Project Level: Summer

Project Duration: 6 weeks

Hours of Engagement: 36

Project Description:

Forecasting and decision-making with optimisation models are two widely used techniques that together can solve many real-world problems. For example, you may be interested to forecast demand for products and then accordingly make a decision to determine your optimal inventory level, or you may be interested in forecasting the electricity demand and optimising the company operations. In this class of problems, the prediction will be used as an input to the optimisation model but more accurate forecasts do not lead to better decisions. We need to be accurate in our forecasts not just on average but where it matters the most. The problem that we would like to answer is how we can integrate these two phases and develop an end-to-end model that can optimise the decisions and forecasts. We can do so by either Bayesian Inference or by using machine learning models that take into account the final decisions in forecasting.

Expected Outcomes: 

You will learn:

• Best forecasting methods.

• Working with real-world data.

• Conducting state-of-the-art research.

There is an opportunity to:

• Present at national or international conferences.

• Writing an article for international journals.

• Collaborate with other researchers in Australian and UK universities.

Suitable for:

Suitable for Masters students with experience in either Bayesian Statistics, optimisation, or Machine learning. Being proficient in R or Python is necessary. Willing to learn best forecasting methods.

Contact for further information:

Dr Mahdi Abolghasemi:

Project members

Dr Mahdi Abolghasemi