 (7.2)	48.8 (6.5)
	3	5	35.3 (6)	27.6 (3.7)
	4	5	37.6 (4.9)	30.1 (2.7)
	5	5	45.1 (7.1)	41.5 (5.3)
	6	6	32.4 (5.2)	28.1 (5.3)
	7	5	41.1 (5.6)	31.7 (6)
	8	5	43 (6.2)	37.8 (6)
	9	5	43 (4.6)	26.8 (3.9)
	10	6	51 (7.3)	56.1 (8.7)
	11	5	42 (5.5)	36.7 (5.3)

Note. NASA-TLX = National Aeronautics and Space Administration—Task Load Index; VR = virtual reality; PMU = physical mock-up.

Wilcoxon, *p ≤ .05.

Furthermore, the RULA score for each task was compared between DHM, VR, and PMU. Five tasks showed a significant difference in the RULA score between DHM, VR, and PMU (Table 5). Only Task 8 had a similar score between the three simulation tools (Table 5). Five tasks showed significant differences between the three simulation tools for safety at work parameters (Table 5). For these five tasks (2, 3, 5, 6, and 7), we observed similar characteristics with sufficient space to move the manikin in various postures (Figure 6). However, the environment was constrained for the rest of the tasks (2, 5, 7, 8, 9, and 10), and the manikin was placed intuitively.

Table 5

Comparison of the Three Simulation Tools Based on RULA Scores and Safety at Work Indicators for Each Task

Ergonomics Parameters	Task	N	Digital Human ModelingM (SE)	Virtual RealityM (SE)	Physical Mock-UpM (SE)
RULA	1*	6	3.33 (0.3)	2.5 (0.3)	2.7 (0.3)
	2	6	6.83 (0.2)	7	7
	3**	5	5.83 (1.2)	7	7
	4*	5	6.33 (0.3)	7	7
	5	5	6.83 (0.2)	7	7
	6**	6	5.5 (0.7)	4.17 (0.2)	4.16 (0.1)
	7	5	7	3.9 (0.2)	4
	8	5	7	7	7
	9	5	6 (0.4)	4	4
	10	6	6.17 (1.3)	7	7
	11*	5	6.67 (0.3)	7	7
Safety at work indicator	1	6	4	4	4
	2*	6	5.7 (0.3)	6.5 (0.3)	6.5 (0.3)
	3***	5	5.4 (0.3)	7	7
	4	5	4.8 (0.2)	5	5
	5**	5	5.6 (0.3)	6.4 (1.8)	4.4 (1.8)
	6***	6	4.83 (0.2)	4	4
	7**	5	6	4.4 (0.8)	4.4 (0.8)
	8	5	5	4.4 (0.8)	4.4 (0.8)
	9	5	6	6	6
	10	6	4	4	4
	11	5	6	6	6

Note. RULA = rapid upper limb assessment.

Friedman test: *p ≤ .05. **p ≤ .01. ***p ≤ .001.

Figure 6.

The operator performing a maintenance task with similar ergonomic characteristics for digital human modeling, the virtual reality platform, and the physical mock-up.

A significant difference existed for Tasks 3 and 6 for both the RULA score and safety at work score. We found the same safety at work score for 54.5% of the tasks for three simulation tools. The safety at work score for Task 3 was 7 on VR and PMU and higher than 5.4 for DHM, indicating the minor risks at the work situation area. However, the safety at work score was less than 6.5 for the rest of the tasks on three simulation tools—reaching 4 for 24% of tasks (indicating a significant risk for the operators).

By comparing three simulation tools, it was possible to demonstrate that some maintenance tasks could not be reproduced satisfactorily (Figure 7), particularly for Tasks 3 and 6. These tasks present a significant difference between DHM, VR, and PMU for the indicators of safety at work and the RULA score. Both Tasks 3 and 6 aimed to manipulate an electrical generator and a hydraulic drain. For Task 6, the manipulation of the drain collector is performed from the platform, without real environmental con-straints (helicopter structure, wire, etc.). Poor accessibility and the resulting awkward posture were the common risk factors for Tasks 3 and 6. The maintenance operators had to work in a kneeling posture and pass their upper body through many electrical wires, hydraulic pipes, and mechanical systems.

Figure 7.

A task with different postures on digital human modeling, virtual reality, and physical mock-up for the same operator.

Moreover, 11 maintenance tasks were performed in the confined space—without the possibility of prolonged displacement and moving around the helicopter—in the work area (limiting the freedom of motion). This condition reduced the variability of postures and the operational leeway of the maintenance operators. However, the maintainability stakeholders could easily make the best choice for placing DHM in the work environment.

Discussion

This study aims to evaluate the difference in estimating the common HFE indicators between three main simulation tools—DHM and VR in the digital category and PMU in the physical category—through a case study in the helicopter industry. We found that the estimation difference for the majority of HFE indicators was not significant between VR and PMU, except for time execution. However, DHM, VR, and PMU estimated different scores for posture assessment (RULA) and safety indicators for several tasks.

The duration of task execution between the theoretical time estimated by the MTM method and the task execution on VR and PMU was different and the tasks performed on both simulation tools required an execution time that was higher than the theoretical time. Only two tasks did not present a significant difference (Tasks 9 and 11) because they were simple tasks (with a few sequences and without significant effort). Theoretical time was estimated correctly. We found that the execution time for more than half of the tasks was higher on the VR platform than on the PMU tool. One of the reasons may be related to the perception of the environment, which could lead to operators being distracted. In our study, the spatial perception was not realistic (Loomis & Philbeck, 2008) due to the gap between perception of the real environment (that maintenance operators are already familiar with) and perception in the virtual platform. The volume of the room limits the workspace in the VR platform, and new technologies (such as mixed reality) have to emerge to perceive the workspace well (Burns, Salter, Sugden, & Sutherland, 2018).

Moreover, the perception of the environment could be improved by the quality of the initial CAD transferred to the VR platform and by the graphics displayed inside the helmet that provides the optimized technical features such as a more realistic field of view (Bowman & McMahan, 2007). Some maintenance tasks (Tasks 1, 3, 6, and 7) required a large workspace in the VR room to ensure the same movement in the workspace volume perceived on the digital immersion. As the physical space of the VR platform limited the movement/displacement of the operators, they had to think and choose a strategy to reach the final destination in the digital immersion and to compensate for this difference.

On the contrary, several tasks required a higher execution time on PMU due to the limitation of physical contacts and the lack of representation of all the physical elements on the VR platform. For example, the torquing bolt was a frequent task in maintenance, but, we could not reproduce this forceful task in VR platforms. Lawson, Salanitri, and Waterfield (2016) have already detailed these specific requirements of VR. The development and integration of physical parts into the VR platform may improve the efficiency of simulation and reduce the gap between real and digital simulation. Therefore, we inserted into the VR platform the physical parts (hatch, table, or wall) to represent the primary physical contacts and mobile parts with a tracker (real torque wrench and an alternator) to reproduce force, similar to previous studies (Menezes, Gouveia, & Patrão, 2018). Several studies proposed the haptic system to improve the quality of VR platform in aeronautic (Savall, Borro, Gil, & Matey, 2002), medical (Wang et al., 2017), and automotive industries (Langley et al., 2016). However, this technology is costly, which justifies exploiting the current potential of tracking technology (Meier & Holzer, 2015; Riley, 2016). Using real parts with trackers is cheaper than the haptic system and provides a more realistic HFE assessment.

Furthermore, although the participants in our study were trained and immersed in the VR platform 1 week before the experiment, they were unsettled at the beginning of the experiment, needed a few minutes to find their bearings, and adapted to a virtual environment by considering their knowledge of the real world. Motion sickness, especially during a long virtual simulation, also appeared to be a considerable issue (Chen et al., 2017). A long break between the tasks was required in the VR platform to enable the operator to continue the experiment under safe conditions. Therefore, the sequence may not be sufficiently realistic because, in reality, an operator does not take a break between each task.

Our findings suggest that managing the perception of the environment and physical contact were primary influences on the work sequences and the organizational aspect. Indeed, the realism of contacts and forces with real parts and maintenance tools tracked on the VR platform superimposed virtual and real contact, reproducing the force and posture perception (Seth, Vance, & Oliver, 2011). Therefore, the assessment of the other HFE indicators (posture, force, safety parameters, mental workload, and MSD symptoms) was not significantly different between VR and PMU in this study. Pontonnier et al. (2014) reported a difference between VR and PMU to assess HFE indicators when the physical contact and force are not present on VR.

The study of the reach envelope between DHM, VR, and PMU did not present any significant difference as it systematically gave an optimal score of 100% for 11 tasks and six operators. Five tasks showed a significant difference for the RULA score and safety indicators between DHM compared with VR and PMU. These tasks concerned specific work situations where the operators must typically be in contact with the environment; however, when these situations were developed on DHM, the engineers did not have sufficient technical information about all the potential contacts between the various parts of an architecture. For example, the engineer put in contact the manikin on DHM with the firewall in Task 4, but this contact is forbidden because the firewall could be broken. Therefore, the posture for Task 4 is more comfortable in DHM than in VR and PMU, which resulted in the lowest posture score on DHM. The lack of communication on technical information between different design office departments and maintainability engineers performing the analysis may be the reason for the differences observed for safety indicators between the tasks in DHM and VR/PMU. For example, we observed the same behavior, posture, and contact with the environment in all three simulation tools (Figure 6) for Task 2 (disassembly/assembly of an alternator weighing 18 kg). The main reason for this similarity, especially on the DHM, might be related to the confined space, in which the maintainability stakeholders had only one choice to place the manikin. However, there were two safety risks that the DHM analysis did not report—cutting the edge of the firewall and slipping due to oil leakage on the floor (explaining the lower safety score on DHM).

The DHM lacked detail in the digital environment, and it may mislead the maintainability stakeholder when performing the simulation. This difficulty might be due to using DHM in the very preliminary stage of the design process, where the definition of the work situation is not optimal, and the information transmitted to the maintainability engineer will change every day because of the iterative process (Pistikopoulos, Vassiliadis, & Papageorgiou, 2000). The maintainability engineer performs the HFE analysis when the architectural structure of the product is not yet frozen on one or several complex tasks in DHM, independently of the other tasks in a work area, meaning that the study excludes other maintenance tasks in this area (Bernard, Zare, Sagot, & Paquin, 2018; De Sa & Zachmann, 1999; Regazzoni & Rizzi, 2014). The architectural structure would probably change throughout the design process, while DHM would not be repeated or would only be repeated for a specific task due to various constraints such as time limitation and the limitations of other departments. By changing this paradigm and studying a zone by DHM instead of the specific complex tasks, we can include several tasks very early in an area, and the efficiency of modification on the new architecture would improve. This paradigm of performing the HFE analysis on a specific task is the same for VR and PMU. This approach may be effective with VR and PMU because these simulation tools are used later in the design process, when the technical information is more stable and complete. However, we propose changing this paradigm for the DHM study and performing the HFE analysis according to the work area. The HFE indicators may be assessed in DHM according to the work area by selecting and simulating various maintenance tasks (the most complex one that affects the operator’s activity). This approach may make it easier to gain an overview of the central issues in the interaction between the digital manikins and the environment.

One of the limitations of this study is the small sample size of maintenance operators, which was due to the industrial constraints. Furthermore, one of the participants could not finish all of the experiments, and we had to exclude him from our study. It may be useful to test these simulation tools and other new technologies (augmented reality and mixed reality) with a large population and in several work areas. The results of this experiment should be generalized only with caution, as working on the other parts of the helicopter (such as the tail rotor, inside the cabin, or the cockpit) may provide different results for each simulation tool. The operator’s perception may be different when completing the maintenance task in a real situation (typically in the cabin within a confined space) compared with the situation in our experiment where the tasks were performed at heights of three meters on the upper deck (engine).

Our findings open up a debate on the application of the simulation tools that significantly influences the integration of HFE into the design process, not only the maintainability department but also all departments that are connected by a single project. Therefore, we propose developing a tool in a further study explicitly aimed at those who are not experts in HFE to use simulation tools (DMU and PMU) in an efficient manner during design development.

Conclusion

Through this industrial experiment on the maintenance activity of a helicopter manufacturer, we confirmed potential interest related to using the digital simulation tool through the VR platform by inserting physical parts to simulate real contacts and forces with the participation of the real operators. We found similar results for biomechanical indicators (such as posture and force) and mental workload in VR and PMU. To improve the assessment of the duration of time execution, we propose improving operator training in VR and better anticipating the physical side added in the VR room. In addition, the DHM produces a significant difference for various maintenance tasks that highlights the need for developing a new paradigm: during a maintainability analysis, do not perform an HFE analysis by individual task, but instead by area, including various maintenance tasks.

Key Points

VR simulation must be better anticipated to enhance the integration of real physical parts in the VR room;

In VR and PMU, we found similar results for biomechanical indicators (such as posture and force) and mental workload;

Change the paradigm of using DHM: Do not perform an HFE analysis for individual tasks, but instead by area, including various maintenance tasks.

Simulation tools (physical and digital), common between ergonomists and engineers, can be better used to assess ergonomics, especially if the right practices are specified and made known to all actors in the design office.

Footnotes

Acknowledgments

I express my deepest and sincere gratitude to all the operators who participated in this experimentation: Jean-Michel, Franck F., Franck R., Charly, Eric, Philippe, and Thierry.

This research was funded through PhD studies between University of Bourgogne Franche-Comté (France) and a French helicopters manufacturer. Funding information: Association Nationale de la Recherche et de la Technologie, Grant/Award Number: N°2015/1306; French National Association of Research and Technology. This paper is based on an earlier presentation at the International Ergonomics Association conference (IEA 2018)

ORCID iD

Fabien Bernard

Fabien Bernard worked 2 years as a research engineer at UTBM in the University of Bourgogne Franche-Comté before starting his PhD studies in the maintainability department of Airbus Helicopters France. He teaches in the Department of Ergonomics, Design, and Mechanical Engineering (EDIM at UTBM) and in the Aix-Marseille University (AMU).

Jean-Claude Sagot is the head of the research team of Ergonomics and System Design (ERCOS) at UTBM in the University of Bourgogne Franche-Comté. He teaches in the Department of Ergonomics, Design, and Mechanical Engineering (EDIM).

Mohsen Zare is an associate professor at the University of Bourgogne Franche-Comté, more precisely, at UTBM on the ERCOS team. He teaches in the Department of Ergonomics, Design, and Mechanical Engineering (EDIM).

Raphael Paquin is maintainability team leader at Airbus Helicopters, and has been working in this department since 2006. He uses Virtual Reality in Design Office to assess and improve maintainability on new aircraft during development phases. He teaches at the Aix-Marseille University (AMU).

References

Ahmed

Gawand

M. S.

Irshad

Demirel

H. O.

(2018, August 26–29). Exploring the design space using a surrogate model approach with digital human modeling simulations. Paper presented at the ASME 2018 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference (paper no. V01BT02A011), Quebec City, Quebec, Canada.

Amundarain

Borro

Matey

Alonso

A. G.

de Guipúzcoa

(2003, November 5–7). Occlusion culling for the visualization of aeronautical engines digital mock-ups. Paper presented at the proceedings of Virtual Concept, Biarritz.

Aromaa

Väänänen

(2016). Suitability of virtual prototypes to support human factors/ergonomics evaluation during the design. Applied Ergonomics, 56, 11–18.

Association française des ingénieurs et responsables de maintenance. (2004). Association française des ingénieurs et responsables de maintenance- santé et sécurité au travail: les métiers de la maintenance en première ligne. Guide nationale de la maintenance [French Association of Maintenance Engineers- Health and safety at work: National maintenance guide]. Paris, France: Author.

Bernard

Bazzaro

Paquin

Sagot

J. C.

(2017, January 23–26). Consideration of human factors in aeronautical maintainability. Paper presented at the Annual Reliability and Maintainability Symposium, Orlando, FL.

Bernard

Zare

Sagot

J. C.

Paquin

(2018). Virtual reality simulation and ergonomics assessment in aviation maintainability. In Bagnara

Tartaglia

Albolino

Alexander

Fujita

(Eds.), Congress of the International Ergonomics Association (pp. 141–154). Cham, Switzerland: Springer.

Bernard

Zare

Sagot

J. C.

Paquin

(2019). Integration of human factors into the design process of helicopter maintainability. Human Factors and Ergonomics in Manufacturing & Service Industries, 29, 350–360.

Bittencourt

J. M.

Duarte

Béguin

(2017). From the past to the future: Integrating work experience into the design process. Work, 57, 379–387.

Borg

(1982). Psychophysical bases of perceived exertion. Medicine and Science in Sport and Exercise, 14, 377–381.

10.

Bowman

D. A.

McMahan

R. P.

(2007). Virtual reality: How much immersion is enough? Computer, 40(7), 36–43.

11.

Boy

G. A.

(2018). From rigid to flexible–from virtual to tangible an evolution of human-centered design. In Bagnara

Tartaglia

Albolino

Alexander

Fujita

(Eds.), Congress of the International Ergonomics Association (pp. 54–63). Cham, Switzerland: Springer.

12.

Broberg

(2007). Integrating ergonomics into engineering: Empirical evidence and implications for the ergonomists. Human Factors and Ergonomics in Manufacturing & Service Industries, 17, 353–366.

13.

Burns

Salter

Sugden

Sutherland

(2018). Virtual reality environment with real world objects. US9865089B2 Patent.

14.

Chang

Y.-H.

Wang

Y.-C.

(2010). Significant human risk factors in aircraft maintenance technicians. Safety Science, 48, 54–62.

15.

Chen

Chao

J. G.

Zhang

Wang

J. K.

Chen

X. W.

Tan

(2017). Orientation preferences and motion sickness induced in a virtual reality environment. Aerospace Medicine and Human Performance, 88, 903–910.

16.

Chitescu

Sagot

J. C.

Gomes

(2003, July). Favoriser l’articulation “Ergonomie/conception de produits” à l’aide de mannequins numériques [“Ergonomics/product design” using digital mannequins]. Dans actes de la conférence 10eme Séminaire CONFERE (Collège d’Etudes et de Recherches en Design et Conception de Produits) sur l’Innovation et la Conception, Belfort, France.

17.

Čokorilo

(2011). Aircraft performance: The effects of the multi attribute decision making of non time dependant maintainability parameters. International Journal for Traffic and Transport Engineering, 1(1), 42–48.

18.

Czerniak

J. N.

Brandl

Mertens

(2017). Designing human-machine interaction concepts for machine tool controls regarding ergonomic requirements. IFAC-PapersOnline, 50(1), 1378–1383.

19.

Das

Sengupta

A. K.

(1996). Industrial workstation design: A systematic ergonomics approach. Applied Ergonomics, 27, 157–163.

20.

De Leon

P. M.

Díaz

V. G. P.

Martínez

L. B.

Marquez

A. C

. (2012). A practical method for the maintainability assessment in industrial devices using indicators and specific attributes. Reliability Engineering & System Safety, 100, 84–92.

21.

De Sa

A. G.

Zachmann

(1999). Virtual reality as a tool for verification of assembly and maintenance processes. Computers & Graphics, 23, 389–403.

22.

Di Gironimo

Monacelli

Patalano

. (2004, April). A design methodology for maintainability of automotive components in virtual environment. Paper presented at the DS 32: Proceedings of DESIGN 2004, the 8th International Design Conference, Dubrovnik, Croatia.

23.

Dmitriev

Burlakov

Popov

(2015). Technological processes and quality control in aircraft engine maintenance. Aviation, 19(3), 133–137.

24.

Duffy

V. G.

(2016). Handbook of digital human modeling: Research for applied ergonomics and human factors engineering. Boca Raton, FL: CRC Press.

25.

EN 13306:2010. (2010). Maintenance-maintenance terminology.

26.

European Agency for Safety and Health at Work. (2010). Safe maintenance in practice. Luxembourg: Publications Offices of the European Union.

27.

Geng

Zhou

Wang

(2013). A modeling approach for maintenance safety evaluation in a virtual maintenance environment. Computer-Aided Design, 45, 937–949.

28.

Gruber

De Leon

George

Thompson

(2015). Managing by design. Academy of Management Journal, 58(1), 1–7.

29.

Guan

T. M.

Lei

J. B.

(2012, June). The human engineering analysis of the racing wheelchair. Advanced Materials Research, 538, 2802–2806.

30.

Hart

S. G.

Staveland

L. E.

(1988). Development of NASA-TLX (Task Load Index): Results of empirical and theoretical research. Advances in Psychology, 52, 139–183.

31.

Hashim

Kamat

S. R.

Halim

Othman

M. S.

(2014). A study on push-pull analysis associated with awkward posture among workers in aerospace industry. International Journal of Research in Engineering and Technology, 3, 233–239.

32.

Hobbs

A. N.

(2000). Maintenance “error,” lessons from the BASI survey. Flight Safety Australia, 4, 36–37.

33.

Horeman

Akhtar

Tuijthof

G. J.

(2015). Physical simulators. In Karahan

Kerkhoffs

G. M. M. J.

Randelli

Tuijthof

(Eds.), Effective training of arthroscopic skills (pp. 57–69). Berlin, Germany: Springer.

34.

International Ergonomics Association. (2000). “What is ergonomics,” “IEA members” and “study programs.” Retrieved from http://www.iea.cc/index.php

35.

ISO 15537. (2004). Principles for selecting and using test persons for testing anthropometric aspects of industrial products and designs.

36.

ISO 7250. (2017). Basic human body measurements for technological design.

37.

Johnson

T. L.

Fletcher

S. R.

(2015). The limitations of using only CAD and DHM in design relating to high value manufacturing. In Contemporary Ergonomics and Human Factors 2015: Proceedings of the International Conference on Ergonomics & Human Factors (pp. 329–332). London, England: Taylor & Francis.

38.

Krüger

Nguyen

T. D.

(2015). Automated vision-based live ergonomics analysis in assembly operations. CIRP Annals-Manufacturing Technology, 64, 9–12.

39.

Kuorinka

Jonsson

Kilbom

Vinterberg

Biering-Sørensen

Andersson

Jørgensen

(1987). Standardised Nordic questionnaires for the analysis of musculoskeletal symptoms. Applied Ergonomics, 18, 233–237.

40.

Lämkull

Hanson

Örtengren

(2009). A comparative study of digital human modelling simulation results and their outcomes in reality: A case study within manual assembly of automobiles. International Journal of Industrial Ergonomics, 39, 428–441.

41.

Langley

Lawson

Hermawati

D’Cruz

Apold

Arlt

Mura

(2016). Establishing the usability of a virtual training system for assembly operations within the automotive industry. Human Factors and Ergonomics in Manufacturing & Service Industries, 26, 667–679.

42.

Laring

Forsman

Kadefors

Örtengren

(2002). MTM-based ergonomic workload analysis. International Journal of Industrial Ergonomics, 30, 135–148.

43.

Lawson

Salanitri

Waterfield

(2016). Future directions for the development of virtual reality within an automotive manufacturer. Applied Ergonomics, 53, 323–330.

44.

Lee

S. G.

Y. S.

Thimm

G. L.

Verstraeten

(2008). Product lifecycle management in aviation maintenance, repair and overhaul. Computers in Industry, 59, 296–303.

45.

Lockett

H. L.

Arvanitopoulos-Darginis

(2017). An automated maintainability prediction tool integrated with computer aided design. Procedia CIRP, 60, 440–445.

46.

Loomis

J. M.

Philbeck

J. W.

(2008). Measuring spatial perception with spatial updating and action. In Carnegie Symposium on Cognition, 2006 (pp. 17–60). Pittsburgh, PA: Psychology Press.

47.

Zhou

(2015). Maintainability fuzzy evaluation based on maintenance task virtual simulation for aircraft system. Eksploatacja i Niezawodność, 17(4), 504–512.

48.

Maurino

D. E.

Reason

Johnston

Lee

R. B.

(2017). Beyond aviation human factors: Safety in high technology systems. New York, NY: Routledge.

49.

McAtamney

Corlett

E. N.

(1993). RULA: A survey method for the investigation of work-related upper limb disorders. Applied Ergonomics, 24(2), 91–99.

50.

Meier

Holzer

(2015). Method and device for illustrating a virtual object in a real environment. US9165405B2 Patent.

51.

Meister

(2014). Human factors testing and evaluation (Vol. 5). Amsterdam, The Netherlands: Elsevier.

52.

Menezes

Gouveia

Patrão

(2018). Touching is believing-adding real objects to virtual reality. In Auer

Zutin

(Eds.), Online Engineering & Internet of Things (pp. 681–688). Cham, Switzerland: Springer.

53.

Neumann

(2006). Inventory of tools for ergonomic evaluation. Stockholm, Sweden: Arbetslivsinstitutet, förlagstjänst.

54.

Nishanth

Muthukumar

M. V.

Arivanantham

(2015). Ergonomic workplace evaluation for assessing occupational risks in multistage pump assembly. International Journal of Computer Applications, 113(9), 9–13.

55.

Perez

de Looze

M. P.

Bosch

Neumann

W. P.

(2014). Discrete event simulation as an ergonomic tool to predict workload exposures during systems design. International Journal of Industrial Ergonomics, 44, 298–306.

56.

Pistikopoulos

E. N.

Vassiliadis

C. G.

Papageorgiou

L. G.

(2000). Process design for maintainability: An optimization approach. Computers & Chemical Engineering, 24, 203–208. doi:10.1016/S0098-1354(00)00514-7

57.

Pontonnier

Dumont

Samani

Madeleine

Badawi

(2014). Designing and evaluating a workstation in real and virtual environment: Toward virtual reality based ergonomic design sessions. Journal on Multimodal User Interfaces, 8, 199–208.

58.

Quintero-Duran

Paul

(2016, June 15–17). Ergonomic assessment of a physical task using two different DHM systems. Paper presented at the 4th International Digital Human Modeling Symposium (DHM2016), Montreal, Quebec, Canada.

59.

Rasmussen

Damsgaard

Christensen

S. T.

Surma

(2002). Design optimization with respect to ergonomic properties. Structural and Multidisciplinary Optimization, 24, 89–97. doi:10.1007/s00158-002-0219-x

60.

Regazzoni

Rizzi

(2014). Digital human models and virtual ergonomics to improve maintainability. Computer-Aided Design and Applications, 11(1), 10–19.

61.

Riley

(2016). System and method of providing access to virtual spaces that are associated with physical analogues in the real world. US9403087 Patent.

62.

Sagot

J. C.

(1999). Ergonomie et conception anthropocentrée (Document pour l’Habilitation à diriger des recherches). Nancy, France: Institut National Polytechnique de Lorraine (INPL).

63.

Savall

Borro

Gil

J. J.

Matey

(2002). Description of a haptic system for virtual maintainability in aeronautics. In IEEE/RSJ International Conference on Intelligent Robots and Systems, 2002 (Vol. 3 pp. 2887–2892). New York, NY: IEEE.

64.

Seth

Vance

J. M.

Oliver

J. H.

(2011). Virtual reality for assembly methods prototyping: A review. Virtual Reality, 15(1), 5–20.

65.

Sharma

H. K.

Singhal

Sonia

(2018). Computer-assisted industrial ergonomics: A review. In Ray

P. K.

Maiti

(Eds.), Ergonomic design of products and worksystems-21st century perspectives of Asia (pp. 37–48). Singapore: Springer.

66.

Shi

Liu

(2012). Application of DELMIA on maintainability design of aircraft. In Proceedings of the 2012 International Conference on Computer Application and System Modeling (pp. 771–774). Paris, France: Atlantis Press.

67.

Spenser

(2008). The airplane. New York, NY: Collins.

68.

Stoffregen

Bardy

B. G.

Smart

Pagulayan

(2003). Virtual and adaptive environments: Applications, implications, and human performance issues. In Hettinger

L. J.

Haas

M. W.

(Eds.), On the nature and evaluation of fidelity in virtual environments (pp. 111–128). Mahwah, NJ: Lawrence Erlbaum.

69.

Vyavahare

R. T.

Kallurkar

S. P.

(2015, December). Ergonomic evaluation of electric hedge trimmer using digital human modeling. Agricultural Engineering International: CIGR Journal, 17(4), 100–108.

70.

Wang

Yao

Wang

Liu

Wang

Zheng

(2017, March). A surgical training system for four medical punctures based on virtual reality and haptic feedback. In 2017 IEEE Symposium on 3D User Interfaces (3DUI) (pp. 215–216). New York, NY: IEEE.

71.

Wiener

E. L.

Nagel

D. C.

(Eds.). (1988). Human factors in aviation. Houston, TX: Gulf Professional Publishing.