About the speaker
Alessandro Hill is an Assistant Professor of Operations Research at California Polytechnic State University. He specializes in combinatorial optimization and applications to networks and scheduling. Alessandro holds a master's in mathematics and computer science from the University of Augsburg and a PhD in Business Administration from the Business School at the University of Hamburg. He has worked both in international research projects and as an industry professional applying operations research to various fields including telecommunications, automotive, petrochemical, logistics, mining, and education. His work has been published in journals such as the European Journal on Operational Research, INFORMS Journal on Computing, and Computers & Operations Research.
Abstract
We first review selected applications of project scheduling models to efficiently manage operations in practice. The main topic of this presentation is the study of a novel class of project scheduling problems that incorporate autonomous learning. In these models, certain jobs can be completed in a reduced amount of time if scheduled after jobs that lead to acquiring relevant experience. We consider single-predecessor learning and present corresponding learning mechanisms. We discuss the structure and complexity of these combinatorial problems and devise approximative and exact algorithms to solve them. In a second part, we extend these models by integrating limited resources. To tackle these NP-hard problems, we devise first constraint programming formulations. In a computational study, we show the potential scheduling benefits that can be obtained when integrating learning compared to classical project scheduling. For all the different models and methods discussed, we provide computational proof of their efficacy.