Phillip Marshall’s Industry placement in the railway industry

Published on

Placements offer the opportunity to experience the workplace environment and how workplace practices differ from academic ones. They also provide the chance to apply your skills in a new setting, solving new problems alongside a new group of people. The LIV.DAT CDT incorporates a 6 month placement for its students, allowing them some industrial experience to implement their big data training and broaden their PhD experience.

At LIV.DAT Phillip Marshall works on a project focussed on Deep Learning on the LHCb Detector. During the first 6 months of 2020 he applied his skills at a placement with OnTrac, a company that creates software products for the railway industry. This is his account of how these 6 months have been.

Phillip Marshall    

Ever since I can remember I have had a passion for all things that run on rails so, when I was given the opportunity to choose a data science placement, I knew what I wanted to do. I found a place at a software company called OnTrac based in Gateshead. They specialise in the area of track worker safety and produce a system that enables contractors to plan work safely on live railway lines. For the last decade they have been collecting data, but only recently decided to put that data to use.

I was asked to take a deep dive into their data to see what I could find. With help from my manager David Smith we produced a dashboard using Microsoft Power BI to visualise the data and prototype a product that could help contractors see if they were meeting targets for productivity, and importantly - safety, without having to collect and compile data manually from their employees, which is the case now. A highlight for me was being able to demonstrate my prototype to a key client in order to get opinions on what they liked and what could be improved.

The placement was also an excellent opportunity to learn about the railway industry itself. The field of track worker safety was one I knew very little about before my placement and the possibilities of big data were easy to see. Perhaps the most important thing I learnt was the challenges of data science. I had to work with colleagues to extract data to analyse, as they had never done this before. I also saw the wider picture in the industry, of how disconnected parties holding incompatible data creates difficulties – for example last minute notices to track workers only existing in pdf documents. Only when we tackle these issues can the big data future become a reality – something OnTrac were passionate about.”