Today’s assembly operations represent about 15-70% of all manufacturing time and about 40% of all manufacturing costs, and manual assembly processes are still a significant portion of today’s assembly operations. Furthermore, today’s manufacturing environment requires a well-trained and flexible workforce that can easily adapt to changing products and processes. Unfortunately, manufacturing training is often performed using the master-apprentice model in the assembly line resulting in unsafe and expensive training conditions as this model is a slow and expensive process. Previous research has considered the use of virtual environments (VEs) for training purposes in different fields such as aviation, driving, construction, medicine, and manufacturing among many others. However, to this date, no assembly studies have been successful in providing a positive transfer of knowledge between virtual environments and real environments. On the other hand, several eye-tracking studies in radiology, air-traffic control, driving, and reading show that participants with higher levels of experience have different eye-scan patterns than participants with lower levels of experience. However, it is unknown how visual scans are affected by practice. Furthermore, several empirical visuomotor studies of task-oriented processes in real environments show that observers fixated their eyes on the areas that are crucial to the required task. However, we do not know the necessary visual elements to observe when performing and when learning how to perform an assembly task, nor the effects of following visual instructions and having visual distractors during this process. Finally, we have yet to establish what observation differences may exist between real and virtual environments with regards to these unknowns. This work presents the results of an assembly task which required participants to follow visual instructions and to select assembly objects among similar distractors. This assembly task was performed for ten cycles in real and virtual environments, and we used an eye-tracking device to register participants’ visual scans. We successfully identified the areas that are needed to observe for an assembly task in both environments and the effect of visual instructions and distractors in a visual scan. We found statistically significant differences for visual scans by assembly cycle and environment, with a p-value of <0.05. We also identified a connection between learning curves and participant eye scan, showing a significant decrease in the incidence of eye tracking metrics (visit count, visit duration, fixation count and fixation duration) between the first and the tenth cycles (ΔΜ), particularly for visual distractors ranging from 37.36% to 48.77%, and for visual instructions ranging from 35.17% to 54.82%. We found that participants’ observations became more efficient with practice, not only in terms of identifying distractors and following visual instructions but also in terms of developing an ability to observe key visual elements. For the RE we found a positive Pearson correlation between the proportion of fixation duration and assembly cycle for the key visual areas with p-values<0.002 and a negative Pearson correlation between the proportion of fixation duration for the non-key visual areas with p-values<0.046. Similar results were obtained for the VE.

This thesis presents mathematical modeling and control techniques that can be used to predict and specify performance of biologically inspired actuation systems called cellular actuators. Cellular actuators are modular units designed to be connected in bundles in manner similar to human muscle fibers. They are characterized by inherent compliance and large numbers of on-off discrete control inputs. In this thesis, mathematical tools are developed that connect the performance to the physical manifestation of the device. A camera positioner inspired by the human eye is designed to demonstrate how these tools can be used to create an actuator with a useful force-displacement characteristic. Finally, control architectures are presented that use discrete switching inputs to produce smooth motion of these systems despite an innate tendency toward oscillation. These are demonstrated in simulation and experiment.

We investigate, model, and design interaction techniques for hand occlusion with direct pen input. Our focus on occlusion follows from a qualitative and quantitative study of direct pen usability with a conventional graphical user interface (GUI). This study reveals overarching problems relating to poor precision, ergonomics, cognitive differences, limited input, and problems resulting from occlusion. To investigate occlusion more closely, we conduct three formal experiments to examine its area and shape, its affect on performance, and compensatory postures. We find that the shape of the occluded area varies across participants with some common characteristics. Our results provide evidence that occlusion affects target selection performance: especially for continuous tasks or when the goal is initially hidden. We observe how users contort their wrist posture during a simultaneous monitoring task, and show this can increase task time. Based on these investigations, we develop a five parameter geometric model to represent the shape of the occluded area and extend this to a user configurable, real-time version. To evaluate our model, we introduce a novel analytic testing methodology using optimization for geometric fitting and precision-recall statistics for comparison; as well as conducting a user study. To address problems with occlusion, we introduce the notion of occlusion-aware interfaces: techniques which can use our configurable model to track currently occluded regions and then counteract potential problems and/or utilize the occluded area. As a case study, we present the Occlusion-Aware Viewer: an interaction technique which displays otherwise missed previews and status messages in a non-occluded area. Within this thesis we also present a number of methodology contributions for quantitative and qualitative study design, multi-faceted study logging using synchronized video, qualitative analysis, image-based analysis, task visualization, optimization-based analytical testing, and user interface image processing.

This thesis evaluated the effects of two types of training on nurses’ ability to safely and efficiently administer IV medications using a smart infusion pump. A high fidelity simulated nursing unit was created in which nurses recruited from the University Health Network programmed a series of infusions after receiving training. A training script modeled after the pump vendor’s training sessions was created and tested first on 24 nurses. The results were analyzed for deficiencies in safety and efficiency from a human factors and education perspective and a new training script was created and tested on a group of 23 nurses. No significant differences were found between training groups on measures related to safety, but significant differences were found in nurse efficiency and behaviour in some aspects of pump programming. This study sets a precedent for human factors evaluation being used in tandem with existing training practices and lays the groundwork for further exploration on this topic.

This research used human factors methods to investigate the relationship between no fault found (NFF) incident frequency and device usability. NFF reporting occurs when a medical device sent for repair is found to be operating normally. NFF incidents are one of the most recurrent failure modes, and therefore have considerable impact on cost, dependability and safety. An analysis of medical equipment maintenance data was conducted and six devices with a high NFF reporting frequency were identified. Semi-structured interviews and heuristics evaluations revealed that usability issues likely caused many of the NFF incidents. Other factors suspected to contribute to increased NFF reporting include accessory issues, intermittent faults and environmental issues. Finally, in order to validate the results, usability testing was conducted on three of the devices. 23 usability-related design flaws were identified. Therefore devices containing latent usability-related design flaws can be identified through analysis of medical equipment maintenance data.

This research used human factors methods to investigate the relationship between no fault found (NFF) incident frequency and device usability. NFF reporting occurs when a medical device sent for repair is found to be operating normally. NFF incidents are one of the most recurrent failure modes, and therefore have considerable impact on cost, dependability and safety. An analysis of medical equipment maintenance data was conducted and six devices with a high NFF reporting frequency were identified. Semi-structured interviews and heuristics evaluations revealed that usability issues likely caused many of the NFF incidents. Other factors suspected to contribute to increased NFF reporting include accessory issues, intermittent faults and environmental issues. Finally, in order to validate the results, usability testing was conducted on three of the devices. 23 usability-related design flaws were identified. Therefore devices containing latent usability-related design flaws can be identified through analysis of medical equipment maintenance data.

People have difficulties using computers. Some have more difficulties than others. There is a need for guidance in how to evaluate and improve the accessibility of systems for users. Since different users have considerably different accessibility needs, accessibility is a very complex issue.<p>ISO 9241-171 defines accessibility as the "usability of a product, service, environment or facility by people with the widest range of capabilities." While this definition can help manufacturers make their products more accessible to more people, it does not ensure that a given product is accessible to a particular individual.<p>A reference model is presented to act as a theoretical foundation. This Universal Access Reference Model (UARM) focuses on the accessibility of the interaction between users and systems, and provides a mechanism to share knowledge and abilities between users and systems. The UARM also suggests the role assistive technologies (ATs) can play in this interaction. The Common Accessibility Profile (CAP), which is based on the UARM, can be used to describe accessibility.<p>The CAP is a framework for identifying the accessibility issues of individual users with particular systems configurations. It profiles the capabilities of systems and users to communicate. The CAP can also profile environmental interference to this communication and the use of ATs to transform communication abilities. The CAP model can be extended as further general or domain specific requirements are standardized.<p>The CAP provides a model that can be used to structure various specifications in a manner that, in the future, will allow computational combination and comparison of profiles.<p>Recognizing its potential impact, the CAP is now being standardized by the User Interface subcommittee the International Organization for Standardization and the International Electrotechnical Commission.

This thesis evaluated the effects of two types of training on nurses’ ability to safely and efficiently administer IV medications using a smart infusion pump. A high fidelity simulated nursing unit was created in which nurses recruited from the University Health Network programmed a series of infusions after receiving training. A training script modeled after the pump vendor’s training sessions was created and tested first on 24 nurses. The results were analyzed for deficiencies in safety and efficiency from a human factors and education perspective and a new training script was created and tested on a group of 23 nurses. No significant differences were found between training groups on measures related to safety, but significant differences were found in nurse efficiency and behaviour in some aspects of pump programming. This study sets a precedent for human factors evaluation being used in tandem with existing training practices and lays the groundwork for further exploration on this topic.

We investigate, model, and design interaction techniques for hand occlusion with direct pen input. Our focus on occlusion follows from a qualitative and quantitative study of direct pen usability with a conventional graphical user interface (GUI). This study reveals overarching problems relating to poor precision, ergonomics, cognitive differences, limited input, and problems resulting from occlusion. To investigate occlusion more closely, we conduct three formal experiments to examine its area and shape, its affect on performance, and compensatory postures. We find that the shape of the occluded area varies across participants with some common characteristics. Our results provide evidence that occlusion affects target selection performance: especially for continuous tasks or when the goal is initially hidden. We observe how users contort their wrist posture during a simultaneous monitoring task, and show this can increase task time. Based on these investigations, we develop a five parameter geometric model to represent the shape of the occluded area and extend this to a user configurable, real-time version. To evaluate our model, we introduce a novel analytic testing methodology using optimization for geometric fitting and precision-recall statistics for comparison; as well as conducting a user study. To address problems with occlusion, we introduce the notion of occlusion-aware interfaces: techniques which can use our configurable model to track currently occluded regions and then counteract potential problems and/or utilize the occluded area. As a case study, we present the Occlusion-Aware Viewer: an interaction technique which displays otherwise missed previews and status messages in a non-occluded area. Within this thesis we also present a number of methodology contributions for quantitative and qualitative study design, multi-faceted study logging using synchronized video, qualitative analysis, image-based analysis, task visualization, optimization-based analytical testing, and user interface image processing.

People have difficulties using computers. Some have more difficulties than others. There is a need for guidance in how to evaluate and improve the accessibility of systems for users. Since different users have considerably different accessibility needs, accessibility is a very complex issue.<p>ISO 9241-171 defines accessibility as the "usability of a product, service, environment or facility by people with the widest range of capabilities." While this definition can help manufacturers make their products more accessible to more people, it does not ensure that a given product is accessible to a particular individual.<p>A reference model is presented to act as a theoretical foundation. This Universal Access Reference Model (UARM) focuses on the accessibility of the interaction between users and systems, and provides a mechanism to share knowledge and abilities between users and systems. The UARM also suggests the role assistive technologies (ATs) can play in this interaction. The Common Accessibility Profile (CAP), which is based on the UARM, can be used to describe accessibility.<p>The CAP is a framework for identifying the accessibility issues of individual users with particular systems configurations. It profiles the capabilities of systems and users to communicate. The CAP can also profile environmental interference to this communication and the use of ATs to transform communication abilities. The CAP model can be extended as further general or domain specific requirements are standardized.<p>The CAP provides a model that can be used to structure various specifications in a manner that, in the future, will allow computational combination and comparison of profiles.<p>Recognizing its potential impact, the CAP is now being standardized by the User Interface subcommittee the International Organization for Standardization and the International Electrotechnical Commission.