[Examiner] Dr. Abdulrahim Ali - Systematic Assessment of Machine Learning for building performance analysis

Congratulations, Dr. Abdulrahim Haroun Ali. On 04/04/2023 at 10h, I attended the Ph.D. defense of ABDULRAHIM ALI as an Examiner at the Department of Industrial and Systems Engineering, College of Engineering of Khalifa University. His dissertation title is: « Systematic Assessment of Machine Learning for building performance analysis: applications, synthesis, and way forward». The Ph.D. is supervised by Prof. Raja Jayaraman (Main Advisor) and Prof. Maher Maalouf (Co-Advisor) at Khalifa University.

The Ph.D. research by Dr. Abdulrahim Ali aims to evaluate the premise of Machine Learning techniques in diverse building performance applications and provide recommendations/guidelines on when to use Machine Learning or traditional statistical methods for different building performance applications and contexts.

Click here for key publications: https://lnkd.in/eJ9SMcuH

The Ph.D. evaluated parameters related to building operation (occupancy, equipment and lighting usage, and thermostat setpoints), which typically influence the building energy consumption and peak loads. Also, the dissertation compared and contrasted the capability and associated computation cost of several black-box methods (random forest, gradient boosting, extreme gradient boosting, and adaptive boosting) with a white-box method such as linear regression. The study uses a dataset representing the extreme climate conditions of the Gulf Cooperation Council countries; very few surrogate models have been developed for the region so far.

Some of the most commonly used machine learning algorithms used in the construction industry by scientists include (based on Ali's literature review):

✅ Logistic regression (60%)

✅ Decision trees (50%)

✅ Random forests (50%)

✅ Support vector machines (SVM) (40%)

✅ K-nearest neighbors (KNN) (40%)

✅ Artificial neural networks (ANN) (40%)

✅ Gradient Boosting (30%)

It's important to note that this list is not exhaustive. The popularity and frequency of use of different machine learning algorithms can vary depending on the industry, application, and data being analyzed. Additionally, machine learning is rapidly evolving, and new algorithms are constantly being developed and refined.

Many thanks for the invitation by Prof. Raja Jayaraman and Prof. Maher Maalouf. The jury committee member included Prof. Elie Azar Carleton University, Prof. Andrei Sleptchenko Khalifa University, Prof. Mecit Can Emre Simsekler, and Prof. Andreas Henschel Khalifa University.

📣 Read and share this 📄 info news and article for further details.

Subscribe to my 📰 newsletter: www.shadyattia.org

Look at my ☑️ recent posts' activities. https://lnkd.in/eR-jiXtR

#machinelearning ; #datasets ; #buildingperformance, #thermalcomfort , #supportvectormachine, #gradientboosting #algorithms #building #constructionindustry