We have different sensors; humidity, CO2, PIR and light. Combining these and other data sources such as room bookings, weather, etc. We think that there is a possibility to use all this data together to draw insights of our office both in how it is used and if there are anomalies in the data that could indicate possible faults in installation or equipment.
One way to reach this goal is through data analysis and anomaly detection.
Understand the data domain and what it represents and possibly propose more data(sensors etc.) if needed.
Define what patterns to recognize
Can certain patterns indicate broken equipment in meeting rooms
What do the CO2 pattern correlated with other sensors in a specific room represent.
When should a booked room be identified as not used.
Investigate methods/algorithms or implementations that recognize patterns in the data.
Recognize new/unknown patterns
Make a prototype for demonstration.
Propose architecture, components to use etc.
Understanding of data analytics and pattern recognition algorithms.
Fluent in one of the following programming languages: Python etc.
Experience of project planning (i.e. ability to plan and work according to the plan).
This thesis work aims at students in Applied Mathematics and statistics , Data science, Electrical engineering, Computer science, Computer engineering or similar. It's suitable for 2 students