In the UK, outdoor air pollution levels receive a great deal of press coverage and research funding. Yet the average person spends 90% of their time indoors. Inadequate ventilation, characterised by high levels of CO2, has been linked to lower cognitive performance, lower exam results and mild health effects such as sore throat and headaches. Being a product of human respiration, CO2 is a very useful indicator of ventilation in a room. Critically, literature suggests that cognitive decline precedes awareness of poor indoor air quality, as occupants become acclimatised to their conditions over time. Over half of all UK building stock is naturally ventilated (i.e. by opening windows and doors), and given expectations of warmth and comfort which correlate to keeping buildings shut tight, I am of the opinion that there are huge productivity losses every winter due to poor ventilation and a lack of awareness of the effects of poor indoor air quality.
A large part of my recent work has been raising awareness of indoor air quality among office workers and how to best present this information to occupants. I have had the pleasure of working with a 3rd year undergraduate Electronic Engineering student called Mike who used participatory design techniques and hardware prototyping to design an ambient air quality monitor for naturally ventilated offices/classrooms. The “Aether” uses a Rasberry Pi Zero to drive sensors for CO2, temperature and humidity and display the values on a LCD. Ambient lighting curtousy of a WS2812B ring changes with increasing CO2 levels from green to orange to red. The casing is an original design which has been 3D printed.
The eventual prototype was based on the conversations and initial paper prototypes which were shown to friends and family. The monitor was purposefully designed to be easily glance-able, with the LED lighting and screen background changing from green to yellow to red with increasing levels of CO2. The reading for CO2 is front and centre with humidity and temperature smaller numbers. Barometric pressure was found to be disliked or misunderstood, so this reading was removed from the display, replaced by a digital clock. The design incorporates a Raspberry Pi Zero sending data to the WIFI dongle and driving the sensors and LCD display, while outsourcing the LED control to an Arduino Pro Mini. All were contained within a purpose-design 3D printed ventilated enclosure. Data is sent via wifi to a server, or if a connection cannot be established, it logs the data to an inbuilt SD card.
The two produced Aethers have been deployed in 8 offices over the recent winter months, with questionnaires and qualitative interviews following each deployment. Preliminary results suggest the devices have potential to change behaviour towards healthier ventilating behaviours. The devices challenged people’s existing habits around keeping comfort were challenged: “When you think this time of year, you think, I want my office to be warm, I never think I want to open the window”. Participants appreciated the fact that the system was not overtly persuasive, in that it was not telling them what to do. “If I was in between things and I got out of my chair and I saw that it was getting amber I would do something. It was my decision, it gave me information to act on but it was my decision”
This study has been a useful proof-of-concept for cheaply built IoT sensors for measuring indoor air quality. It has produced some encouraging results which suggest that by visualising CO2 in an aesthetically pleasing and unobtrusive fashion, we may be able to foster healthier behaviours around ventilation in offices. Hardware prototyping has been key to the design process. Rather than jumping straight to production, we ran initial interviews related to information placement and aesthetic design of information displayed on the LCD display. The system is flexible and adaptable and parameters be easily configured in the code to change the brightness or colour of the LED lights and later be adapted to send notifications to users regarding unhealthy indoor air quality rather than relying on people noticing the device. An important part of the appeal of the device so far has been its unobtrusiveness and minimal human-engagement required to retain its usefulness. Future work involves further user-testing over a broader range of offices and a larger sample size, in order to determine the likely effects of the system and further interrogate preferences for long-term office-situated monitoring technology such as this. We also plan to develop of a number of networked sensors which feed information back to a central screen, to better accommodate larger offices and allow information to be viewed, captured and analysed remotely (i.e. on the cloud).
My eventual aim is much more than sensors alone. I (somewhat optimistically) envision CO2 monitors to be part of every home, classroom and office, with occupants having the knowledge to interpret the values and be prompted to open a window or a door if they see the CO2 to be creeping towards unhealthy thresholds (i.e. above 1,000ppm). The first step in this process for our work however, is to focus on hardware prototyping to ensure our final designs are as informative, useful as possible, whilst remaining unobtrusive over their journey towards becoming an integral part of the office furniture.