Exactly Here

We liked the potential of the idea to make a difference. We also liked the many directions the analysis could go in once the initial challenge of mapping the two sets of data onto each other had been achieved.

We built a scalable statistical analysis tool that displays both berth and circuit activation data on a single UI. It can be used immediately to determine where areas of special interest are on the tracks. Exactly Here is built for data analysts and train operators (initially). The dashboard removes many of the difficulties of working with inaccurate timestamps on the circuit data by using an approach that doesn't rely on them.

How I built it

Our project was primarily one of data and statistical analysis. Our goal was to merge the two datasets. We worked in python and pandas to extract, clean and load the data. We loaded a subset of the cleaned and sequentially merged the data combining it with a graphical user interface to allow the users to explore the data and extract new insights.

Challenges I ran into

Lots of data was missing from one dataset. Many of the timestamps were not aligned. The data was only semi-labeled.

Accomplishments that I'm proud of

We have created a scalable automated solution to merge these two datasets and gain new insights from a previously unused data soure.

What I learned

We have gained in-depth familiarity with this dataset and how it can be used to solve real problems within Network Rail.

What's next for Exactly Here

Given more time to work on this project, we believe we can scale this solution to work on a much larger level. We have lots of ideas about how this dataset can generate valuable insights for Network Rail, but due to time limitations, we were not able to pursue these ideas to their full potential. There are still some challenges to overcome but we have a clear idea of how we would tackle these issues and ensure that the solution is rigorous and reliable. We have not had an opportunity to fully test our proposed solution and would welcome the opportunity to test this on other data from across the UK.