Integrated services company Downer EDI has deployed a Microsoft Azure intelligent solutions which ingests sensor data from a fleet of Waratah trains in Sydney to allow predictive maintenance and data-driven decision making.
The company believes this solution will enhance passenger safety, boost public transport reliability and keep tight control of costs over the contract lifespan.
Each train is equipped with over 300 sensors and around 90 cameras. Every ten minutes 30,000 signals are sent from the train to Downer, which holds a 30-year contract with the NSW Government to manage and maintain the existing fleet of 78 trains.
In December 2016 the NSW Government ordered 24 Waratah Series 2 trains under its Sydney Growth Trains Project.
The trains are being delivered and maintained by Downer.
Mike Ayling, general manager of digital technology and innovation at Downer said, “The new intelligence-based solution has transformed the way train maintenance is tackled injecting huge efficiencies and improving the reliability of the fleet for the millions of passengers using the public transport network.”
The Microsoft Azure based solution has been rolled out to capture and store IoT data, and use advanced data analytics and visualisation tools to make sense of the information, allowing Downer engineers to make data-driven decisions about train maintenance.
Downer said its Rollingstock Services business is one of the first adopters of the Azure based solution, used as the backend for their TrainDNA product.
Ayling said the impact of the digital transformation is already clear and that over the contract lifespan the more intelligently scheduled bogie overhauls alone has the potential to rein in costs significantly.
Tim Young, executive general manager – Rollingstock Services, Transport and Infrastructure said, “This is a data analytics platform on steroids. With such massive volumes of data it will allow us to establish trends in relative real time, allow us to predict failures in advance and calculate the remaining life of an asset more effectively.
“The advantage to our customers is that all of this takes place whilst the train is in service without interrupting the operation and at the same time enhances worker safety through the potential of removing high risk inspections.
“These enhancements in our asset management capability will boost our ability to better predict failure rates and reduce unscheduled down times of our train fleet and in turn result in fantastic outcomes for our customers and our business.”
Using predictive analytics
Downer noted machine learning and intelligent data analysis means it can predict the likelihood of failures, sometimes months in advance, scheduling preventive maintenance well in advance.
The front end of the solution is an Angular web app built on top of ASP.NET core services. The solution is hosted through Azure’s service fabric ensuring scale and resilience.
The Azure IoT Hub feeds stream analytics into an Azure Data Lake Store and Azure SQL database. Access is managed by Azure Active Directory with Power BI providing analytics and reporting.
Downer explained that its TrainDNA platform ingests the data and leverages Azure machine learning to make sense of it.
The company has ensured total flexibility by allowing data scientists to operationalise the algorithms they develop in Python or Julia or R, then containerise them and run them through Azure Batch.
Lee Hickin, national technology officer, Microsoft Australia said, “Downer has taken a complex IoT data collection and with the support of Microsoft used cloud and artificial intelligence to extract real meaning from the data and get it into the hands of engineers so that they can schedule and carry out maintenance on time, every time.
Ayling said essentially these are trains with brains, “We’re getting 30,000 signals from each train every 10 minutes. You extrapolate that out, we now have billions of data points since the inception of the fleet.
“We’re using those sensors to tell us about the health of the train – it’s almost like having a blood pressure reading,” which Ayling says provides Downer with the insights it and its engineers need to ensure trains continue to operate correctly and reliably.
“This is an absolute game changer for us in terms of how we operate, and being able to diagnose the trains early, being able to get to them before they fail.”
Downer is also preparing to leverage more unstructured data, such as video from the trains’ cameras, using Azure’s cognitive services API.
“This automation and digitisation moves us away from being an inspector and maintainer to a differentiated world class asset maintainer, driving condition-based maintenance,” Ayling added.