Trackify

Inspiration

One of the most important, yet easily unnoticed, aspects about fleet management is preventative care and maintenance. It's not the quality of a vehicle's build that causes a premature demise, but rather improper maintenance. Given a lot of fleet data, our team would analyze it, so we can promote a more pro-active, instead of reactive, approach to maintenance, thus ultimately saving a company money and preventing unnecessary waste.

What it does

A user can log into the app and check up on their fleet of vehicles. The user can add new vehicle by simply scanning the QR code containing vehicle number .From there, a python script manipulated the data given to find vehicles that need service/maintenance. The app will then tell the user which vehicles need what service, and the user can easily locate what vehicles need service by using augmented reality and location data! It chiefly takes into consideration the work laod over engine to compute the next required service check but also the other two main features are that it will use the latitude longitude data along with timestamp to check for weather log at that day, and also it would leverage the location data to find out terrain of the location. The weather and terrain form two important factor alongside of mileage/fuel efficiency of vehicle these all parameters are taken into consideration to find out when is next maintenance required and what type of maintainance is required. Based on location it suggests the nearest service station at the time of required maintainance.

How we built it

The python script communicates to the app by a firebase database. The authentication is also done from firebase. We used a t-test to find what vehicles need maintenance.

Challenges we ran into

Dealing with a lot of data was difficult as it caused for very long run times just to work with the data.

Accomplishments that we're proud of

We managed to find vehicles that needed maintenance and present that in a very user-friendly way! We also learned a lot about data analysis techniques and managed to work efficiently as a team the whole time.

What we learned

We learned many types of data analysis techniques, and how to collaborate between back and front end.

What's next for Trackify

Currently, we had only ran 1 type of t-test on the data, and found some outliers in the data. We want to do more tests in the future and implement a cluster algorithm to find out more about the data. We also want to upload more information to the databases so, for example, the app knows the last location of a vehicle for the AR to flash "service required" or "not required."

Built With

• android-studio
• ar
• firebase
• java
• pyrebase
• python