Who can Enter:

Anyone who is enrolled in CDU IT Courses (short course, VET, undergraduate and postgraduate courses) in Academic year 2023.

Important dates

  • Challenge and registration open: Monday, 11 September 2023
  • Submission deadline: Wednesday, 18 October 2023 at 12:00 noon ACST
  • Presentation and judging: Wednesday, 25 October 2023 at 9 am to 1:00 pm ACST
  • Venue: Purple 12.1.15

Info Session:

Panel of judges

Ewan Perrin GAICD
Executive Director, Digital Government at Northern Territory Government

Donald Young
Senior Director, Digital Strategy at Northern Territory Government

Dr Guzyal Hill
Senior Lecturer (Law) and Deputy Chair Academic Board and Academic Programs Committee

Liam Ma
Solutions Architect, The Arnhem Land Progress Aboriginal Corporation (ALPA)

Background

Australia’s Information Technology (IT) landscape has been marred by several high-profile IT project failures. Billions of dollars were wasted on projects that never delivered on their promises. The Robodebt scheme, the Queensland Health payroll system, and Victoria’s Myki Smart Card are just a few examples.

Are there patterns behind these failures? Can these failures be anticipated? And what does the data tell us? 

Mission

To begin, your team should collect data on high-profile IT project failures from Australia and worldwide. These may include both governmental and non-governmental projects. By applying machine learning techniques to this data, you will identify patterns and factors that may have contributed to these failures and determine the extent to which failed projects could have been predicted.

Datasets

Start by identifying a list of failed IT projects that are publicly available. Sources may include but are not limited to:

Your team should consider how to perform featurisation on these datasets. There is no restriction on how this should be done, although a high level of automation is strongly desirable.

There is also no restriction on which predictive algorithms to use. Experimentations are encouraged.

Rules of competition

  1. Participants must be in teams of 2 to 4 students.
  2. The data science solution must be developed in Python. Results must be submitted as .py scripts or Jupyter Notebooks written in Python.
  3. There is no restriction on the specific Python packages.

Submission

The following items must be submitted before the closing date of the submission (not the presentation date):

  • One .zip file containing .py scripts / Jupyter Notebooks, all processed datasets, and the requirements.txt file to replicate your Python environment.

Label your submission with your team’s name.

Upload your submission SUBMISSION IS CLOSED

Team presentation

Your team must pitch your work to a panel of judges on the stated presentation date at Charles Darwin University Casuarina Campus.

Duration: 10 minutes pitching + 5 mins Q&A per team

Judging criteria

Participating teams will be judged along five criteria:

  1. Innovation (20%)
  2. Sophistication (20%)
  3. Results (30%)
  4. Scalability and automation (10%)
  5. Presentation (20%)

Further readings

Willcocks, L. and Griffiths, C., 1994. Predicting risk of failure in large-scale information technology projects. Technological forecasting and social change47(2), pp.205-228.

Chen, H.L., 2015. Performance measurement and the prediction of capital project failure. International Journal of Project Management33(6), pp.1393-1404.

Owolabi, H.A., Bilal, M., Oyedele, L.O., Alaka, H.A., Ajayi, S.O. and Akinade, O.O., 2018. Predicting completion risk in PPP projects using big data analytics. IEEE Transactions on Engineering Management67(2), pp.430-453.

Wang, Y., Shao, Z. and Tiong, R.L., 2021. Data-driven prediction of contract failure of public-private partnership projects. Journal of Construction Engineering and Management147(8), p.04021089.

Basurto, N., Jiménez, A., Bayraktar, S. and Herrero, Á., 2023. Improving the prediction of project success in the telecom sector by means of advanced data balancing. Cybernetics and Systems54(3), pp.306-320.