Embedded Files
Research Philosophy and Expertise
Research Philosophy and Expertise
My research philosophy is to work on real-world problems and develop advanced customised state-of-the-art solutions that have value for both industry and academia.
My research philosophy is to work on real-world problems and develop advanced customised state-of-the-art solutions that have value for both industry and academia.
My research expertise is in developing forecasting and decision-making models. We have been mostly focused on energy and retail supply chains.
My research expertise is in developing forecasting and decision-making models. We have been mostly focused on energy and retail supply chains.
Since my appointment as a tenure faculty member, I have worked closely with several industries. This includes projects on developing a predictive model for wind farms, forecasting and decision-making tools for energy and renewable energies, developing a sales forecasting model for food and beverage companies, and developing a predictive and prescriptive model for water treatment companies, among others.
Since my appointment as a tenure faculty member, I have worked closely with several industries. This includes projects on developing a predictive model for wind farms, forecasting and decision-making tools for energy and renewable energies, developing a sales forecasting model for food and beverage companies, and developing a predictive and prescriptive model for water treatment companies, among others.
Here are some of the projects that I have worked on. If you have a problem with forecasting and you want to build decision-support systems for your industry, I may be able to help you. Feel free to reach out and set a free consultation time with me via mahdi.abolghasemi@qut.edu.au.
Selected Industry projects
Selected Industry projects
Virtual Power Plant: Forecasting extreme events in electricity price, CSR Co, Australia
Problem: Considering a virtual power plant to reduce high energy costs.
Project and Outcomes: Developed forecasting models for electricity price and demand as well as solar power, that were used for scheduling. Estimated to save $100K/yr in costs.
Forecasting Promotional sales of food: Sanitarium Health and Wellbeing, Australia
Problem: Forecasting sales during promotional periods and optimising inventory
Project and Outcomes: Developed forecasting models for sales across states and stores. Resulted in 19% improvement in accuracy and 12% reduction in inventory & operations costs.
Forecasting renewable energies: Power generation from wind and solar farms, Worley Co, Aug. 2020 and Australian Energy Market Operator (AEMO), Australia.
Problem: Inaccurate short-term forecasts incurred operational costs/penalties (wind farm) & caused grid instability (energy system).
Project and Outcomes: Developed machine learning forecasting models and tested for 8 weeks by live data in production, which improved accuracy by 45%, saving est. $200K/yr in operational costs. See Media release.
Page updated
Google Sites
Report abuse