Weather Influence on Assets Failures

We were interested in seeing how the railway data and how the weather can actually impact railway assets. That is why we chose to tackle this challenge. The main aim is to understand whether there is a correlation between the weather conditions and railway assets failures

This project has an analysis of the correlation between the weather data and railway assets failures as well as two predictive models that predicts if an asset is going to fail as well as what kind of failure type does it fall into.

All the analysis are done in Python and Jupyter Notebook, we built two predictive models with one being Random Forest model and the other being K-Means Clustering.

Dataset needs a lot of cleaning and understanding. It took us a lot of time to complete that.

Three Days. Two Man. One Hackathon. We are proud that we managed to complete our analysis on this and built two predictive models.

We have learnt about how weather can influence railway assets failures as well as how severe the impact can actually be.

Firstly, to understand the data at a lower level where models should be built targeting specific railway assets and section so problems can be pinpointed. Then, build a predictive maintenance dashboard in order to present the result in an easily interpretable way, so the maintenance team can be well informed of all times.