Workplace Stress

Sponsored by NSF

Workplace stress has been identified as the health epidemic of the 21st century. Among office workers, stress is one of the most common reasons for missing work (absenteeism) and a leading cause of underperformance while at work (presenteeism). Moreover, in the U.S., there are 81 million office workers who spend 75% or more of their day working at a desk. Unfortunately, increased daily sitting time is linked to significant health conditions, including cardiovascular diseases and diabetes. Multiple other health-related issues arise in these workers due to poor ergonomic habits. Health conditions are further exacerbated by building-level control of environmental conditions, most specifically lighting and temperature.

Advancing sensing technologies and computational techniques, including machine learning and artificial intelligence, provide opportunities for us to evaluate individualized mechanisms of these health and well-being concerns, monitor changes and impacts over time, and provide precise just-in-time notifications or automated adjustments. Our transdisciplinary team which includes occupational scientists, psychologists, civil and environmental engineers, electrical engineers, and computer scientists is working on various projects focused on leveraging these technologies to promote positive worker health, well-being, and performance within the workplace.

Mapping Positive and Negtive Workplace Stress

Through our work, we aim to generate new analytic models to uncover and map the patterns and pathways that influence work-related stress to understand the primary contributing factors to stress across space and time. The project will develop methods for integrating different data types from the physical and social environment (e.g., temperature, lighting, conversational tones), physiology (e.g., heart rate data, electrodermal activity, movement), and personal experiences (e.g., ecologic momentary assessment) to identify patterns that inform personalized solutions for improving self-awareness and managing work-related health and well-being. We will develop individually contextualized understandings of stress among office workers using machine learning methods that incorporate heterogeneous and noisy multimodal data streams at multiple temporal resolutions while enabling the unsupervised discovery of behavioral routines.