Inspiration

After speaking to the challenge owners we saw a host of issues they were faced with, our heads began racing through all possible solutions. When we brainstormed together, we came up with this solution that combines they key data for all these issue in a great UI, and focused in on one - highlight catches we feel are 'fishy'.

What it does

We have developed an application to plot trip data vs catch event data vs fishing areas to hunt for FishyCoordinates.

Comparing the two datasets we were able rate each Catch Area Tokenization with a simple Red/Amber/Green for regulatory bodies to review. Comparing what the vessel categorised the catch area as vs the likely hood they were in this area based on their collision data.

Example of classification rules:

Green - Previous and next pings confirm they were in the area at this time.

Amber - If they recently entered or exited a catch area around the time of the catch

Red - They were not in the area before or after the catch

How we built it

We implemented a .Net Core API backend with a React front end.

Data at the minute is a mixture of hard coded & reading from XML/JSON files - database layer to be implemented.

Challenges we ran into

Our biggest challenges would have been

  • Mocking multiple datasets in to one dataset to work with, and also working with these limited datasets.
  • Time constraints - obvious one but working to a tight schedule

Accomplishments that we're proud of

Proud to have completed a full working demo in a short time that could be extended in a lot of different ways to solve many issues.

We also feel this is a realistic solution that could be piloted without too much development so to produce this in 24 hours was very satisfying.

What we learned

With a team made up for developers it was good to learn more about the problems the seafood industry is having around traceability.

What's next for FishyCoordinates

Blockchain Integration

  • Potential to write elements to a blockchain – Location Data, Catch Data, Regulatory sign off.

Machine Learning

  • Potential to train a model to become a self approving regulatory body to sign off flagged catches, trained based on this vessels historical data and previously flagged catches.

Internet Of Things

  • Device to capture GPS co-ordinates and send them to the API.

Data Capture

  • Capture your catch data via the app and push to API.

Data Analysis

  • With all this data in one place there is endless ways it could be used, one example would be to highlight areas that are currently being over-fished.

Built With

Share this project:
×

Updates