We used to be in charge of operations in a coworking space and we found it difficult to keep track of the usage patterns of the tenants. The printer would sometimes be out of paper/ink, the trash can would be filled to the brim, and sometimes the toilet would be choked and no one reports it. Even though there was a reporting app, seldom do the tenants use it to report issues as it is too much of a hassle.
What it does
The hardware IoT button adopts a plug-and-play approach, where the buttons can be deployed in a multitude of use cases—deploying alongside printers, trash can, paper towel, toilet sinks—to collect crowdsourced data. Beyond that, building owners could associate a RE:BUTTON with anything he wants to track. For instance, he could setup the button, drag it on the floor plan map, and finally place the button in its physical location. Tenants who see that the item needs to be replenished would simply just press the button and this data would be sent to our web application. Managers can also immediately fix the issue and send automated emails.
Our data analytics would allow the building owner to look into the usage patterns of their items and facilities over time, and we have a machine learning algorithm in the backend to predict the lifespan of these items. Most importantly, instead of spending a few hundreds or even thousands to collect data using new IoT devices or like the toilet feedback iPad that we are all familiar with, Re:Button provides a cheap and effective alternative.
How we built it
We have re-purposed the Amazon Dash button by looking for the packets of data that gets transmitted from the amazon dash button in the wifi network. For our web application, we used d3.js, angular and recharts and also prototyped a machine learning prediction model using prophet, a library released by Facebook for time series modelling.
Challenges we ran into
Our idea involved using Amazon Dash (can be bought for about 4 dollars) as our button and it was difficult to find a way to hack the Dash button such that we can detect the button press from our backend. We had to use Wireshark to find out all the packets of data that were transmitted in the network and it took us quite some time to find the packet that was sent from the Amazon Dash button. After we found it and managed to listen for that packet of data sent when the button was pressed, we were SUPER excited about the possibilities of REBUTTON.
As it was our first time using D3 and SVG, it was difficult for us to integrate the floorplan map — we had some issues initially to detect the mouse click on the SVG but we managed to solve it in the end.
Accomplishments that we're proud of
We managed to come up with an interesting idea, hacked the Amazon Dash and even build a web application prototype that we are proud of! Most importantly, we found the time as a team to participate in this hackathon despite it being our exam week!
What we learned
It is important for great time management skills and to ensure that the strengths of every individual is properly utilised.
What's next for RE:BUTTON
As a team we are excited about bringing this idea to greater heights. Jing Shun and Benjamin are both flying to Shen Zhen as part of the school programme where they would visit the HAX Accelerator and makerspaces around Shen Zhen. Using Jing Shun’s mechanical engineering background, we would redesign the form factor of REBUTTON to make it as small and as invisible as possible such that it can be retrofitted to different use cases and invisibly sending data back to the backend. The possibilities of REBUTTON is only limited to our imagination.