top of page

I am grateful to have been awarded approximately $20 million in research grants (more than $10 million as first chief investigator) from the Australian Research Council, and other sources external to my university, to conduct the research described below. Excluded from this list are the industrial optimisation projects from our industry partners that form the core of the ARC Industrial Transformation Training Centre in Optimisation Technologies, Integrated Methodologies and Applications (OPTIMA) that I lead as Director. Further details about the research undertaken in OPTIMA is found here.

Current and Recent Projects

Australian Research Council grants
  • Numerous industrial optimisation research projects funded under the ARC Industrial Transformation Training Centre in Optimisation Technologies, Integrated Methodologies and Applications (OPTIMA - see optima.org.au), 2021-2026, over $8m from ARC, industry partners and Monash and Melbourne Universities.

​​

  • "Stress-testing algorithms: generating new test instances to elicit insights", ARC Laureate Fellowship 2014-2019, $2.83m

 

This project aims to develop a new paradigm in algorithm testing, creating novel test instances and tools to elicit insights into algorithm strengths and weaknesses. Such advances are urgently needed to support good research practice in academia, and to avoid disasters when deploying algorithms in practice. Extending our recent work in algorithm testing for combinatorial optimisation, described as 'ground-breaking,' this project aims to tackle the challenges needed to generalise the paradigm to other fields such as machine learning, forecasting, software testing, and other branches of optimisation. An online repository of test instances and tools aim to provide a valuable resource to improve research practice and support new insights into algorithm performance.

 

Project website MATILDA

Recent media: 

buscard2.jpg

At ARC Laureate Award Ceremony in Adelaide, August 2014, with Prof Aidan Byrne, CEO of the ARC (standing in front of bust of Nobel Laureate Robin Warren)

  • "Optimising progress towards elimination of malaria", ARC Discovery Projects 2020-2023, $520,000 (with Jennifer Flegg, Jonathan Keith and Jack Richards)

​​

The project aims to advance mathematical knowledge by developing novel tools appropriate for modelling disease elimination. We will apply these new mathematical tools to the significant problem of malaria elimination in Vietnam. The expected outcomes are new tools for modelling disease elimination on a fine spatial resolution with heterogeneities in individual patient characteristics, calibrating models to household level data on disease transmission and designing intervention strategies for maximum effect on disease transmission. The innovative combination of modelling, inference and optimisation ensures that the mathematical methods developed will be broadly applicable to modelling elimination strategies for other infectious diseases.

​​

  • "Faster, cheaper, better: mathematical advances for improved design and scheduling of robotic instrumentation", ARC Linkage Projects 2014-2016, $421,276 (with Industry Partner Leica Biosystems)

 

​This project extends a successful Researchers-in-Business project, addressing mathematical challenges in the optimal design and scheduling of robotic instrumentation. The Partner Organisation manufactures instruments for cancer diagnostics, and designs instruments that need to produce rapid, high quality results, at a reasonable cost in a competitive market. Powerful new scheduling algorithms will be devised to handle their complex problem, which is more challenging than standard problems. The developed methodologies will reduce the product development cycle and boost the competitiveness of Australian manufacturers. In addition, new theoretical and algorithmic contributions will enable improved scheduling in other application areas.

 

  • "Optimising experimental design for robust product development: a case study for high-efficiency energy generation", ARC Linkage Projects 2012-2014, $342,214 (with Rob Hyndman and Industry Partner Ceramic Fuel Cells Limited)

​​

The development of robust products can be a tedious and expensive procedure. Our industry partner manufactures a world-leading alternative energy product, with around 200 variables affecting product quality. The prohibitive cost and time to build and test trial products creates a challenging multi-criteria optimisation problem: find the optimal setting of numerous decision variables, subject to many constraints, while minimizing the effect of noise factors and relying on only a limited number of experiments to simultaneously optimise several objective functions related to cost, quality, and functionality. Our new methodology and tools aim to transform the competitiveness of our industry partner and the manufacturing sector in general.

​

  • "Footprints in Instance Space: Visualising the Suitability of Optimisation Algorithms", ARC Discovery Projects 2012-2014, $365,000 (with Edmund Burke)

​​

Knowing precisely when and why an optimisation (or any) algorithm will perform well is crucial to avoiding deployment disasters, gaining theoretical insights to improve algorithm design, and ensuring that algorithm performance is robustly described independent of chosen test instances. The project is developing powerful methodologies for learning and visualising the boundaries of algorithm performance (footprints) in a high-dimensional instance space. Using the instance space representations we can understand the unique strengths and weaknesses of algorithms, predict performance on unseen instances, identify where existing benchmark instances lie, and generate new instances that fill the gaps in the instance space. The figure to the right shows an instance space for graph colouring problems. Each instance is represented as a point in this 2-d space, calculated as a projection from a higher dimensional feature space. Each instance is coloured according to which algorithm is predicted to perform best. More details are available here in Smith-Miles, K. et al. "Towards objective measures of algorithm performance across instance space", Computers & Operations Research, vol. 45, pp. 12-24, 2014.

aims.png
Other projects

There are also several projects that I am currently involved in through MAXIMA - collaborating with interdisciplinary colleagues who need mathematical solutions, or with industry partners. A few of these are listed below:

  • Mathematical metrics of improvement and diagnosis in children with developmental delays when playing an interactive game - with Professor Kim Cornish and team in the School of Psychological Sciences at Monash University, in partnership with Grey Innovation and Torus Games

  • Predicting the Outcomes of Hypervelocity Impact Events via Computational Intelligence - with Defence Science Technology Organisation

  • Music as a canary: predicting onset of depression via music and social media usage analysis - with A/Professor Nicki Rickard and team in the School of Psychological Sciences at Monash University, in partnership with Beyond Blue

  • Extracting and classifying shapes from noisy time series, particularly arising in atmospheric turbulence - with Dr. Danijel Belusic and Yanfei Kang

Earlier Projects

Australian Research Council grants
  • "Direct Stimulation of the Visual Cortex: A Flexible Strategy for Restoring High-Acuity Pattern Vision", ARC Special Research Initiatives (2010-2014), $8.1 million (with Arthur Lowery and 16 other investigators)

 

  • "Automatic Human Age Estimation Based on Visual Information", ARC Discovery Projects 2009-2011, $245,000 (with Zhi-Hua Zhou and Xin Geng)

 

  • "Concept-Based Multilingual Web Content Mining", ARC Discovery Projects 2004-2006, $237,466 (with Chung-Hsing Yeh and Rowena Chau)

 

  • "Improving the modelling of insolvency risk and financial health assessment of global companies using hybrid intelligent techniques", ARC Linkage Projects 2004-2006, $164,253 (with Andrew Flitman, Vincent Lee, and industry partner Lincoln Indicators)

 

  • "Intelligent techniques to exploit the dynamic temporal structure in detection of attacks in credit application fraud", ARC Linkage Projects 2004-2006, $164,253 (with Vincent Lee, Ross Gayler and industry partner Veda Advantage)

 

  • "Realising the promise of neural networks for practical optimisation: improving their efficiency and effectiveness through chaotic dynamics and hardware implementation", ARC Discovery Projects 2002-2004, $235,000 (with Terence Kwok)

 

  • "A new approach to premium pricing in the insurance industry based on data mining", ARC SPIRT Scheme 2001-2003, $205,561 (with Rob Willis and industry partner AAMI)

 

  • "Dispatch optimisation and constraint modelling in the Australian National Electricity Market", ARC SPIRT Scheme 2001-2003, $97,335 (with Lyfie Sugianto and industry partner NEMMCO)

 

  • "Real-time metaheuristics for industrial problems: applications to cellular manufacturing, telecommunications, and logistics", ARC IREX Scheme, 2000 (with David Abramson, Sebastian Lozano and Fernando Guerrero)

 

 

 

 

Other past projects

I have also completed numerous industry projects with organisations including: DSTO, CSIRO,  Hughes Network Systems (USA),  TAB (NSW), AXA, NAB, Telstra, Alfred Hospital, Frankston Hospital, GlaxoSmithKline, Grey Innovation, Melbourne Storm Rugby Club, Agilent Technologies.

 

My academic collaborations have included researchers at University of Edinburgh (UK), Sterling University (UK), IITB (India), Nanjing University (China), South Eastern University (China), University of Caterbury (NZ), University of Seville (Spain), Ball State University (USA), Nanyang Technological University (Singapore), University of Montreal (Canada), Leiden Institute of Advanced Computer Science (Netherlands), University of Chile, as well as many Australian universities, and collaborators from other disciplines at Monash University.

bottom of page