Making Energy Savings Easier: Usability Metrics for Thermostats
Daniel Perry, Cecilia Aragon, Alan Meier, Therese Peffer, and Marco Pritoni
Journal of Usability Studies, Volume 6, Issue 4, August 2011, pp. 226 - 244
Abstract
U.S. residential thermostats control approximately 9% of the nation’s energy use. Many building codes now require programmable thermostats (PTs) because of their assumed energy savings. However, several recent field studies have shown no significant savings or even higher energy use in households using PTs compared to those using non-PTs. These studies point to usability problems that lead to incorrect use and wasted energy. However, the lack of clear, consistent metrics has hampered the acceptance of usability concerns by thermostat manufacturers. Thus there is a need for metrics specific to PTs that manufacturers can use to evaluate their products.
In this paper, we report on the results of a usability study conducted on five commercially available PTs and the development of four new metrics suitable for use in evaluating thermostat usability. Our study confirmed usability deficits in the current generation of PTs and showed the metrics are correlated with each other as well as agreeing with the qualitative results of the study.
Practitioner’s Take Away
Our usability study confirmed a number of usability issues with current programmable thermostats (PTs) and identified four novel metrics that can be used to quantitatively evaluate the usability of PTs. Our results could be generalized to embedded devices in other domains where quantitative metrics are desired.
The following are key points from our research:
- PTs and other small appliances often exhibit poor usability.
- Manufacturers of such appliances often prefer quantitative measurements of usability to qualitative evaluations.
- The metrics we devised show reasonable correlations with each other and with the NIST industry standard while having the additional benefit of being normalized for easier interface comparison.
- Our four metrics contain several input options allowing manufacturers to freely choose between them based on convenience.
- Metrics that combine time-on-task with success/failure rates to yield a single usability score may have greater uptake among PT manufacturers.
- User satisfaction ratings alone are not a complete measure of usability when the device does not have sufficient and direct feedback on task success.
- Usability evaluations are recommended for future PT designs to realize energy savings.
Article Contents
- Introduction
- Methods
- Description of the Metrics
- Results
- Discussion
- Future Research Opportunities
- Conclusion
- References
Download Article
Making Energy Savings Easier: Usability Metrics for Thermostats