Solving Rooming & Timetabling Problems With Azure IoT
If you have worked in schools for long enough, you will know there is always a premium placed on being as efficient and smart as possible with available room space. Often, it’s the Head of Mathematics who is tasked with the annual challenge of assigning the enormous number of permutations around organizing all of classes needing to go into a limited number of available classrooms.
With modern learning environments leading to a reduction in closed “single cell” classrooms, finding an available space for teaching or break out areas is increasingly difficult, with both clusters of students and even entire classes being more mobile during the day.
This was a problem faced by Drenthe College in the Netherlands, where students and faculty were finding it difficult to find available space during the day.
Measurements of afternoon usage revealed that, in some buildings, up to 35 percent of the rooms were empty. Paradoxically, students and faculty were still spending up to 15 minutes searching for an available room. Clearly, something was wrong.
The solution relied on the use of 500 bluetooth enabled sensors that could track and monitor:
- movement
- sound
- temperature
- humidity
- CO2 levels
All of these were connected back to the Azure IoT Hub which was further connected to the school’s timetabling records, before using Azure Stream Analytics to provide real time updates on the status of each room. Through the use of a mobile app, faculty and students were able to identify free spaces for teaching and learning, a task that had previously involved 10-15 minutes of wandering around the campus looking for a classroom. Through the application of Azure Machine Learning, Drenthe College were able to add a predictive element providing a level of certainty as to room availability up to 2-3 days in advance.
Prior to the solution’s launch in September 2016, overall room usage stood around 65 percent. Since then, it has jumped to as high as 85 percent. Whereas students and teachers were previously spending 10 to 15 minutes trying to find a free space, they are now able to easily look ahead and book an available room immediately.
This is a great use of Azure services to provide a real and measurable return on the investment. Not only was teaching and learning able to start faster (up to 10-15mins), with a 20% improvement in room utilization, by understanding room usage more comprehensively, the College saved around USD$163,000 a year in infrastructure and maintenance costs.