Machine Learning in Train Network Operations

About this course 

This taster course explores how UTS and Sydney Trains are collaboratively working towards developing a timetable robustness evaluation model, using machine learning.  This is the first data-driven model that can provide detailed, station-level, line-level and network-level analysis and evaluation results and predict in real-time, the delay effect after capturing delays.

Course structure

The outcome of this innovative application of the intelligent timetable evaluation technology significantly reduces delay-caused losses and increases operation efficiency. This enables the train operating system to meet performance metrics and pursue timely recovery from incidents.

The model will be able to assess timetables and response plans to ensure that they are operationally robust and resilient. Based on the statistics, improving the on-time running rate by 1% could potentially save customer ‘lost-minute’ value by $5 million.

The proposed generic model can be easily applied to other traffic scenes, with subtle refinements. This work has demonstrated to train operating companies that they can produce highly detailed and granular information to develop targeted timetable design and real-time scheduling strategies. Importantly, for rail managers and controllers, end-to-end timetable evaluation and delay prediction is automatically achieved by data-driven techniques. This eliminates the need for domain expertise and hard-core feature extraction.

Learning outcomes

This course will provide an opportunity for participants to engage with the following learning outcomes:

• The ability to describe how data can be used to inform decision making in real-world contexts

• The ability to describe machine learning and advances in machine learning

• The ability to explain how data science can be applied in real-world contexts to provide innovative solutions.

Who is this course for?

This taster course is suitable for anyone who is interested in machine learning and a real-life application of its benefits, as demonstrated through a case study of the Sydney train network.