Workload
Novel mental workload scale application using fuzzy sets theory
| Document
|
Author Hanna J. Neroj & Jim Nixon |
| Abstract A new scale measuring pilot subjective mental workload was developed as part of the Horizon 2020, Future Sky Safety Project: The validity of new measures needs to be established for effective and reliable implementation. This article investigates the validity of this new scale through assessment of psychological distances between linguistic variables on the scale and subsequent transformation from discrete to continuous scale using fuzzy sets theory. Although the new scale has claimed ordinal properties for the items, no evidence exists to support the interval properties of the scale and any subsequent analysis using statistical methods. The results in this article show linear progression of the items on the scale, supporting the order of the items. To establish the interval properties of the scale, the distances between the ordered items were evaluated using fuzzy sets theory using the Fuzzy Logic Designer in MATLAB. Transformation to continuous scale using fuzzy sets may allow the capture of the dispersion of the data around specific points, showing the degree to which each score belongs to different linguistic category of mental workload. Finally, a proposal for incorporating the validated performance measure was presented. |
Technical and collaborative work in the management of acute kidney injury
| Document
|
Author Denham PHIPPS, Rebecca MORRIS, Thomas BLAKEMAN and Darren ASHCROFT |
| Abstract Patient care may become complicated by acute kidney injury (AKI), a syndrome that affects a patient’s renal functioning. Our study aimed to explore the work involved in dealing with clinical situations where AKI may be present in primary or secondary care. From interviews with 54 doctors and pharmacists in England, we describe their work under three themes: the clinical context; the organisational context; and meeting challenges arising from these contexts. Our findings reflect the role of cognitive work, in particular decision making and collaboration, in facilitating clinical tasks. These should be the focus of any interventions to improve AKI management. |
The quantity and nature of in-vehicle cognitive demands experienced by real-world drivers
| Document
|
Author Madeleine CONAGHAN, Edward ELTON, Periklis CHARCHALAKIS, and Elias STIPIDIS |
| Abstract A real-world driving study was conducted into the cognitive demands within the cabin of a modern domestic car. The quantity and nature of the interactions were recorded, along with dashboard and centre console glances. Ethnographic data was collected from a sample of 8 drivers using remote video analysis and a journey diary. The results suggest that glancing at the dashboard is the highest singularly demanding task, and the highest cognitive demands occur when several types of visuospatial sketchpad representational information are presented to the driver. Therefore, the type of information presented may be more demanding than the area it comes from. |
Staffing models in ergonomics/human factors
| Document
|
Author Colin G. Drury and Catherine Drury Barnes |
| Abstract The level of staffing in an organisation is an important determinant of both the organisation’s effectiveness and of the workforce wellbeing. To predict staffing we need to know the work content of the enterprise. This was comparatively simple when the main work content was repetitive tasks, with total output measured in numbers of identical items produced per unit time. This is not the situation with most modern work, where more complex tasks performed by teams are more common, for example nursing or maintenance. Unfortunately staffing is not well-represented in the ergonomics/human factors (subsequently referred to as human factors) literature, for example in an Ergonomics Abstracts search. More literature is in trade journals, and in the industrial/organisational psychology literature. Specific staffing models have been developed for specific situations, and later applied to analogous enterprises. There are two types of model, depending on how the data are collected and analysed. Models of existing staffing levels use staffing data from a number of enterprises to predict the staffing needs of other enterprises with analogous tasks. For example, in nursing there is much data on existing staffing levels across sites in a health service, such as the UK’s National Health Service. We can examine the staffing levels and concomitant input and process variables to determine how input variables correlate with overall nursing staff levels. The alternative method is summation of task times: if we know the time for each task, then we can sum across all of the tasks in a job to find the total workload and hence the number of staff to meet that workload. The estimates of time per task can come from historical data such as timesheets and work order records, from estimates by subject matter experts or from one of the various time study techniques. Staffing models are needed in human factors, so we need to access literature outside the immediate profession. |
Highly Automated Platooning: Effects on Mental Workload, Stress, and Fatigue
| Document
|
Author Daniël D. HEIKOOP, Joost C. F. DE WINTER, Bart van AREM and Neville A. STANTON |
| Abstract Automatically driving platoons of vehicles are a likely candidate for solving many existing issues of road safety and congestion. However, the psychological effects of such technology are yet to be understood. Therefore, by means of a driving simulator experiment, we aimed to assess the psychological effects of driving in a highly automated platoon. The results showed that the type of task had no substantial effect on heart rate and self-reported stress, fatigue, and workload. However, time-on-task substantially reduced participants’ heart rate. |
Driver interaction with a semi-autonomous vehicle on the road
| Document
|
Author Jisun Kim, Kirsten Revell, Pat Langdon, Simon Thompson, Lee Skrypchuk, Jim O-Donoghue, Joy Richardson, Jediah Clark and Neville A. Stanton |
| Abstract The current Society of Automotive Engineers (SAE) Level 2 (automated acceleration and steering capabilities) autonomous vehicles on the market require constant monitoring and intermittent interventions by the driver whenever the system reaches its limitations. Although drivers’ workload could be increased in the processes, interactions associated with them have rarely been explored in a naturalistic environment. Therefore, our understanding about drivers’ interaction with the current generation of autonomous vehicles seems limited. In order to extend the relevant knowledge, this paper investigates drivers’ interaction with a semi-autonomous vehicle in a complex naturalistic urban environment. Five drivers with varying levels of experience in autonomous cars participated in the experiment. Findings show that active monitoring was required in automated driving for safety to gain sufficient situation awareness that preceded manual interventions in certain conditions. The process seemed to be smoother for the experienced drivers as they had a better understanding about capabilities and limitations of the autonomy. Insights to automotive manufacturers on how driver-vehicle interaction could be enhanced are discussed. |
Quantifying mental workload in performance driving: The motor racing load index (MRLIN)
| Document
|
Author James W.H. Brown, Kirsten M. Revell and Neville A. Stanton |
| Abstract Within the upper echelons of motorsport, it is usual for teams to spend tens or even hundreds of millions of pounds developing aspects of their cars to achieve best performance. Despite this, costly, and potentially dangerous errors occasionally occur due to the complexity of steering wheel-based interfaces. Investigations into improving usability of steering wheel-based controls should employ a range of metrics, one important aspect being the measurement of mental workload. The NASA task load index (NASA-TLX) represents a well-validated method that has been applied to many domains, a derivative of which, called the driving activity load index (DALI) was developed to focus solely on the driving task. However, motorsport differs considerably from conventional road driving – the competitive nature of the task, combined with high levels of physical and emotional stresses indicates that the domain may benefit from a specific derivative of the NASA-TLX. This paper examines the motorsport domain with respect to the NASA-TLX Index and DALI to identify suitable adaptions that may improve the suitability and usefulness of the workload evaluation. A motor racing load index (MRLIN) might provide a more domain-focussed description of racing drivers’ mental workloads, and therefore provide insight into potential areas of improvement, not just in terms of steering wheel-based control usability, but also other aspects of car design and team function. |
Effects of work flexibilisation: The role of interruptions and unfinished work tasks
| Document
|
Author Christian Korunka and Edo Meyer |
| Abstract The organisation and management of work is constantly changing. The growing impact of information and communication technology (ICT) and the growing importance of knowledge-based services have enabled an increase in individual flexibility at work. Core elements of flexible work are flexible time schedules, with or without core times, flexibilisation in work locations, ranging from conventional telework to highly flexible mobile work, and flexibility in work organisation. Many studies examined the effects of these different aspects of work flexibility on quality of working life indicators, showing a wide range of – often contradictory – results. Some studies found mostly positive effects of flexibility at work, for instance on work-life balance or job satisfaction, whereas other studies reported work intensification and/or increases in employees’ strain due to work flexibility. Paradoxes, such as the telecommuting paradox, the connectivity paradox and the autonomy paradox are sometimes used to explain these results. These paradoxes describe phenomena which unexpectedly contradict what is generally expected in a certain situation or which lead to a contradiction (for example, positive and negative effects at the same time). A similar paradox is the relation between flexibility at work and work interruptions. On the one hand, flexibility at work may reduce work interruptions by the possibility to select undisturbed work locations and by using time slots uninterrupted by meetings and coworkers. On the other hand, flexible work organisation may increase work interruptions, because of increased ICT use and increases in communication efforts. Work interruptions are important work stressors, which may result in unfinished work tasks and finally in reduced detachment from work and decreases in wellbeing. |
D-MOD Dynamic Modelling of Operator Demand – A new simulator module for the evaluation of signaller’s demand
| Document
|
Author Lise DELAMARE, David GOLIGHTLY, Andy LUMBY, Peter TREBLE |
| Abstract Estimating demand in signalling operations can inform both decisions about current operational practice as well as future technology. The ideal situation is to have a flexible, automated tool that can estimate demand parameters based on models of timetable and infrastructure to capture dynamic (as opposed to static or averaged) characteristics of a workstation. The Dynamic Modelling of Operator Demand (D-MOD) project has developed a new workstation evaluation tool by drawing on existing simulation models of traffic, timetable and infrastructure. This paper presents the first phase of work to dynamically model parameters from an existing paper based tool showing savings of time and effort, and highlighting directions for future dynamic demand modelling. |
Effect of mental games as cognitive ergonomic intervention on subjective workload and objective productivity of call centre employees
| Document
|
Author Silvia Ahmed Khattak, Laurence Clift and Michael Fray |
| Abstract Call centre work is tedious and stressful. The nature of call centre work lacks the flexibility to bring variety of tasks in a typical workflow of an employee’s work shift. The stringent time limitations on the call monitoring key performance indicators makes work challenging for employees. A previous study which highlighted the ergonomic problems that call centre employees face, led to further investigation in this area. This study was an intervention study which rose from the interest in finding ways to make the work of call centre employees more enjoyable to enhance employee wellbeing and efficiency. This study attempted to alter the design of the workflow of the employees by placing mental games (such as crosswords, jigsaw puzzles, hangman and board games) in the break areas of employees. Employees were encouraged to play the games in their breaks. Because this study was based on mental games as an intervention (not physical), and involved employee’s perceptions, hence its association with cognitive ergonomics. The game playing was kept within the entitled breaks, avoiding other times, to bring minimum disruption to the normal workflow. The aim of the study was to mentally engage employees in mental games rather than work related thoughts. It was assumed that this kind of short but enjoyable disengagement might reset the mind of the employee and enhance wellbeing when they reassume their work after the break. This wellbeing in turn might affect outcomes such as objective productivity and subjective workload. |
Exploring workload and performance through the use of visual analytics
| Document
|
Author Joanne Kitchin & Chris Baber |
| Abstract In Visual Analytics, the output of automated analysis is presented to users in an interactive visualisation. By responding to this, the user can modify the parameters of the computer visualisation. This raises questions about the design of the visualisation and the appropriate level of interaction for users. This paper focuses on the impact of visualisation on user performance. A simple air target detection task (in which automated support identified possible threat aircraft) was combined with a secondary task (in which target letters had to be detected against a background). Four visual analytic displays were used to complete a target detection task over two studies. The first study explored how the displays affected workload, attentional demand and performance, and the second how workload, attentional demand and performance are affected by task load (using the same displays). Results show that the use of visual analytic displays maintains response time and primary task performance when task load increases. This suggests that the demand on attention is easier to manage when visual analytic displays are used. |
Examining the effect of reduced crew sizes on submarine command team workload
| Document
|
Author Sophie G. Hart, Kiome A. Pope, Victoria A. Banks, Aaron P. J. Roberts & Neville A. Stanton |
| Abstract The Command Team Experimental Test-Bed (ComTET) is a body of work that aims to understand the current functionality of submarine command teams and future ways of working. To facilitate their research, the ComTET team built a submarine control room simulator for testing purposes. Previous research from the ComTET research program ran baseline testing to understand current functionalities in the control room. This generated a number of recommendations that informed the design of a new control room configuration, whereby the crew size was reduced by removing one Sonar Operator and one Target Motion Analyst Operator. This study tested 70 participants (ten teams of seven individuals), all of which received general maritime and role-specific training using the ComTET tutorial package. Each team was required to complete high and low demand versions of three scenario types: Return to Periscope Depth, Inshore Operations, and Dived Tracking. Following the completion of each scenario, the participants’ subjective workload was self-reported using an electronic version of the Bedford Workload scale. The results provided in this paper show a preliminary analysis of workload scores from a subset of four teams. Teams were taken from both the baseline and reduced crew size configuration. Results suggested that the subjective workload of operators was affected by scenario demand and type. Additionally, the results indicate that the reduction of crew size increased the workload placed on operators in the command team. Therefore, submarine command teams with reduced crew sizes are more likely to be faced with higher workload levels during submarine operations. |
Task Switching – Managing Workload within Digital AFV Systems
| Document
|
Author Trevor Dobbins, Ryan Meeks, Dan Evans, Stuart Howe & Stephen Barrett |
| Abstract The Armoured Fighting Vehicle Commander’s role is characterised by having multiple mission critical tasks. They are required to rapidly redirect their attention at short notice as events change. This paper describes how this task-switching is modelled and analysed, within the system model, to manage workload and develop/deliver a useable system |
Assessing the Benefits of Virtual Reality Training for Manual Assembly
| Document
|
Author Charlotte Temmink & Yee Mey Goh |
| Abstract The aim of this study is to assess, through experimentation, the effectiveness of Virtual Reality training when compared to a traditional written instructions in a manufacturing assembly process. The effectiveness was evaluated in terms of perceived workload and task performance, with participants asked to self-report on the experience during both the written and VR training. Findings indicate that VR training reduced the user’s workload perception and errors made during the task, however there was no significant impact on time taken to complete the task. |
EEG-based Mental Workload Level Estimation using Deep Learning Models
| Document
|
Author Kunjira Kingphai & Yashar Moshfeghi |
| Abstract Deep learning-based approaches have recently received much attention and managed to accurately capture variance characteristics in the Electroencephalography (EEG) signals. In this paper, we aim to classify the subject’s mental workload (MWL) level from EEG signal by using deep learning models named Stacked Gated Recurrent Unit (GRU), Bidirectional GRU (BGRU), BGRU-GRU, Stacked Long-Short Term Memory (LSTM), Bidirectional LSTM (BLSTM), BLSTM-LSTM and Convolutional Neural Network (CNN). The classification was performed on a publicly available mental workload dataset, STEW. Our encouraging results show the potential of deep learning models for MWL level detection. |
An assessment of workload in a simulated submarine control room
| Document
|
Author Kiome A. Pope, Aaron P.J. Roberts & Neville A. Stanton |
| Abstract The Command Team Experimental Testbed (ComTET) is a programme of work designed to study the operation of current and future submarine command teams. As part of this work a submarine control room simulator was designed and built for testing purposes. Baseline testing was conducted to establish current functionalities, with recommendations forming the basis for Manipulation One where some operators were co-located. During testing, 32 participants (four teams of eight individuals) were given general maritime, and role specific training. Each team completed a high and low demand version of three scenario types: return to periscope depth, inshore operations, and dived tracking. On completing each scenario, the workload of participants was assessed using an electronic version of the Bedford scale. Preliminary results suggest the workload of operators was affected by scenario demand and type. Results also suggest that the co-location of operators had a positive effect on the demand placed on them. The results are discussed, along with future analysis plans. |
Assessing pilots’ mental workload using touchscreen inceptor for future flight deck design
| Document
|
Author Joao Paulo Macedo, Kyle Hu, Rani Quiram & Samarth Vilas Burande |
| Abstract Touchscreen displays are one of the pillars of future flight deck design and it is foreseen that at some point traditional flight control inceptors will be modified to a touchscreen version. However, this transition can only be safe and successful with due regard for human performance implications. This study addresses it by comparing pilots’ mental workload for a traditional sidestick and an innovative touchscreen control inceptor. The results indicate that the new technology increases pilot workload, suggesting that further development is required to use it in future flight decks. |
Pilots’ Scan-path and Attention Shifts during Tactical Manoeuvres
| Document
|
Author Wen-Chin LI, Chen-Kai HSU, Graham BRAITHWAITE and Matthew GREAVES |
| Abstract Understanding the cognitive processes of pilots’ fixation shifts and attention distributions can facilitate training of tactical maneuvers and improve both performance and flight safety. Thirty-five military pilots participated in the current study. By analyzing eye movement data, it is possible to identify what the military pilot was looking at during particular moments in the flight. A pilot’s attention is likely to remain one of the critical and limited resources that will need to be confronted in the development of new technology. It is important that the nature of pilots’ attention and fixation distributions is understood in order to assess and manage the implications of new technology in the cockpit. |
Adaptations to everyday work amongst nuclear operators: A safety-II approach
| Document
|
Author Robin Hamer, Gyuchan Thomas Jun and Patrick Waterson |
| Abstract The safety-II movement has recently been gaining momentum. Most of the work has been conceptual in nature; however, some efforts have been made to apply Safety-II thinking in a practical manner. Current safety-II based methods have been criticised as being difficult to administer and complicated to understand. Although some research has been carried out to address safety-II, there is still some ambiguity about how the construct can be applied in practice. One way forward is to understand adaptations that occur in normal everyday work. This can be achieved by contrasting work-as-imagined and work-as-done in the context of the work environment. One understanding of this is to capture and analyse such adaptations and demonstrate value by sharing good practice. The current case study was undertaken at a major British nuclear installation. The research aims were to understand and learn from normal every day work done by nuclear operators, more specifically the adaptations they make to successfully navigate their work. A mixed methods study design (document analysis, contextual inquiry and interviews) was selected to obtain a complete understanding of work-as-imagined and work-as-done from two selected tasks, common in a nuclear power plant. Semi-structured interviews were administered to nuclear personnel who performed the tasks. These were subsequently transcribed and coded to extract the adaptations. 47 adaptations were elicited and interpreted using an analysis framework, adapted for use in the nuclear industry. The framework enabled adaptations to be classified and provided new insights into how and why the adaptations occurred. The findings were reviewed by human factors practitioners at a major UK nuclear operator. The consensus was positive, and a potential benefit of application was recognised – in particular the new information on adaptations to review tasks and rewrite standard operating procedures. |
Performance and workload using an audible intelligent assistant during pilot training
| Document
|
Author David A. Hudson & Michael A. Bromfield |
| Abstract Boeing forecast that globally an additional 602,000 pilots will be needed by 2041 to meet year on year growth of 3.6% in passenger traffic. All pilots need to be trained in accordance with accepted regulatory standards. New technology is continuously being developed to enhance training and reduce training time. Research into the effectiveness of training technologies and how these impact pilot performance and workload is key to future growth. One such technology under review at the University of Birmingham is the ‘audible intelligent assistant’. This artificial voice that provides real-time feedback to the pilot during flight training. The system provides warnings, cautions and instruction to the human pilot to enhance pilot training. Preliminary results of the research showed an improvement in pilot performance against a specified set of target parameters of airspeed and altitude and a corresponding decrease in workload for 80% of pilots when using the audible intelligent assistant. Without AIA, pilot performance improved by 35% due to learning effects alone, with AIA pilot performance showed an improvement of 65%. |
Driver Perception Using A Camera-Based Digital Side Mirror: An On-Road Study
| Document
|
Author Anastasia Schkreba, Sanna Pampel, Gary Burnett & David R. Large |
| Abstract Camera-based digital mirrors purport to offer a range of benefits yet may influence drivers’ ability to quickly and accurately extract salient information pertaining to driving. In an on-road study, fifteen experienced drivers (seated in the front, left, passenger seat) undertook an orientation-discrimination task requiring the extraction of real-world information using either a digital mirror (placed internally) or a conventional, external, reflective mirror. Participants were asked to complete each task as quickly and accurately as possible, and then return their attention to the driving scene, as if they were driving. Although there was no difference in performance accuracy or reported workload between conditions, participants responded sooner when using the digital mirror – suspected to be due to the wider field-of-view intrinsic to its design, although participants also reported feeling “rushed” when using the digital mirror. The majority of participants (9 out of 15) indicated a preference for the conventional mirror, raising numerous concerns associated with the digital mirror, relating to image quality, field-of-view, focal depth (particularly for wearers of varifocal or reading glasses), and potential deleterious effects of ambient weather conditions, demonstrating important human factors issues still requiring attention in this context. |
