Cloud-based IoT platform delivers predictive maintenance for Sydney’s train fleet

Microsoft, Downer detail TrainDNA rollout

Infrastructure services company Downer and Microsoft have revealed details of an Azure-based IoT platform delivering predictive maintenance capabilities to Sydney’s train fleet.

Downer and Microsoft in 2017 established a strategic alliance to develop cloud-based solutions for specific industries. Downer Rolling Stock Services is one of the first adopters of the ‘TrainDNA’ platform: An IoT solution produced by the two companies that is underpinned by Microsoft’s Azure cloud.

The platform can help estimate the risk of failure of a carriage and pre-emptively schedule preventative maintenance. As part of the use of TrainDNA in Sydney, Waratah trains have been equipped with more than 300 sensors and some 90 cameras.

“Essentially these are trains with brains,” said to Mike Ayling, Downer’s general manager of digital technology and innovation at Downer. “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.”

Downer EDI is part of a consortium that in 2006 entered into a rolling stock public-private partnership with the NSW government to build and maintain Waratah carriages. Under a 30-year contract with the state government, Downer oversees maintenance for a fleet of 78 trains. The company has also been overseeing the delivery and maintenance of the new Waratah Series 2 trains.

Microsoft said that Azure IoT Hub feeds stream analytics into an Azure Data Lake Store and Azure SQL database. TrainDNA uses Azure machine learning services to analyse the data, but the platform can also allow the use of algorithms developed by an operator and run through Azure Batch.

The platform can detect anomalies in variables such as temperatures, voltages and currents, and the opening and closing of doors.

“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,” said Lee Hickin, national technology officer, Microsoft Australia. 

“There are clear safety and reliability benefits for the millions of people that rely on Sydney trains every day – and efficiency and cost benefits for Downer.”