Understanding Smart Home Sensor Data for Ageing in Place Through Everyday Household Routines: A Mixed Method Case...

Area of Science:

• Gerontology
• Human-Computer Interaction
• Ubiquitous Computing

Background:

• Smart home technology generates vast amounts of data, posing challenges for real-time monitoring and alert systems in aging-in-place research.
• Interpreting this sensor data effectively is crucial for supporting elderly individuals living independently.

Purpose of the Study:

• To investigate the utility of a routine-based methodology for analyzing smart home sensor data.
• To enhance the understanding of elderly individuals' daily routines through sensor data analysis.

Main Methods:

• Employed maximum variation sampling to select three elderly participants for a mixed-methods study.
• Collected 180 days of power usage and motion sensor data, supplemented with longitudinal interview data.
• Analyzed daily routines by comparing sensor data with self-reported activities.

Main Results:

• Sensor data demonstrated a strong correlation with participants' self-reported daily routines.
• Changes in routine, indicative of illness, recovery, or sleep quality decline, were identified by comparing movement data with established routines.
• External factors such as temperature and daylight influenced routines, as evidenced by both sensor and interview data.

Conclusions:

• A routine-based approach offers a clear, intuitive, and transparent method for interpreting smart home sensor data.
• Acknowledging individual differences in routinization, daily rhythms, and seasonal variations is vital for accurate data interpretation.
• This research validates the routine-based approach, advancing the development of real-time monitoring and effective alerts for smart homes.