Data driven prediction, prescription and optimisation models

Project description

Optimisation of existing preventative maintenance processes and techniques often leads to diminishing returns. Preventative maintenance typically includes a contingency to allow for unknowns, which can result in unnecessary or early maintenance. To achieve improvement in outcomes, a new way of prescribing the required maintenance and its timing is essential. This research reviews the various condition and maintenance data sources available to develop predictive and prescriptive tools to ascertain the optimal time and optimal intervention activity to increase the life of the asset while minimising cost and any impact on operations. The project will develop exploratory, predictive and prescriptive tools that will work with railway condition data to provide data driven insights and decisions.

Industry Partner

The project will be conducted in collaboration with the industry partner BHP.

Doctoral Candidate

Mr Dean Van Der Woude

Dean Ashley Van Der Woude

Supervisors

Academic supervisors:

  • Prof. Ye Lu, Department of Civil Engineering

Co-supervisors:

  • A/Prof. Guido Tack, Department of Data Science and Artificial Intelligence
  • Dr. Nithurshan Nadarajah, Monash Institute of Railway Technology, Faculty of Engineering

Industry supervisor:

  • Mr Romain Saporito, Principal Engineer-Rail, BHP Iron Ore
Faculties Involved

Engineering and Information Technology (IT)

Contact Details & Additional Information