Dr. Abolghasemi has a multi-disciplinary background in Engineering, Business, Statistics, and Machine Learning. He is passionate about higher education and has studied and worked on three continents. His research interests and expertise lie in time series forecasting, predictive analytics, decision making and machine learning, with applications in supply chain management and renewable energies optimisation. Dr. Abolghasemi has consulted dozens of companies and developed forecasting and optimisation models for companies such as Sanitarium, Coca Cola, CSR, AEMO, Waterloo Wind Farm, Urban Utility, and Australian Renewable Energy Agency(ARENA). Mahdi's research is closely tied to evidence-based analytics for real-world problems.
Mahdi has published over 40 articles in top-tier academic journals and presented at numerous national and international conferences. He serves on the editorial board of the International Journal of Forecasting and is a regular reviewer for the International Journal of Production Economics, the International Journal of Production Research, and several international conferences. Mahdi is a member of national and international professional societies such as the International Institute of Forecasters, the Australian Mathematical Society, and INFORMS. He is also the founding chair and host of an international scientific podcast, "Forecasting Impact," which reaches audiences in over 120 countries around the world.
Dr Mahdi Abolghasemi is passionate about applied research, i.e, research with applications in solving real-world problems. Dr Abolghasemi has worked on several industry research projects during his working experience in the industry and academic journey. His research has been used in several sectors including the food and beverage supply chain, renewable energy, automotive, mining and construction industries. Mahdi has developed and optimised several forecasting decision support systems that are currently in use by industries.
Conference Publication: Approximating solutions to the Knapsack Problem using the Lagrangian Dual Framework
Keegan, Mitchell and Abolghasemi, Mahdi (2023). Approximating solutions to the Knapsack Problem using the Lagrangian Dual Framework. 36th Australasian Joint Conference on Artificial Intelligence, AI 2023, Brisbane, QLD, Australia, 28 November - 1 December 2023. Singapore, Singapore: Springer Nature Singapore. doi: 10.1007/978-981-99-8388-9_37
Journal Article: Book review
Abolghasemi, Mahdi (2023). Book review. International Journal of Forecasting, 39 (4), 1949-1951. doi: 10.1016/j.ijforecast.2022.12.002
Journal Article: Machine learning for satisficing operational decision making: a case study in blood supply chain
Abolghasemi, Mahdi, Abbasi, Babak and HosseiniFard, Zahra (2023). Machine learning for satisficing operational decision making: a case study in blood supply chain. International Journal of Forecasting. doi: 10.1016/j.ijforecast.2023.05.004
Analytics for the Australian Grains Industry (AAGI)
(2023–2027) Grains Research & Development Corporation
Predict and optimise with machine learning
Doctor Philosophy
Forecasting and optimising decisions with machine learing
Master Philosophy
A Review of Quarry Mine Datasets in Queensland
Doctor Philosophy
Statistical and machine learning applications for probabilistic energy forecasting
We aim to improve our understanding of Renewable Energy Sources (RES, wind and solar) over two timescales: 1) forecasting generation from hours to months, and 2) studying past and future climates to quantify projected changes in RES generation. Use physical models of the atmosphere (Numerical Weather Prediction and Climate Modelling) alongside historical observed data and new-generation data to drive the development of new forecasting methods and energy climate datasets. The forecasting component will develop seamless probabilistic RES forecasts with time horizons from hours to months using a blend of statistical and machine learning methods. Climate projections will develop high spatial and temporal (hourly) resolution RES data for onshore wind, offshore wind and PV from 2020-2050, using historical system data and ensembles of recent atmospheric model data, coupled with machine learning algorithms. This data will be used to investigate generation for different climate scenarios across RES alternatives. Statistical methods can rely on frequentist or Bayesian methods to help generate outcomes over short- and long-time scales, while machine learning techniques can range from simple decision trees to complex deep learning approaches.
This is an industry-supported project. It is suitable for Honors, and PhD students.
Forecasting sales and optimising inventory in food supply chains via machine learning
Probabilistic forecasting associates a probability of occurrence with the predicted values, making it a useful technique for determining decisions based on the level of risk one can take. It is a powerful technique that unlike point forecast gives you a complete view of the future unknown values. In this project, we aim to use Bayesian approaches for probabilistic forecasting to predict the demand for products/services and accordingly determine a better decision whatever that may be, e.g., inventory of product, or optimal allocation of resources. We investigate the association between these two using real-world data.
This is suitable for Honors and Masters students.
Forecasting a slection of time series using integer programming and optimisation methods
Time series often come in the form of many related time series known as hierarchical or group time series. Hierarchical time series refers to a collection of time series that have a natural and structural connection, e.g, time series are gathered across different locations such as sales across different stores and states in a country. This project aims to use optimisation methods like integer programming in the setting of hierarchical time series to forecast the entire hierarchy optimally.
This is suitable for Honors or Masters students.
Abolghasemi, Mahdi (2023). The intersection of machine learning with forecasting and optimisation: theory and applications. Palgrave Advances in the Economics of Innovation and Technology. (pp. 313-339) Cham, Switzerland: Springer. doi: 10.1007/978-3-031-35879-1_12
Abolghasemi, Mahdi (2023). Book review. International Journal of Forecasting, 39 (4), 1949-1951. doi: 10.1016/j.ijforecast.2022.12.002
Machine learning for satisficing operational decision making: a case study in blood supply chain
Abolghasemi, Mahdi, Abbasi, Babak and HosseiniFard, Zahra (2023). Machine learning for satisficing operational decision making: a case study in blood supply chain. International Journal of Forecasting. doi: 10.1016/j.ijforecast.2023.05.004
Abolghasemi, Mahdi, Tarr, Garth and Bergmeir, Christoph (2022). Machine learning applications in hierarchical time series forecasting: Investigating the impact of promotions. International Journal of Forecasting, 40 (2), 597-615. doi: 10.1016/j.ijforecast.2022.07.004
Abolghasemi, Mahdi, Rostami-Tabar, Bahman and Syntetos, Aris (2022). The value of point of sales information in upstream supply chain forecasting: an empirical investigation. International Journal of Production Research, 61 (7), 1-16. doi: 10.1080/00207543.2022.2063086
Model selection in reconciling hierarchical time series
Abolghasemi, Mahdi, Hyndman, Rob J., Spiliotis, Evangelos and Bergmeir, Christoph (2022). Model selection in reconciling hierarchical time series. Machine Learning, 111 (2), 739-789. doi: 10.1007/s10994-021-06126-z
Hierarchical forecast reconciliation with machine learning
Spiliotis, Evangelos, Abolghasemi, Mahdi, Hyndman, Rob J., Petropoulos, Fotios and Assimakopoulos, Vassilios (2021). Hierarchical forecast reconciliation with machine learning. Applied Soft Computing, 112 107756, 107756. doi: 10.1016/j.asoc.2021.107756
Demand forecasting in the presence of systematic events: cases in capturing sales promotions
Abolghasemi, Mahdi, Hurley, Jason, Eshragh, Ali and Fahimnia, Behnam (2020). Demand forecasting in the presence of systematic events: cases in capturing sales promotions. International Journal of Production Economics, 230. doi: 10.1016/j.ijpe.2020.107892
Considering pricing and uncertainty in designing a reverse logistics network
Zamani, Mohsen, Abolghasemi, Mahdi, Hosseini, Seyed Mohammad Seyed and Pishvaee, Mir Saman (2020). Considering pricing and uncertainty in designing a reverse logistics network. International Journal of Industrial and Systems Engineering, 35 (2), 158-182. doi: 10.1504/IJISE.2020.107554
Demand forecasting in supply chain: the impact of demand volatility in the presence of promotion
Abolghasemi, Mahdi, Beh, Eric, Tarr, Garth and Gerlach, Richard (2020). Demand forecasting in supply chain: the impact of demand volatility in the presence of promotion. Computers & Industrial Engineering, 142 106380, 106380. doi: 10.1016/j.cie.2020.106380
Alizadeh, Reza, Lund, Peter D., Beynaghi, Ali, Abolghasemi, Mandi and Maknoon, Reza (2016). An integrated scenario-based robust planning approach for foresight and strategic management with application to energy industry. Technological Forecasting and Social Change, 104, 162-171. doi: 10.1016/j.techfore.2015.11.030
A new approach for supply chain risk management: mapping SCOR into Bayesian network
Abolghasemi, Mahdi, Khodakarami, Vahid and Tehranifard, Hamid (2015). A new approach for supply chain risk management: mapping SCOR into Bayesian network. Journal of Industrial Engineering and Management, 8 (1), 280-302. doi: 10.3926/jiem.1281
A Bayesian framework for strategic management in energy industry
Abolghasemi, Mahdi and Alizadeh, Reza (2014). A Bayesian framework for strategic management in energy industry. International Journal of Science, Engineering and Technology, 3 (11), 1360-1366.
Approximating solutions to the Knapsack Problem using the Lagrangian Dual Framework
Keegan, Mitchell and Abolghasemi, Mahdi (2023). Approximating solutions to the Knapsack Problem using the Lagrangian Dual Framework. 36th Australasian Joint Conference on Artificial Intelligence, AI 2023, Brisbane, QLD, Australia, 28 November - 1 December 2023. Singapore, Singapore: Springer Nature Singapore. doi: 10.1007/978-981-99-8388-9_37
Hamza, M., Abolghasemi, M. and Alvandi, A.O. (2021). Forecasting sales with Bayesian networks: a case study of a supermarket product in the presence of promotions. Modeling and Simulation Conference (MODSIM), Sydney, Australia, 5-10 December 2021. Modelling and Simulation Society of Australia and New Zealand. doi: 10.36334/modsim.2021.M9.hamza
Measuring down stream supply chain performance using Bayesian network
Abolghasemi, Mahdi and Khodakarami, Vahid (2014). Measuring down stream supply chain performance using Bayesian network. 44th International Conference of Computer and Industrial Engineering, Istanbul, Turkey, 14-16 October 2014. Los Angeles, CA, United States: Computers and Industrial Engineering.
Analytics for the Australian Grains Industry (AAGI)
(2023–2027) Grains Research & Development Corporation
Predict and optimise with machine learning
Doctor Philosophy — Principal Advisor
Other advisors:
Forecasting and optimising decisions with machine learing
Master Philosophy — Principal Advisor
Other advisors:
A Review of Quarry Mine Datasets in Queensland
Doctor Philosophy — Associate Advisor
Other advisors:
Note for students: The possible research projects listed on this page may not be comprehensive or up to date. Always feel free to contact the staff for more information, and also with your own research ideas.
Statistical and machine learning applications for probabilistic energy forecasting
We aim to improve our understanding of Renewable Energy Sources (RES, wind and solar) over two timescales: 1) forecasting generation from hours to months, and 2) studying past and future climates to quantify projected changes in RES generation. Use physical models of the atmosphere (Numerical Weather Prediction and Climate Modelling) alongside historical observed data and new-generation data to drive the development of new forecasting methods and energy climate datasets. The forecasting component will develop seamless probabilistic RES forecasts with time horizons from hours to months using a blend of statistical and machine learning methods. Climate projections will develop high spatial and temporal (hourly) resolution RES data for onshore wind, offshore wind and PV from 2020-2050, using historical system data and ensembles of recent atmospheric model data, coupled with machine learning algorithms. This data will be used to investigate generation for different climate scenarios across RES alternatives. Statistical methods can rely on frequentist or Bayesian methods to help generate outcomes over short- and long-time scales, while machine learning techniques can range from simple decision trees to complex deep learning approaches.
This is an industry-supported project. It is suitable for Honors, and PhD students.
Forecasting sales and optimising inventory in food supply chains via machine learning
Probabilistic forecasting associates a probability of occurrence with the predicted values, making it a useful technique for determining decisions based on the level of risk one can take. It is a powerful technique that unlike point forecast gives you a complete view of the future unknown values. In this project, we aim to use Bayesian approaches for probabilistic forecasting to predict the demand for products/services and accordingly determine a better decision whatever that may be, e.g., inventory of product, or optimal allocation of resources. We investigate the association between these two using real-world data.
This is suitable for Honors and Masters students.
Forecasting a slection of time series using integer programming and optimisation methods
Time series often come in the form of many related time series known as hierarchical or group time series. Hierarchical time series refers to a collection of time series that have a natural and structural connection, e.g, time series are gathered across different locations such as sales across different stores and states in a country. This project aims to use optimisation methods like integer programming in the setting of hierarchical time series to forecast the entire hierarchy optimally.
This is suitable for Honors or Masters students.
Predicting cyber-attacks using Machine learning
Outlier detection is an important problem in many fields including in time series forecasting. Applications include detecting large spikes in transactions, or security breaches. This research project explores the application of machine learning techniques in the fields of cybersecurity forecasting and anomaly detection. With the ever-growing sophistication of cyber threats, traditional security measures are often insufficient to protect systems and networks effectively. By leveraging machine learning algorithms, this study aims to develop accurate and efficient models for predicting cyber attacks and identifying anomalous behaviour. The project involves analyzing large datasets of historical cybersecurity incidents, extracting relevant features, and training models to recognize patterns indicative of malicious activities. The findings of this research have the potential to enhance proactive cybersecurity measures and bolster defence mechanisms against evolving cyber threats.
This is suitable for Honors and Masters students.
Wind Power Forecasting
This project aims to develop a forecasting model for wind farms. We will use Machine Learning algorithms to predict power. The successful applicant will work on developing an R package for wind power forecasting.
This is suitable for Honors and Master students.
Decision Support systems for Forecasting and Decision Making
This project aims to develop advanced forecasting models using machine learning, and advanced decision-making models using mixed integer programming for various problems in the agriculture supply chain, retail, and energy sectors.
This is suitable for PhD and Masters students.