AirCondor

Team Name: 
Team GoshAwks

After much deliberation, and many blocked turns, our team finally nailed down a plan to develop an air conditioner recommendation application late on Saturday night. 

To do this, we collated the air conditioner cooling data and energy rating for air conditioners datasets. 

This provided us with a large dataset that we were able to query and interpret to determine the best air conditioners based on location and climate. 

Driven by a desire to create a tool for residents that can assists them to not only save money, but also impact the environment more positively, we created a web application that allows the private user to enter their postcode location and view ratings and efficiency details for air conditioners. 

Once the postcode information was recorded by the system we were able to query the database that we created to see which air conditioners matched the climate (according to recorded Beaureau of Meterology information). 

Implementation

We used R to deconstruct and join the existing datasets mentioned above.  We also used Python to convert geographic data, and csv files.  QGIS was used initially to examine and view geographic datasets while designing our project. 

The transformed data was then entered into Tableau to present it in an appropriate visiual manner.  

Postcode data taken from the airconditioning data sets was used to generate a Postcode Layer within the Mapbox toolkit.  We were then able to associate the air conditioner data with the postcode markers.  Our prototype relies on a single example image of a table due to our time constraints, however we also have solid proof and understanding of how to query and load our transformed data into tables within the marker popups dynamically. There's a voice recognition feature using Google open source code as well, which takes user's voice of an Australian postcode and outputs a table with php and json (Since our demo website is a static website base on Github, the php dynamic feature tested on UTAS server is not well presented on the demo website. A screencapture picture might be able to show how it works: link). 

 

Discussion

We believe this application could benefit many citizens across Australia, showing them the difference between air conditioner efficiency in different climates and helping them to understand appliance running cost. With the growing number of Tasmanians investing in air conditioners it is especially helpful to find the best appliances for the cooler weather.  We had anticipated a plan to include forecast data from the Beaureau of Meterology to provide examples of efficient usage in real-time (ie. 'The weather will be quite cold later in the week. The best way to prepare is to maintain a constant temperature within your home.  Based on the predicted forecast, we recommend you maintain a range of 19-21 degrees celcius this week.").  Unfortunately, due to time constraints, we were not able to implement this particular feature.

 

Datasets Used: 
Air-conditioner Location Running Hours Data https://data.gov.au/dataset/air-conditioner-location-running-hours-data Energy Rating Data for household appliances – Labelled Products http://data.gov.au/dataset/energy-rating-for-household-appliances

Local Event Location: