OmniEye

• 

  OMNIEYE - Perceive.Predict.Prevent.

Inspiration

• A lot of money is spent on maintaining the rail operations to ensure smooth operations
• However, asset failure causes severe delays and losses to the company and to the people
• There is a huge opportunity to predict failure rate of railway assets

What it does

• A predictive model that is able to predict asset failure based on weather data
• Fuzzy logic recommendation engine to suggest remedial actions based on actions taken in the past

How I built it

• Python
• Jupyter Notebook
• Microsoft Azure Machine Learning Platform
• Random Jungle

Challenges I ran into

• Dirty Data
• Choosing the right parameters

Accomplishments that I'm proud of

• High accuracy (~70%)

What's next for OmniEye

• Automated maintenance ticket issuance for engineers based on priority

Built With

• azure
• python