How Downer is using sensors to predict Sydney Trains maintenance
Australian giant Downer has a 30-year contract with the New South Wales government to manage and maintain its fleet of 78 Waratah trains that operate in the greater Sydney metro area. With 2041 not approaching any time soon, the company recognised a perfect opportunity to maximise technology to make the most of its data and plan for proactive, rather than reactive, maintenance of Sydney's trains.
In December 2016, the NSW government ordered 24 Waratah Series 2 trains under its Sydney Growth Trains Project and in February 2019, announced the decision to order 17 more trains. The new trains are touted as providing passengers with improved safety and comfort, fitted with air-con, more CCTV cameras, and improved accessibility.
Downer general manager of Digital Technology and Innovation Mike Ayling said his company saw this as the perfect opportunity to leverage additional sensor data from the fleet.
As each Waratah train pulls in and out of a Sydney station, more than 300 Internet of Things (IoT) sensors and almost 90 cameras silently capture data and record video. Every 10 minutes, 30,000 signals are sent from the train to Downer.
According to Ayling, those 30,000 signals represent the train's digital DNA.
"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," Ayling said.
"We're using those sensors to tell us about the health of the train -- it's almost like having a blood pressure reading."
Downer deployed a Microsoft Azure-based intelligent solution that ingests sensor data from Sydney's fleet of Waratah trains and spits out something useful that is easily digestible to engineers and other staff.
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's TrainDNA platform ingests the data and uses Azure machine learning to make sense of it.
The collaboration followed Downer turning to Microsoft in 2017 to form a partnership aimed at developing and marketing cloud-based solutions and services for specific industry sectors. The alliance, according to Microsoft, sees "both parties bring their technology and sector specific know-how to the table", and was designed to help "accelerate the rate at which transformational value could be unlocked for business".
Speaking with media at the Auburn Maintenance Centre in Sydney, Ayling said the platform was built with two audiences in mind: Firstly for the fleet of support officers who are working with Sydney Trains in their operations centre making sure the trains that are on the network are running all the time; and for the fleet engineers who are looking at the longer-term trends.
Data from the trains are received by the platform every 10 minutes, but Ayling told ZDNet there are hopes to reduce that to real-time, at least for the important information.
"At the moment, we're getting everything every 10 minutes, we're going to change that profile to say, 'We want these more frequently'," he said. "We've got to be careful, because if we say we want everything within a millisecond, the infrastructure ... it might not be able to cope."
Essentially, TrainDNA is helping to automate inspections, while also giving Downer the opportunity to optimise operations and introduce predictive maintenance, Ayling explained.
"This is the Holy Grail -- you bring it in and you maintain it based upon its condition. Now you imagine the time taken to maintain a train every 30 days, as opposed to bringing it in only when it needs to be brought in. So that's where we would get significant cost savings, plus the fact that there's a 30-year contract, there's some significant pieces of work where we need to overhaul major assets like bogies," Ayling explained.
Inside Downer's maintenance centre
Although the company had the raw data, it struggled to make sense of it or get it to the people who needed it, when they needed it.
With the overarching idea to reduce maintenance time and cost, Ayling said being able to make sense of the data it holds is already proving to be beneficial, offering an example of train carriage doors and understanding why exactly they take up such a decent chunk of maintenance time.
"Door spatial analysis -- this has got the opening and closing times of every door, there's also a geospatial map that shows where there are issues with door opening and closing, at particular stations," he told ZDNet.
"This is stuff we were sitting on, but not thought to look at it and not been able to look at some correlation between door opening times slowing down and particular parts of the network ... we'll be able to give that information to Sydney Trains and find out what is happening [for example] in Hornsby ... and work together -- because what we tend to do is deal with the consequences, when something goes wrong, we fix it ... but we can hopefully now get to the source and find out what was causing that."
It isn't just the Waratah fleet that Downer has the maintenance contract for, as the plans for TrainDNA also involve Downer's High Capacity Metro Trains (HCMT) fleet in Melbourne, with the company looking at how it can utilise what it is doing with the Waratah fleet elsewhere in the business.
Downer employs around 56,000 people across more than 300 sites across Australia, New Zealand, the greater Asia-Pacific region, South America, and Southern Africa.
In addition to rail, the company has business units spanning utilities, engineering, construction, and mining, to name a few. Although not a requirement of the project, Ayling said TrainDNA could easily be tweaked to fit other areas of the business.
"What we're trying to achieve -- TrainDNA is essentially an AI platform, built on top of with AI services on datasets ... today we have the Waratah trains, in the future it could be any other train set," Microsoft Australia national technology officer Lee Hickin added. "But the principle of it is how do we unlock the value of the data, how do we apply reusable AI assets -- intelligence services and tools that we built in Azure machine learning -- and then visualise them.
"I think the platform will continue to grow and grow."
With sensors already in place, and more to follow, Ayling is keen to explore how technology such as Microsoft HoloLens could help engineers with maintenance.
Read also: How Microsoft is making its most sensitive HoloLens depth sensor yet
Wearing a head-mounted device, engineers will be able to look at a train and see an overlay of the data, the train blueprints, technical drawings, and insights about what maintenance is required.
"We have some research going on with HoloLens, mixed reality, where they're getting live information that they can see in front of them, on how to maintain the train," Ayling explained.
"The tablets can provide data and insights to the maintainers while they're actually on the job, and HoloLens I guess is the next step where they can actually visualise the maintenance and see how the procedure actually carried out.
"When 5G comes, part of me is thinking we might be able to get more and more data," Ayling added, with Hickin also noting that the pair are looking at how they can offload some of the business process to the edge -- that is, on the train through its sensors.
MORE FROM MICROSOFT IN AUSTRALIA
- Microsoft Technology Centre opens in Sydney
- Canberra gives Microsoft protected-level cloud classification
- Microsoft and NAB trial biometrics for cardless ATMs
- RMIT moves to Office 365
THE RISE OF INDUSTRIAL IOT
- What is the IIoT? Everything you need to know about the Industrial Internet of Things
- 3 things you should know about Industrial IoT (TechRepublic)
- How IoT can transform four industries this year
- How IoT will drive the fourth industrial revolution
- How machine learning and the Internet of Things could transform your business