Welcome to my page -- I am Charalampos P. Andriotis, Assistant Professor of Artificial Intelligence in Structural Design & Mechanics at TU Delft, co-directing AiDAPT, the AI-Lab for Design, Analysis, and Optimization in the Faculty of Architecture & the Built Environment. My group's research is on decision-making under uncertainty for life-cycle control of engineering systems and infrastructure. Within this line of research, we focus on prediction, inference, learning, and optimization methodologies supporting resilient and autonomous planning against deterioration and hazards. Computational and theoretical themes that underpin this research include structural mechanics, uncertainty quantification, systems risk and reliability, and stochastic optimal control, especially as these emerge at the confluence of structural engineering and data science. Applications of interest to our research lie in the broader areas of performance-based engineering; stochastic modeling of structures subject to multiple stressors; life-cycle assessment and optimization; risk mitigation planning; value of information quantification; and infrastructure resilience. Shaping novel algorithmic approaches in AI and machine learning, our agenda is motivated by bridging physics-based engineering with data-driven intelligence, towards new paradigms, able to meet the complex decision-making challenges posed by the dynamically changing, expanding, and intertwining urban, natural, and digital environments, at large, real-world scales.

Recent News

  • Applications are invited for two PhD positions in the areas of AI-driven decision-support analytics for (i) structural life-cycle assessment and optimization and (ii) adaptive infrastructure management against multiple stressors, read more - Feb 24, 2021

  • TU Delft announces our new Artificial Intelligence Lab, AiDAPT: The AI-Lab for Design, Analysis, and Optimization in Architecture & the Built Environment, read more - Feb 01, 2021

  • Excited to soon be joining the Structural Design & Mechanics Group in the Faculty of Architecture & the Built Environment, at TU Delft, read more - Jan 15, 2021

  • ​Preprint published on coupling of dynamic Bayesian networks and Markov decision processes for planning against deterioration, with P.G. Morato, K.G. Papakonstantinou, N.J. Nielsen and P. Rigo, entitled "Optimal inspection and maintenance planning for deteriorating structures through dynamic Bayesian networks and Markov decision processes", read more - Sept 18, 2020

  • Story available on PSU College of Engineering website, entitled "How to train your infrastructure", covering developments on our deep reinforcement learning project for infrastructure life-cycle control, read more - Aug 8, 2020

  • Preprint published on constrained partially observable Markov decision processes and decentralized deep reinforcement learning for planning of large engineering systems, with K.G. Papakonstantinou, entitled "Deep reinforcement learning driven inspection and maintenance planning under incomplete information and constraints", read more - Jul 15, 2020

  • Preprint published on the mathematical foundations of the value-of-information of structural health data in the context stochastic optimal control under partial observability, with K.G. Papakonstantinou and E.N. Chatzi, entitled "Value of structural health information in partially observable stochastic environments", read more - Jul 07, 2020

Contact

Faculty of Architecture & the Built Environment

Delft University of Technology

Julianalaan 134, 2628 BL, Delft 

email: c.andriotis [at] tudelft [dot] nl

Copyright © 2020-21 by C.P. Andriotis