Sensor networks in workplaces: Correlating comfort and productivity Steven van der Valk #* , Trina Myers #*± , Ian Atkinson *± , Karl Mohring #* # Discipline of IT, College of Business, Law and Governance, James Cook University * eResearch Centre, James Cook University ± Centre of Tropical Biodiversity and Climate Change, James Cook University (steven.vandervalk, trina.myers, ian.atkinson, karl.mohring)@jcu.edu.au Abstract—Current designs of modern office workplaces focus on health and safety standards but minimally address worker comfort and satisfaction. However, physical and mental wellness indicators directly affect the output of the worker according to existing studies that illustrate the strong relationship between workplace conditions and productivity. Personal and environmental sensed data can inform smarter workplace design and operations to potentially increase productivity and benefit all stakeholders in the modern workplace. This paper introduces a project that will measure environmental variables and personal physical and mental parameters of employees in different office environments to determine correlations with productivity. In- vivo measurements are obtained via a combination of environmental sensor modules and personal monitoring devices that record the interplay between variables affecting comfort and productivity in a workplace. The office spaces included in the study vary in age, heating/cooling management systems and the internal environmental efficiency of the buildings' design. This paper presents the pilot study that obtained bio-metric data from body area network sensors for each participant and environmental data on the state for workplace including thermal, humidity and light measurements. The participants were given self-reflective micro-surveys on productivity during the time of data collection. The preliminary results indicate that despite climate controlled conditions (i.e., a sealed air conditioned building), natural variance and sub-optimal conditions directly affected reported productivity. Keywords—Internet of Things; sensor networks; personal monitoring device; productivity monitoring I. INTRODUCTION The modern day office workplace is where many individuals spend the bulk of their day. A large proportion of the population spends 23 out of 24 hours indoors [1], [2]. Research illustrating the strong relationship between workplace conditions and productivity could inform smarter design, operation and benefit all stakeholders in the modern workplace. Studies have shown the comfort and quality of the indoor environment are closely related to work efficiency, long-term health and well-being [3], [4]. Currently, workplaces adhere to occupational and safety standards to minimise injury, but very few are intended to increase occupant productivity. Workspaces that are design to be conducive to individual comfort and satisfaction could increase worker productivity [5]. Thermal comfort, personal well-being, workplace satisfaction and output determinants are elements to explore in occupant productivity [6-8]. Sensing technologies have developed rapidly in the past decade and has made many potential areas of study possible. Sensors can be used to monitor office environmental conditions such as temperature, air pressure, sound, light and humidity. Whereas, Body Area Network (BAN) sensors like ‘smart’ watches and Personal Monitoring Devices (PMDs) can be deployed to gauge biometric data such as heart rate, activity levels, sleep patterns and skin perspiration. This paper describes a project that aims to measure and correlate the personal and thermal comfort, workplace satisfaction and productivity determinants and their inter- relationships. In-vivo measurements will be obtained via a combination of environmental sensor modules and PMDs that record the interplay between variables affecting comfort and productivity in a workplace. Further, qualitative data will be gathered via mobile and Web application micro-surveys to gauge self-reported productivity from the participants at two hour intervals. The data will be gathered from a sample of the population from a tropical region that cover age and gender demographics as well as varying building specifications (age, Heating Ventilation and Air Conditioning (HVAC) management, etc.). Correlations in the data can then be used to show relationships between variables and examine the determining factors on workplace productivity. A preliminary pilot study has been conducted that monitors a sample of two participants to test the sensor networks and the results are presented in this paper. During the pilot, initial data was gathered to explore early correlations from the subset of participants and is presented in this paper. The early findings show workers are consciously and physiologically affected by environmental variations. The full study will involve a sample of eight participants. This paper is structured as follows. Section 2 provides background to the relevant domains of productivity and thermal comfort. Section 3 explains the methodology of the study. Section 4 describes the sensor network components and architecture in the context of a work place environment. Section 5 presents the preliminary results from the pilot study and section 6 gives some concluding remarks. II. BACKGROUND Research has shown that stress can have a negative effect on physiological[9], [10], psychological [11], behavioural [9] and productivity outcomes [5]. The physical environment is a