Inspiration
There are many applications working in the smart building based on the occupancy, presence, and flow of the population in smart buildings. These applications can get population data sample set and predict future distribution of people to provide services such as cleaning, cooling, heating and ventilation. The tangible outcome of these applications is cost reduction of electricity price for light, heating and cooling systems to enhance indoor environmental quality.
What it does
We have fetched sensory data for desk occupancy from the web server. Then we cleaned the data, and removed the anomalies. Later the visualization is provided to illustrate flow of people in the building in different hours of the day.
How I built it
We used python urllib2 library to fetch the data for all the desk sensors from the server. Then we cleaned it to remove the duplicate. Then the time stamp and date for each desk occupancy was calculated. We have computed and monitored the percentage of occupancy of a desk over time, and visualize it on the top of building plan.
Challenges I ran into
It took one hour to fetch whole data for all desks from the server. There were many artifact contaminating the sensory data; and the amount of data before duplicates removal was huge.
Accomplishments that I'm proud of
Explore the whole data from the server, clean it and get a sensible meaning from it.
What I learned
It is interesting to generate different scenario with the sensory data, which make sense and solve a real life problem. During our hacking session, we came up with different crazy ideas to benefit people in different areas.
What's next for Demo (MOnitoring Employee's Desk)
We are currently planning to apply a clustering machine-learning algorithm to label different areas in the building with time and population intensity in favor of energy management and make a better work place.
Log in or sign up for Devpost to join the conversation.