Gustavo de Almeida

I have been working on data-driven analysis applied to continuous industrial systems—both in simulated environments and, more significantly, in real-world scenarios—using statistical methods and AI/ML/DL techniques. These applications primarily span the pulp and paper sector, as well as cement, steel, and oil and gas, among others. They encompass tasks such as scenario (what-if) analysis, sensitivity analysis, virtual sensing, process improvement, monitoring and optimization, as well as abnormal situation management, fault detection, and operational diagnostics. Such applications combine data-centric analysis, model interpretability, and human-in-the-loop to develop more effective AI models. The ultimate goal is to achieve more reliable, efficient, and sustainable industrial operations.

I am also involved in the development of online courses for higher education in engineering. In the field of research, I have a particular interest in "Learning Analytics", utilizing statistical methods and AI/ML/DL/LLM techniques. One example of this is the development of ways to automatically evaluate and provide feedback on student responses to open-ended questions, such as short essays.