Advances in Human Factors in Simulation and Modeling

This paper discusses the development and use of an analytical assessment methodology that applies Systems Engineering principles, Ecological Affordance Theory, and Human Abilities, to measure the potential of integrated simulation training environments (ITEs) to support the development of competence in the execution of specific military missions. The results of this research include the development and use of the integrated training environment assessment methodology (ITEAM). ITEAM was used to re-evaluate the ecological validity of several ITEs ability to support the development of specific competencies during training.

We aim to better inform the scientific community regarding test and evaluation techniques for validating devices that will potentially be used by individuals interfacing with autonomous robotic teammates (particularly, members of the U.S. Military). Testing within immersive virtual environments (IVRs) similar to those experienced in military operations will be discussed with focus on the use of a commercial gaming engine for task development. Highlights of using commercial gaming engines will be illustrated throughout the paper to emphasize their utility for evaluating future technologies with attention given to testing efficiency and ecological validity. The study of interactions with simulated agents and future communication devices will be described in the context of the Robotics Collaborative Technology Alliance (RCTA) research program.

Spatial layout of workplaces and geometric analysis of future products is a prominent application of Digital Human Models (DHMs). DHMs describe characteristics of body dimensions, body shape and motions of the future workers and users. Further applications of DHM are animated, computer-generated and photo-realistic figures for populating computer games and movies. In contrast to these applications, other areas of human modeling, e.g. human behavior modeling or cognitive modeling, have not been applied broadly.These models address different levels of human behavior, including human information processing. This paper presents several of these models and uses them as a basis for generating the idea of a comprehensive digital human model. Such a model is applicable for optimizing complex work processes as well as populating virtual environments. It is concluded that there will be no single solution for modeling and simulation all variations and aspects of human behavior, but that there is a need for a reference architecture as a generic interface between the different models.

Rendering virtual environments for simulated reality applications often proves to be a challenge due to the high demands of simulations capable of imparting enough detail to appease the human eye. Traditional simulated environments typically require hardware capable of providing enormous graphical processing power in order to render entire scenes in high detail, limiting simulations to high end computers. However, most of this detail is wasted, as human eyes are only capable of perceiving detailed information in a small central field of view. In the peripheral regions of human vision, the majority of the final image that is seen by the human is filled in by the brain, based on context and the minimal detail provided by the eyes. This three-phase research effort attempts to identify whether reducing the level of detail of objects in the peripherals of a virtual reality simulation affects the perceived quality of the simulation.

When developing military capabilities, it is essential to determine the effect that equipment, tasks, and training will have on Soldier and unit readiness. Human factors processes require technology enablers that streamline the systems engineering design process. Toward this end, a collaboration between the U.S. Army Science and Technology Objectives for the Soldier Systems Engineering Architecture (SSEA) and Training Effectiveness for Simulations (TEfS) was established in 2015. An operational SSEA is envisioned to enable analysts to implement systems engineering procedures to achieve more effective results in predicting best system designs. Best practices from TEfS is critical to improving the predictive analyses. In this paper we describe best practices from a squad training use case and propose how the training method could be applied as a test protocol in SSEA SaaS modeling.

Research into the human factors of augmented reality (AR) systems goes back nearly as far as research into AR. This makes intuitive sense for an interactive system, but human factors investigations are by most estimates still relatively rare in the field. Our AR research used the human-centered design paradigm and thus was driven by human factors issues for significant portions of the development of our prototype system. As a result of early research and more recent prototype development, mobile AR is now being incorporated into military training and studied for operational uses. In this presentation, we will review human factors evaluations conducted for military applications of mobile AR systems as well as other relevant evaluations.

Thermal performance of three bicycle helmets for latent heat loss was evaluated through a virtual testing methodology using Computational fluid dynamics (CFD) simulations. The virtual thermal manikin was prescribed with a constant sweat rate of 2 g/h and a constant sweat film thickness of 0.3 mm. The simulations were carried out at 6 m/s until convergence was achieved. The results from steady state simulations show heat loss of 158 W from manikin without helmet and approximately 135 W with helmets. However, the thermal performance of helmets with a sweating manikin has been reduced from 89–93% to 84–87%. These results imply that evaporative/latent heat loss plays a significant role in thermal performance of helmets. Therefore, thermal performance tests for helmets should also include testing of helmets for evaporative heat loss.

Collaborative immersive environments are an important way for communication and interaction between people located in distant areas. This paper presents the development of virtual worlds for this class of 3D applications. It focuses at their use in a collaborative virtual classroom for the improvement of the teaching methods. The paper includes a review of virtual worlds and 3D modeling studies, it discusses the role of ergonomics, with the purpose of designing 3D models that follow ergonomics guidelines and principles and the technical specifications. The developed worlds are expected to convey a realistic and immersive experience to the user.

Manned-unmanned teaming is the synchronization of Soldiers, manned and unmanned vehicles, and sensors that may improve situational understanding, greater lethality, and improved survivability during military operations. However, since unmanned vehicle autonomy capabilities are constantly advancing, it is difficult to integrate the human team and assess the performance of the team during early design. This work provides an overview of the US Army Wingman program and the human factors integration and assessment capabilities that support improved manned-unmanned teaming performance during joint gunnery operations. The discussion culminates with human integration and team assessment capabilities for interaction with respect to both fielded and software-in-the-loop simulation systems.

This article reviews advancements in methods for detection of task-induced driver fatigue. Early detection of the onset of fatigue may be enhanced by spectral frequency analysis of the electrocardiogram (ECG) and analysis of eye fixation durations. Validity may also be improved by developing algorithms that accommodate driver sleep history assessed using mobile actigraphic methods. Challenges to development of fatigue indices include ensuring that metrics are valid across the range of task demands encountered by drivers. Future autonomous vehicles will place novel demands on the driver, and research is needed to test the applicability of current fatigue metrics.

This work proposes a process for formulating a model and estimation scheme to predict changes in decision authority with a simulated autonomous driving assistant. The unique component of this modeling approach is the use of direct estimation of governing mental decision states via recursive psychophysiological inference. Treating characteristic quantities of the environment as inputs, and behavioral and physiological signals as outputs, we propose the estimation of intermediate or underlying psychological states of the human can be used to predict the decision to engage or disengage a driving assistant, using methods of stochastic filtering. Such a framework should enable techniques to optimally fuse information and thereby improve performance in human-autonomy driving interactions.

In this driving study, participants were assigned to a driver-passenger dyad and performed two drives along Interstate-95 in normal traffic conditions. During the driving session, the driver had to safely navigate the route while listening and discussing news stories that were relayed by the passenger. The driver then performed a set of memory tasks to evaluate how well they retained information from the discussion in a multitask context. We report preliminary analyses that examined subjective factors which may influence success in social communication, including trait and state similarity derived from questionnaires as well as physiological synchrony from implicit state measurements derived from brain activity data. Although this dataset is still in collection, these initial findings suggest potential metrics that capture the contextual complexity in naturalistic, multitask environments, providing a rich opportunity to study how successful communication reflects shared social and emotional experiences.

There are a number of reasons to use computer-based simulation in human-robot interaction research. Most predominant is the assessment of humanin-the-loop interactions for robotic technologies that do not yet exist, are in prototype development, or are in early test and evaluation stages of development. In these cases, simulation can provide insight into how the human may interact with the real world robotic vehicle. However, there are a number of challenges to designing and developing these simulations so that findings translate to interaction with real-world systems. This work specifically looks at the incorporation of human factors into driving simulations for robotic military vehicles. Three use cases address a number of these challenges.

Trust is important in group interactions; however, little is known about trust in wellness-related telehealth training. This study examined self-reported trust in U.S. military active duty and veterans (n = 45) who participated in an 8-week mindfulness course offered in the Virtual World (VW) of Second Life. Participants completed a VW Trust Questionnaire (VWT, measuring relational trust such as communication, confidentiality, and self-representation) and a Trust in Automation Questionnaire (TIA, measuring confidence in system and perceived system security, integrity, dependability, and reliability) post training. Participants reported moderately high levels of TIA and high relational trust (VWT). Higher class attendance was associated with being comfortable speaking in the VW and belief in confidentiality (relational trust). Higher attendance was also associated with higher TIA. These results demonstrate that individuals are more likely to participate in virtual world telehealth interventions, and complete more of their training, when their trust is high.

As part of an olfactory adaptation experiment, researchers collected NASA Task Load Index (TLX) data from United States Military Academy cadets who completed two complex tasks in either the presence or absence of a simulated malodor. Results showed that participants exposed to the odor twice tended to show a decrease in perceived mental demand during the second task. Furthermore, these participants also showed a higher correlation coefficient between decrease in mental demand and improvement in task performance. Taken together, these results indicate a possible link between olfactory adaptation and perceived mental demand, at least in the presence of a malodor.

Statistical body shape modeling (SBSM) is a well-known technique to map out the variability of body shapes and is commonly used in 3D anthropometric analyses. In this paper, a new approach to integrate movement acquired by a motion capture system with a body shape is proposed. This was done by selecting landmarks on a body shape model, and predicting a body shape based on features. Then, a virtual skeleton was generated relative to those landmarks. This skeleton was parented to a body shape, allowing to modify its pose and to add pre-recorded motion to different body shapes in a realistic way.

Parametric 3D human body models are valuable tools for ergonomic product design and statistical shape modelling (SSM) is a powerful technique to build realistic body models from a database of 3D scans. Like the underlying 3D scans, body models built from SSMs are typically represented with triangle meshes. Unfortunately, triangle meshes are not well supported by CAD software where spline geometry dominates. Therefore, we propose a methodology to convert databases of pre-corresponded triangle meshes into multi-patch B-spline SSMs. An evaluation on four 3D scan databases shows that our method is able to generate accurate and water-tight models while preserving inter-subject correspondences by construction. In addition, we demonstrate that such SSMs can be used to generate design manikins which can be readily used in SolidWorks for designing well conforming product parts.

Complex unmanned aerial vehicle (UAV) operations place high information processing demands on operators. Dynamic automation or function allocation (DFA) has been proposed as a method to address operator “overload” situations without compromising system/situation awareness. This study made use of a high-fidelity UAV simulation to investigate any benefits of DFA when applied to decision making tasks. Three modes of UAV control automation, including DFA and static high and low level of automation, were compared. A preliminary analysis of operator responses to situation awareness queries showed no significant differences; however, response accuracy in an aircraft detection task indicated superior situation comprehension under DFA as compared to static high level of automation. Results of an embedded secondary task indicated that participants experienced lower cognitive workload when operating under DFA as compared to static low level of automation. Findings of this study could be used for guiding design of DFA approaches for UAV control.

In recent years, understanding and modeling the human driver in the scope of traffic simulations has received considerable attention. With the advent and the ongoing development of new technologies in the field of Intelligent Transportation Systems, we are consequently moving towards an era where a majority of driving-related tasks will presumably be carried out by autonomous systems rather than humans. Notwithstanding, the transition from today’s conventional traffic to tomorrow’s highly automated traffic will not take place overnight. Up to that point, the available transportation infrastructure will most likely be shared among both human-driven and (partially) automated vehicles. Considering such scenarios of mixed traffic is therefore inevitable when developing new concepts and applications for the use in ITS, and requires a proper modeling of the human driver for simulation purposes. Although there have been diverse ways of integrating human factors with traffic simulation models, most existing studies focus on the impacts of human driving behavior in very constrained scenarios such as isolated platoons or bottleneck situations rather than on their large-scale effects. In this paper, we address this particular issue by performing large-scale simulations to investigate the impacts of human behavior on vehicular traffic flow under varying traffic conditions. We show how specific factors such as delayed reaction, distracted or anticipatory driving affect traffic efficiency and safety in terms of travel time, fuel consumption and accident frequency.

The purpose of this paper is to spatially validate an agent-based predictive analytics model of energy siting policy in a techno-social space. This allows us to simulate the multitude of human factors at each level (e.g. individual, county, region, and so on). Energy infrastructure siting is a complex and contentious process that can have major impacts on citizens, communities, and society as a whole. Furthermore, the process is sensitive to varying degrees of human input, of differing complexity, at multiple levels. When it comes to validating ABMs, the virtual cornucopia of techniques can easily confuse the modeler. As useful as historical data validation is, it seems to be underutilized, most likely due to the fact that it is hard to find data suitable data for many models. For the purpose of In-Site, historical data availability is excellent due to Environmental Impact Assessments (EIA) providing us with citizen and community based organization (CBO) preferences, and regulatory decisions being public. For the model, citizen and CBO preferences were decided by coding comments on the EIA procedure so as to allow for quantitative analysis, and then geocoding the locations of the commenters. The end results of this is that, we can literally overlay our simulation results with the actual, real world, results of the historical project. This will allow for a high degree of confidence in the validation procedure, as well as the ability to deal with the complexity of the networks of human interactions.

In the real-world, manned and unmanned vehicles may be used for a number of applications. Visual technologies like indirect visual display (IVD) and virtual reality (VR) have been used to train operators in both manned and unmanned environments. The main objective of this research was to evaluate the effectiveness of manned and unmanned interfaces in IVD and VR display designs. Using an underwater search-and-shoot scenario, we developed two variations in display designs (IVD and VR) and two variations in type of interface-based training (manned and unmanned). A total of 60 subjects participated in the experiment, where 30 subjects were randomly assigned to simulations in IVD and the rest in VR. In both the simulations, 15 randomly selected participants executed the manned interface first and the remaining 15 executed the unmanned interface first. Results revealed that the subjects performed better in VR compared to IVD, and also performed better when they executed the unmanned interface first. We highlight the implications of our results for training personnel in scenarios involving manned and unmanned operations in IVD and VR interfaces.

Workload management is integral to the success of human-machine teams, and involves measuring and predicting workload and implementing proactive interventions to mitigate the adverse effects of degraded performance. Common approaches to workload measurement rely on the use of subjective, behavioral, and physiological metrics. These approaches suffer from two important limitations. First, the mapping between workload, subjective ratings, behavior, and physiology is complex and noisy, resulting in high uncertainty. Second, metrics based on subjective ratings, behavior, and physiology often fail to explain why performance degrades, and consequentially does not inform the development of mitigation strategies. As an alternative, we propose using cognitive metrics profiling (CMP) to improve the measurement and prediction of workload. This approach uses computational cognitive models to simulate the activity within individual cognitive systems, such as vision, audition, memory, and motor, to measure and understand workload. We discuss how CMP can be used in an unmanned vehicle control task.

This paper presents an agent-based model (ABM) to model systemic risk in the housing market from 1986 to 2017. We provide a unique approach to simulating the financial market along with demonstrating the phenomenon of emergence resulting from the interconnected-behavior of consumers, banks and the Federal Reserve. Consumers can buy or rent properties, and these agents own characteristics such as income and may be employed or unemployed. Banks own balance sheets to monitor their assets and liabilities and participate in the interbank lending market with one another. This tool can assess the complexity of the United States’ housing market, conduct stress tests as interest rates fluctuate, and explore the landmark financial crisis and epidemic of foreclosures. This is important because understanding the impact from increases in foreclosed properties and changes in rates can help policymakers and bankers have a better understanding of the complexity within the housing market.

For evaluating the redistributive effects of the reform of China’s personal income tax system, this research attempts to break through the limitations of traditional research methods and provide a micro, coherent and structural framework for experimental research on policy reforms. Based on the microcosmic survey data of CHIPs and the system of personal income tax in China, a micro-simulation model was established to comprehensively analyze and quantify the redistributive effect of the reform of the personal income tax system and to clarify the final destination of the effect of the policy. This research provides the modeling tools and the quantitative basis for the design and evaluation of the reform of redistribution system.

We addressed the measurement and modeling of contact forces during intense, short impact loading of the shoulder. Experimental recoil force measurements show significant intra- and inter subject dispersion, advocating a an approach without real shooters but based on a biomechanical model for the dynamic analysis of the recoil-shoulder interaction. This mechanical prototype would replace actual shooters and is being developed and optimized for the standard, military, standing shooting position, but the framework could easily be extended to other positions and applications like clay shooting and hunting. The obtained recoil parameters are more realistic than current measures. The latter are not registered in realistic conditions as they are not based on direct measurements of the recoil forces on the shoulder but evaluated in a free-to-move or blocked setup. We illustrate the effects of these conditions on the recorded recoil force.

Cognitive psychology and Artificial Intelligence (AI) have long been intertwined in the study of problem solving, learning, and perception. The early pioneers of AI, Herbert Simon and Allen Newell, drew as inspiration chess masters and from their study developed computer programs to mimic the problem solving abilities identified in chess masters. The understanding of chess strategies relied heavily upon characterizing the problem space as a combination of symbolic inference and statistical pattern matching, which allowed for a quick understanding of the environment by computer systems. Recently, robotics has emerged as an AI domain, and the problem space has proven a difficult one due to the sub-symbolic nature of the knowledge. As robotics has emerged as a field in AI, cognitive architecture researchers have continued to refine their understanding of cognition in new ways that allow for the duplication of human problem solving with limited resources. The goal of this manuscript is to inform the AI world of the successes cognitive architectures have produced with the hope that this knowledge can be transferred to AI, and more specifically, robotics.

Cyber analysts must work long hours and under intense pressure all while performing complex mental tasks. Research is needed to improve the execution of their tasks. We describe a research plan that can be used to provide cyber analysts with the assistance they need using adaptive aids. The Multi-Level Cognitive Cybernetics (MLCC) [1] approach provides a methodological approach to studying adaptive automation and advancing its development across multiple levels of analysis. We follow up on a previous paper [1] by focusing on how MLCC can be used to improve cyber analyst tasks. We use the breakdown of cyber situation awareness by D’Amico and Whitley [2] to analyze the different cyber tasks and provide general examples of aids that may assist with each of the three stages. A more in-depth example of a specific type of cyber adaptive aid is then provided. We conclude with a call for action to expand adaptive automaton into the cyber realm.

Agent-based models are commonplace in the simulation-based analysis of cyber security. But as useful as it is to model, for example, adversarial tactics in a simulated cyber attack or realistic traffic in a study of network vulnerability, it is increasingly clear that human error is one of the greatest threats to cyber security. From this perspective, the salient features of behavior are those of an agent making decisions about how to use a system, rather than an agent acting as an adversary or as a “chat bot” which functions merely as a statistical message generator. In this paper, we describe work to model a human dimension of the cyber operator, a user subject to different motivations that lead directly to differences in cyber behavior which, ultimately, lead to differences in the risk of suffering a “drive-by” malware infection.

We offer a revised conditional deterrence agent based model applied to global and regional nuclear proliferation issues. Further extending the dyadic logic already established in the deterrence literature helps anticipate more recent 21st century challenges generated by the proliferation of nuclear capabilities and their acquisitions by dissatisfied non-state actors. Key elements include relative capabilities, risk propensity associated with the status quo, and physical exposure to preemptive-attack or retaliation. This work continues to extend our previous complex adaptive system framework to generalize insights to deterrence environments with multiple competing actors. Our preliminary analysis suggests that deterrence is stable when the capabilities of a dissatisfied challenger are inferior to that of a dominant and satisfied defender. Conversely, deterrence is tenuous when a dissatisfied challenger approaches parity in capability with a more dominant and satisfied defender, or when a violent non-state actor obtains nuclear capability or other WMDs.

An emergency is a serious, unexpected, and potentially life-threatening situation requiring immediate action. Emergency evacuation is the most critical step to save people’s lives. The purpose of this paper is to provide a review of various factors to investigate human behavior under emergency situations. Computational modeling and simulation as a practical way to replicate human behavior change requires quantifying psychological and physical parameters. Previous studies on humans and animals, as well as common simulation approaches were reviewed. According to the results of this literature review, future experiments or simulations can consider not only physical parameters such as human dynamics, but also quantifying psychological parameters such as interpersonal relationship, goal-seeking behavior, decision-making differences, and many more.

Today, military decision-making is dependent on the ability to amalgamate information across sources of varying degrees of agreement. Given the increasing volume of information, automated methods to assist in the identification and prioritization of the most valuable or relevant information has become paramount. Relevant information is not only critical to situational awareness and the military decision-making process, but vital to mission success. Towards this end, the US Army Research Laboratory (ARL) has undertaken a research initiative to model and test how analysts perceive the Value of Information (VoI) in varying military context. The goal of this effort is to develop methodologies useful in the development of automated information agents. As a part of the VoI initiative, ARL conducted an experiment with Subject Matter Experts (SMEs) at the US Army Intelligence Center of Excellence (ICOE), where data was collected on how intelligence analysts’ amalgamate information given information content and source reliability within complementary and contradictory conditional associations. The resulting experimental data was incorporated into an Adaptive Control of Thought-Rational (ACT-R) model. Exercising the ACT-R cognitive model resulted in some interesting response behaviors not observed in the initial experiment. In an effort to better understand the perceptions (cognitive underpinnings) of a military intelligence analyst, this paper extends the previous effort and utilizes a crowdsourced experiment within Amazon Mechanical Turk (Mturk). The experiment captures many of the same conditional ratings encountered by the military analysts. Data gained from the Mturk experiment will be examined using the ACT-R model as a simulation to determine whether the same data distributions exist within a wider audience and as a direct comparison to the analyst’s responses. This paper will examine the Mturk experimental design, discuss the experimental apparatus implementation and provide an overview of the ACT-R model utilized to replicate the amalgamation strategies.

Historically, cognitive modeling has been an exercise in theory confirmation. “Cognitive architectures” were advanced as computational instantiations of theories that could be used to model various aspects of cognition and then be put to empirical test by comparing the simulation-based predictions of the model against the actual performance of human subjects. More recently, cognitive architectures have been recognized as potentially valuable tools in the development of software agents—intelligent routines that can either mimic or support human performance in complex domains. While the introduction of cognitive architectures to what has been regarded as the exclusive province of artificial intelligence is a welcome turn, the history of cognitive modeling casts a long shadow. In particular, there is a tendency to apply cognitive architectures as monolithic, one-off solutions. This runs counter to many of the best practices of modern software engineering, which puts a premium on developing modular and reusable solutions. This paper describes the development of a novel software infrastructure that supports interoperability among cognitive architectures.

Active mining operations are complex, dynamic environments that can present workers with an array of potential safety and health challenges. From missing fire extinguishers to large equipment and falling rocks, hazards exist that mineworkers must be cognizant of to keep themselves and their coworkers safe. While hazard identification is a key skill that mineworkers must possess to ensure workplace safety, the location and perceived risk of the hazards may alter this ability. To further explore these effects, NIOSH researchers conducted a study to characterize how mineworkers search for and identify hazards. Researchers asked participants to search 32 static panoramic scenes depicting typical locations at a surface stone mine—pit, plant, roadway, and shop—with each containing zero to seven hazards. Mineworkers tended to miss hazards when they were in clusters—i.e., where two or more hazards appeared within the worker’s central field of view. This paper examines the relationship of clustered hazards, perceived risk and identification accuracy and how location and experience affect it. Based on the results, strategies will be suggested that mineworkers can use to help identify hazards in their workplace.

To perform a successful workplace examination, mineworkers must be able to find and fix hazards at their workplace. NIOSH recently completed a laboratory study to identify differences in hazard recognition performance for mineworkers, safety professionals, and mining engineering students tasked with performing a simulated workplace examination in a virtual environment. The laboratory methodology and study results were used to develop a training product aimed at improving mineworker safety. The purpose of the current chapter is to describe the efforts that were taken to modify the laboratory workplace examination simulation into a portable software tool called EXAMiner, which can be used for data collection and training purposes in the field. This chapter provides an explanation of the literature and results from the NIOSH laboratory research studies used to inform and motivate development of EXAMiner. In addition, the software specifications are explained.

The shoe-floor interface is a key element in preventing slips and falls. The design of footwear and the floor surface should be considered to ensure worker safety. Testing various floor surface materials and boots in a real-world setting would impose unnecessary risk to participants and limit the extent of testing possible. Hence, two examples of high-fidelity physical simulation—an inclined grated metal walkway and a grated metal stairway—were built to evaluate risk factors for slips and falls associated with various walkway materials and boots with metatarsal guards. This paper discusses details and findings of the two studies. Also discussed are the advantages and disadvantages of using physical simulations of the environments, including decreased risk for participants and large space requirements for the experiment. Findings of the research can help select appropriate floor surface materials and boots for the mining industry and inform the use of future high-fidelity physical simulations.

In military environments, maintaining Auditory Situational Awareness (ASA) and providing protection from hearing hazard are often dueling priorities. Traditional passive hearing protection devices (HPD’s) can provide adequate protection to the soldier from impulsive and continuous noise hazards, but this can come at the cost of reduced ASA. Anecdotal reports indicate that many soldiers may forego HPD use entirely in an attempt to maximize ASA. However, unprotected ears can lead to temporary threshold shifts (TTS) and permanent threshold shifts (PTS) that can be much worse for ASA than HPD use. Tactical Communication and Hearing Protection Systems (TCAPS) are active electronic systems that can provide a potential solution to this problem by giving the soldier protection, environmental hearing and radio communications. This paper discusses evaluating the effectiveness of these systems through objective measures of attenuation and subject human sound localization.

New approaches are needed to improve outcomes for safety training in hazardous industries. We use an evidence-driven approach to identify the key attributes of serious games that have the potential to improve safety training. Following a detailed needs assessment, we identified four major themes of usability problems which may be addressed through gamification: Limited accessibility, lack of context, lack of consequence, and absence of practicum. Based on our analysis, a series of application prototypes was developed to improve safety training in the mining industry. In particular, we discuss Harry’s Hard Choices, a game for mining emergency response training. Pilot tests indicate high levels of user satisfaction and engagement and anecdotal evidence of training transfer.

The ultimate aim of training is to improve task performance towards expert level. Novices and experts differ in their capability to understand and make sense of sensory information (for example, perception on environmental hazard). Computer-aided training, from online course to immersive simulation such as Virtual Reality (VR) [1]. van Wyk and de Villiers [2] define VR-based training environments as “real-time computer simulations of the real world, in which visual realism, object behavior and user interaction are essential elements”. The use of VR-based training environments assumes that Human-Machine interaction stimulates learning processes through better experiencing and improved memorization, leading to a more effective transfer of the learning outcomes into workplace environments. However, there are many human factors (internally and externally), which have impact on the quality of the training and learning process which need to be identified and investigated. The present study was conducted with Coal Services Pty Ltd, a pioneering training provider for the coal mining industry in NSW, Australia. The research focussed on 288 rescuers and the specific training programs developed for them. In this article, initially factors affecting the quality of the training and learning process for underground mine rescuers have been identified and then measured by using pre- and post-training questionnaires. We attempted to determine how much of the trainees’ perceived learning could be explained by pre-training (9 in total) and post-training (16 in total) factors. The relatively small size of the sample (288 observations for 17 predictors) and the high level of correlation between variables led us to Principal Component Analysis (PCA). Principle Component Analysis (PCA) has been used to investigate the underlying relationship among different variables. This technique results in factor reduction based on hidden relationships. Based on the nature of the pre-training factors mostly contributing to each component we have used the first 3 Components to create 3 new aggregated variables: “Positive State of Mind” (Component 1), “Negative State of Mind” (Component 2) and “Technology Experience” (Component 3). Similarly, based on the nature of the post-training factors mostly contributing to each component we have used the first 3 Components to create 3 new aggregated variables: “Positive Learning Experience” (Component 1), “Negative Learning Experience” (Component 2) and “Learning Context” (Component 3).

Adaptive systems typically comprise components that interrelate and interact to enable the whole system to respond and adjust to changes in the environment, operator, and task in order to regulate or maintain a level of performance or homeostasis. In so doing, they enable a degree of individualization and customization for many technological innovations such as managing the use of automation. Adaptive systems often involve some kind of feedback or closed-loop which requires a criteria for determining invoking thresholds, as well as some type of classification algorithm that models the type of changes to which the system has to adapt. This paper outlines the issues, considerations, and challenges associated with classification in adaptive systems, and reviews several algorithms that implement the feedback loop in neuro-ergonomic applications. These include logistic regression, Naïve-Bayes, artificial neural networks (ANN), and support vector machines (SVM) techniques.

This research effort is to develop human simulation methods to predict and assess injuries due to the fatigue of a repetitive loading. Over the past few years, we sought to integrate high-fidelity computational methods for stress/strain analysis, namely finite element analysis (FEA), in combination with biomechanics predictions through digital human modeling and simulation (DHMS). Our previous work for the past 12 years has culminated with the development of a simulation environment called Santos™ that enables the prediction of human motion including many aspects of its biomechanics and physiological systems. The Santos environment provides a joint- and physics-based predictive simulation environment. Cumulative load theory states that repetitive activities precipitate musculoskeletal injury and suggests that cyclic load application may result in cumulative fatigue, reducing their stress-bearing capacity. Such changes may reduce the threshold stress at which the tissues of the joint components fail. Repetitiveness of the work activity has shown to be a strong risk factor for cumulative trauma disorders (repetitive strain injury). Hence, repetitive load cycling is a leading factor in the propensity for injury. Santos, and the developed method, are able to characterize the motion using an optimization algorithm that calculates the motion profiles (i.e., the kinematics of the motion across time for each degree of freedom for the body) and external forces during the task. An OpenSim model uses motion profiles and external forces to calculate forces of the muscles that articulate the joint. A multi-scale FEA system uses external forces and muscle forces for each task over a repetitive cycle as input. The FEA model of the selected joint (ankle) computes the stresses of all joint components at limited and random frames of the motion cycle. The Artificial Neural Network program (ANN) estimates the stress over the full motion cycle and compares current stresses of the components with the newly calculated yield strength that been affected by cyclic loading. As a result, the program indicates the injury status of the joint components. This paper presents promising results for this approach to predict and quantify injury in the ankle joint that is undergoing a cyclic repetitive motion. This integrated system is capable of showing the effects of various motions and task-parameters on the selected joint and by this we can modify tasks, save analysis time, and reduce the likelihood of injury.

This paper presents a registration framework for the construction of a statistical shape model of the human hand in a standard pose. It brings a skeletonized reference model of an individual human hand into correspondence with optical 3D surface scans of hands by sequentially applying articulation-based registration and elastic surface registration. Registered surfaces are then fed into a statistical shape modelling algorithm based on principal component analysis. The model-building technique has been evaluated on a dataset of optical scans from 100 healthy individuals, acquired with a 3dMD scanning system. It is shown that our registration framework provides accurate geometric and anatomical alignment, and that the shape basis of the resulting statistical model provides a compact representation of the specific population. The model also provides insight into the anatomical variation of the lower arm and hand, which is useful information for the design of well-fitting products.

Successful grasp, transfer, and release of objects with the hand are important movements for completing everyday tasks. This study’s objective is to understand the effect of an object’s surface and shape on hand grip function for heavy objects in a young age group. For their functional prevalence and significance, grasp and release movements have been incorporated into many clinical hand function assessments, such as the Box and Block Test (BBT), a common test used to assess people’s hand grip function. In the BBT, subjects transport a block from one side of a box to the other while crossing a partition and repeat the procedure as fast as possible. In this study we measured performance on the BBT in 20 right handed subjects between ages 20 to 30 years old. There were no statistically significant effects of object surfaces and shapes on hand grip function for heavy objects.

A large population will likely experience lower back pain during their lifetime. Severe cases of lower back pain can sometimes be caused by back conditions or diseases, eventually being alleviated through surgical procedure. Skilled surgeons can make educated decisions on the best procedure for their patients, but the development of a spine model that can estimate biomechanical properties of the spine could aid in surgical decision-making. This paper discusses the current state of the art of four approaches used to study spine biomechanics: in vivo experimentation, in vitro cadaveric testing, finite element analysis, and musculoskeletal modeling. It is concluded that using a combination of these methods can lead to more accurate spine models that could possibly lead to clinical use.

Due to certain demanding manual tasks the loads on the human hand can be high, which can cause several disorders, among which are also tendon disorders. Many researchers tried to quantify the loads and provide mathematical models for the tendons of the hand. Since experiments and measurements in vivo are complex and usually not viable, we developed a finite element model of a finger joint, which utilizes tendon/muscle force for the joint movement. Initial simulations of the fingertip finite element model with the developed tendon joint model have shown accurate biomechanical behavior of finger movement and soft tissue deformation. We also compared the results in terms of relationship between tendon force and resulting fingertip (reaction) force from the simulation to an in vivo experiment and have confirmed that the results of the developed finite element model correspond well to the experimental results.

Conventional musculoskeletal model often uses the optimization method to estimate muscle forces. However, the optimization method unfortunately does not usually consider the role of antagonistic muscles, which act in opposite direction to the prime motion or for restriction of rotational joint motion. Therefore, this study proposes a new method to estimate muscle forces considering the role of the antagonistic muscle during 3-dimensional motion. In this model, it is assumed that the agonist muscle is connected to the antagonistic muscle by a coupled spring. Joint torque is defined as the summation of the torques derived from the agonist muscles and the torques derived from the antagonistic muscles. Each muscle force can be estimated to keep balance among the torques generated by the agonist muscles and the antagonistic muscles respectively. The experiments were conducted to validate the proposed method to estimate muscle forces. Surface electromyograms (sEMG) were measured to compare with the estimated muscle forces. The experimental results showed that the estimated muscle forces had a good agreement with the sEMG of muscles.

Millions of children die from preventable injuries every year around the world. Environmental modification is one of the most effective ways to prevent these fatal injuries. The environment should be modified and products should be designed in ways that will reduce the risk of injury by taking child–environment and child–product interactions into account. However, it is still very difficult even for advanced simulation systems to predict how children interact with products in everyday life situations. In this study, we explored a data-driven method as a promising approach for simulating children’s interaction with products in everyday life situations. We conducted an observational study to collect data on children’s climbing behavior and developed a database on children’s climbing behavior to clarify a climbing configuration space, which enables the prediction and simulation of the possible climbing postures of children.

Opening and closing the mouth is one of the most important biomechanical movements of the human being, being one of the first to be performed even before birth. This movement is accomplished by a set joint called the temporomandibular Joint (TMJ). Their dysfunction causes a number of problems always accompanied by pain, in which much of the world population have disorders in this system, requiring the search for treatment. Their dysfunction causes a number of problems always accompanied by pain, in which much of the world population have disorders in this system, requiring the search for treatment. In this way, this work contributes to varieties of knowledge, clarification and care of patients suffering from temporomandibular dysfunction (TMD), with its main objective, the construction of a device able to measure and diagnose abnormalities during biomechanical movement of the opening and closing the mouth, through a low cost imaging system and easy to handle.

Background and objective: The wheelhouse on large ships is the place where personnel, equipment, environmental factor and navigation monitoring interaction concentrate. It’s extremely easy to have human factor problems in the work area. The rationality of wheelhouse design scheme needs to be evaluated in design phase by ergonomics simulation. Methods: Firstly, the vitual cabin scene assembly models are built by using the ship 3D design software CADDS5. They’re converted into intermediate format models of VRML and then imported into the simulation software DELMIA. Secondly the virtual crew human bodies are set up by 5%, 50% and 95% percentile according to the male body sizes of national standard GB/T 10000-1988. Thirdly the sailing commander, steersman and telegraph operator are selected to verify according to the typical tasks such as: (1) the sailing commander’s riding comfort, convenience on up and down, lookout visibility; (2) the steersman’s operability and reachability of the wheel, console panel, phone etc., as well as steering fatigue and lookout visibility; (3) the telegraph operator’s operability and reachability of the telegraph, console panel, phone etc. Finally, via setting the crew of frequently-used walking path, the rationality of channel settings in wheelhouse was validated through collision and interference analysis. Results: Aiming at the typical position operators and space layout of wheelhouse on the large ships, the simulated validation was carried out with the human factors design requirements. The analysis and evaluation work includes the visibility and reachability, the working gesture of crew, the collision and interference checking, the rationality and comfort of cabin operation space. The concrete improvement suggestions are put forward to optimize design. Conclusion: The method adopted in this paper can be applied to ergonomics simulation analysis and evaluation for the wheelhouse design on large ships. This study has important significance in solving the relationship between the crew, equipment and cabin space in the ship design phase.

The huge number of workers affected by musculoskeletal diseases and disorders demonstrates that prevention in working environment is an important issue. Large scale retail trade involves lots of workers, with a multiplicity of tasks and activities. The working tasks and environment condition the onset of these diseases. Digital Human Modeling (DHM) can support the design of the working environment, thus to avoid the risks of work-related musculoskeletal diseases and disorders (WRMSD). The aim of this study was to analyze and assess existing furniture by using DHM to evaluate the risk of developing WRMSD among supermarket clerks and cashiers. Two case studies are specifically presented. Assessment was realized in terms of reaching maps comparing different genders and height/weight percentiles. Preliminary findings suggested the use of dedicated guidelines to choose and set-up furniture in these specific applications, underling the variety of issues present in the large-scale retail trade.