Designing the user experience of a spatiotemporal automated home heating system: a holistic design and implementation process

Kruusimagi, Martin (2017) Designing the user experience of a spatiotemporal automated home heating system: a holistic design and implementation process. PhD thesis, University of Nottingham.

PDF (Thesis for reader access - any sensitive & copyright infringing material removed) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (13MB) | Preview


This research explores technological interventions to reduce energy use in the domestic sector, a notable contributor to the global energy footprint. In the UK elevated challenges associated with renovating an outdated, poorly performing housing stock render a search for alternatives to provide immediate energy saving at low cost. To solve this problem, this thesis takes a holistic design approach to designing and implementing a spatiotemporal heating solution, and aims to investigate experiences of comfort, thermal comfort concepts for automated home heating, users’ interactions and experiences of living with such a system in context, and the underlying utility of quasi-autonomous spatiotemporal home heating.

The mixed-methods research process was employed to explore and answer four questions: 1) what is the context within which these home heating interfaces are used, 2) to what extent can spatiotemporal automated heating minimise energy use while providing thermal comfort, 3) how are different heating strategies experienced by users, and 4) How do visibility of feedback, and intelligibility affect the user experience related to understanding and control? Ideation techniques were used to explore the context within which the designs are used with regard to all factors and actors in play and resulted in a conceptual model of the context to be used as a UX design brief. This developed model used mismatches between users’ expectations and reality to indicate potential thermal comfort behaviour actions and mapped the factors within the home context that affected these mismatches. Potential user inclusion through participatory design provided stakeholder insight and interface designs concepts to be developed into prototypes. The results of a prototype probe study using these prototypes showed that intelligibility should not be an interface design goal in itself, but rather fit in with broader UX design agenda regarding data levels, context specificity, and timescales. Increased autonomy in the system was shown not to directly diminish the experience of control, but rather, control or the lack of originated from an alignment of expectations and reality.

A quasi-autonomous spatiotemporal heating system design (including a novel heating control algorithm) was coupled with the design of a smartphone interface and the resultant system was deployed in a low-technology solution demonstrating the potential for academic studies to explore such automated systems in-situ in the intended environment over a long period of time. Assessment of the novel control algorithm in an emulated environment demonstrated its fitness for purpose in reducing the amount of energy required to provide adequate levels of thermal comfort (by a factor of seven compared with EnergyStar recommended settings for programmable thermostats), and that these savings can be increased by including occupants’ thermal preference as a variable in the control algorithm.

Field deployment of that algorithm in a low-tech sensor-based heating system assessed the user experience of the automated heating system and its mobile application-based control interface, as well as demonstrated the user thermal comfort experience of two different heating strategies. The results highlighted the potential to utilise the lower energy-use “minimise discomfort” strategy without compromising user thermal comfort in comparison to a “maximise comfort” strategy. Diverse heating system use behaviours were also identified and conceptualised alongside users’ experiences in line with the developed conceptual model. A rich picture analysis of all previous findings was utilised to provide a model of the design space for home automated heating systems, and was used to draw interface design guidelines for a broader range of home automation control interfaces.

The work presented here served as important first steps in demonstrating the importance of assessing UX of automated home heating systems in situ over elongated periods of time. Novel contributions of (i) conceptual model of automated systems’ domestic context and thermal comfort behaviours within, (ii) nudging this behaviour by selecting a “minimise discomfort” heating strategy over “maximise comfort”, (iii) using UX to influence user expectations and subsequently energy behaviour, and (iv) inclusion of thermal preference in domestic heating control algorithm were all resultant of examining naturally occurring behaviours in their natural setting. As such, they are important exploratory discoveries and require replication, but provide new research directions that would allow reduction of domestic energy use without compromise.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Sharples, Sarah
Robinson, Darren
Keywords: home heating, energy saving, automated heating, user experience
Subjects: T Technology > TH Building construction > TH7005 Heating and ventilation. Air conditioning
Faculties/Schools: UK Campuses > Faculty of Engineering
UK Campuses > Faculty of Engineering > Built Environment
Item ID: 40024
Depositing User: Kruusimagi, Martin
Date Deposited: 13 Jul 2017 04:40
Last Modified: 14 Jul 2017 05:09

Actions (Archive Staff Only)

Edit View Edit View