11 Jun Design, Modelling and Simulation | Intelligent data fusion and analytics framework
Design, Modelling and Simulation
Intelligent data fusion and analytics framework
Downer (Saad Khan, Mike Ayling, Loic Ayoul) / Deakin University (Doug Creighton)
Downer has built a digital platform called TrainDNA, which seeks to digitise data taken from internal and external train monitoring systems. This platform uses cognitive computing tools to display up-to-date information on train location and status. Building upon this platform, the data science capability was augmented using Deakin University research methodologies to develop algorithms for predicting maintenance needs.
18 April 2019 to 22 March 2020
Total contracted budget (including in-kind)
- The team developed algorithms to assess train components to predict more accurate maintenance regimes and identify potential failures prior to occurring.
- The data analytics work being conducted has been implemented into Downer’s TrainDNA system, enabling engineers real-time access to data analytics.
- The analytics framework enables new models to be built that will assist Downer and its customers reduce down-time and optimise rollingstock maintenance.
Chen, J., Jindal, H., Philpot, D., Khan, S., Lim, C. and Creighton, D. Can I Trust Machine Learning Analytics? Downer’s Approach to Predictive Maintenance, AusRAIL PLUS 2019, 3-5 December 2019, Sydney Australia.