Email stress: what are its causes? how is it measured? can it be solved? The literature review revealed that, despite the term being well used and recognised, discussions surrounding the root cause of email stress had reached little consensus and the concept was not well understood. By its very nature, email stress theory had fallen victim to the academic debate between psychological vs. physiological interpretations of stress which, as a result of either choice, limited more progressive research. Likewise an array of email management strategies had been identified however, whilst some generated quick successes, they appeared to suffer longevity issues and were not maintained a few months after implementation in the workplace. The purpose of this research was to determine whether email communication causes employees psychological and physiological stress and investigate the impact of email management strategies in the workplace. A pragmatic philosophy placed the research problem as central and valued the differences between paradigms to promote a mixed-method approach to research. The decision to pair both case studies and action research methods ensured a framework for presenting results and an actionable solution was achieved. In direct response to the research aims an original email stress measuring methodology was devised that combined various data collection tools to measure and investigate email stress. This research design was applied and evaluated 'email free time' and email filing. Results of the study showed an increased stress response to occur during email use, i.e. caused employees' increased blood pressure, heart rate, cortisol and perceived stress, and a number of adverse effects such as managing staff via email, social detachment, blame and cover-your-back culture were identified. Findings revealed 'email free time' was not a desirable strategy to manage email stress and related stressors, whereas email filing was found more beneficial to workers well-being. Consolidation of the data gathered from the literature review and research findings were used to develop an initial conceptualisation of email stress in the form of two models, i.e. explanatory and action. A focus group was conducted to validate the proposed models and a further investigation at the ? was carried out to critique the use of an email training intervention. The results showed some improvements to employees' behaviour after the training, e.g. improved writing style, email checked on fewer occasions each day and fewer sufferers of email addiction. The initial models devised, alongside the latter findings, were synthesised to create a single integrative multidimensional model of email stress and management strategies. The model made an original contribution to knowledge in terms of theory, i.e. to conceptualise email stress, and practice, i.e. to offer practical solutions to the email worker.

Stress and overload are strongly associated with unsafe behaviour, which motivated various studies to detect them automatically in workplaces. This study aims to advance safety research by developing a data-driven stress and overload detection method. An unsupervised deep learning-based anomaly detection method is developed to detect stress. The proposed method performs with convolutional neural network encoder-decoder and long short-term memory equipped with an attention layer. Data from a field experiment with 18 participants was used to train and test the developed method. The field experiment was designed to include a pre-defined sequence of activities triggering mental and physical stress, while a wristband biosensor was used to collect physiological signals. The collected contextual and physiological data were pre-processed and then resampled into correlation matrices of 14 features. Correlation matrices are used as an input to the unsupervised Deep Learning (DL) based anomaly detection method. The developed method is validated, offering accuracy and F-measures close to 0.98. The technique employed captures the input data attributes correlation, promoting higher interpretability of the DL method for easier comprehension. Over-reliance on uncertain absolute truth, the need for a high number of training samples, and the requirement of a threshold for detecting anomalies are identified as shortcomings of the proposed method. To overcome these shortcomings, an Adaptive Neuro-Fuzzy Inference System (ANFIS) was designed and developed. While the ANFIS method did not improve the overall accuracy, it outperformed the DL-based method in detecting anomalies precisely. The overall performance of the ANFIS method is better than the DL-based method for the anomalous class, and the method results in lower false alarms. However, the DL-based method is suitable for circumstances where false alarms are tolerated. / Dissertation / Doctor of Philosophy (PhD)