top of page

Decision-making optimization 

Engineered and natural environments are exposed to adverse operating conditions during their life-cycle, with deterioration and hazards posing a constant threat to their reliability and resilience. Optimal decision-making for such systems is concerned with controlling the extent and consequences of this exposure through scheduling of efficient strategies, from preventive planning to post-disaster recovery, with the objective to minimize the involved socioeconomic and environmental risks. To be successful in this objective, decision optimization needs to holistically alleviate multi-faceted complexities arising, among others, by the curse of dimensionality related to large-scale multi-component systems; the curse of history related to long-term sequential decisions; the model and data uncertainties; the operational constraints; the presence of multiple agents and decision layers;  and, finally, the sheer connection of physics-based engineering models with the optimization process. This line of research studies solutions to the above challenges through novel conceptual and computational frameworks within the contexts of stochastic optimization, systems reliability & control, and artificial intelligence.


Dynamic decisions for deteriorating systems through Partially Observable Markov Decision Processes

Integrating POMDPs with Deep Reinforcement Learning for planning in high-dimensional environments

Inspection & maintenance policies subject to stochastic resource constraints and long-term reliability goals

Assessing the Value of Information for decision-making in partially observable stochastic environments 



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

bottom of page