Saad Khan

At just 28 years old, Saad Khan joined engineering firm Downer to turn predictive maintenance into a reality within the company’s rollingstock services (RSS) unit.

Apart from having to define the project, create a team and compete for budget, Khan had an incredible engineering challenge: to predict the need for maintenance before a failure manifested as a problem that affected service.

Khan put together the framework for his predictive maintenance program, which consisted of an asset management strategy, a technical study to define the exact data required from the train, a data analytics platform and a reliability-centred maintenance model.

After joining Downer in 2017, he spent the remainder of the year finding talent to execute his program.

Managing multiple teams spread across Australia, including various partnerships with universities and local businesses, he developed the program, which finally launched this past March.

But building the data-analytics platform in an organisation focused on industry services was a significant challenge. Khan’s managers provided support as he proved the need to invest in a platform that catered to multiple asset types.

The result was TrainDNA.

TrainDNA is a highly scalable asset-agnostic data analytics platform.

The culmination of hundreds of hours of work, its foundation is
Microsoft’s software stack, packaged into a cloud-based, scalable,
software-as-a-service solution called Neuroverse, which was internally developed by Downer’s D3S team in Perth.

To build TrainDNA, Khan needed data scientists, which he addressed through a long-term RMCRC partnership with Deakin University in Geelong. This built significant machine learning and statistical data science capability in TrainDNA.

Downer RSS has already seen benefits in the form of managing critical components of its train during its testing phase.