ESR1 University of Strathclyde
Project title: Evidence-Based Robust Optimisation for Large Scale Evolvable Processes

Gianluca Filippi
Objectives: To develop a unified computational framework for Optimisation Under Uncertainty with uncertainty represented with Imprecise Probability Theory; To use expert judgement to quantify epistemic uncertainty and develop reliable elicitation and information fusion methods; To apply the unified framework to the optimisation of large scale evolvable processes in the context of MMBCSE.
Expected Results: Demonstration that the unified computational approach can consistently incorporate uncertainty quantifications based on different Imprecise Probability Theories. Computationally efficient algorithm for the treatment of many-objective optimisation of systems and processes under mixed aleatoric and epistemic uncertainty in model and parameters. Demonstration of the benefits of IP in the optimisation under uncertainty of aerospace systems. Demonstration of the advantage of these methodology in the design and control of evolvable processes.
Planned Secondments: JSI (M21-23) to work on multi-level optimisation for EBRO and RRBMDO within WP3.1 and 3.3, ESTECO (M31-33) to integrate the methodology and computation framework in the context of multidisciplinary model-based collaborative system engineering within WP1.1.

This position will be supervised by Prof Massimiliano Vasile. There will be 3 co-supervisors: Dr Akartunali, Dr Revie and Dr Minisci.

ESR2 University of Strathclyde
Project title: Optimal Multi-sensor Multi-target Tracking of Aerospace Systems Under Uncertainty

Cristian Greco
Objectives: To develop new approaches for the tracking from ground of space objects during re-entry or the tracking from space of vehicles on Earth; To develop enhanced-HDMR techniques to propagate uncertainty through complex time-varying processes; To optimise the use of observations from heterogeneous sources (FFS, ground stations) to reduce estimation uncertainty and maximise detection and identification probability; To develop efficient strategies for optimal resource allocation and scheduling under uncertainties.
Expected Results: Implementation of new numerical algorithms for the minimisation of the uncertainties and costs of tracking and monitoring systems for aeronautical and space applications. Test of the implemented tracking and monitoring system on two industrial cases defined in WP1.2.
Planned Secondments: THK (M15-17) to work on many objective MINLP methods for resource allocation and optimal observation scheduling, VKI (M35-39) to work on uncertainty characterisation during debris re-entry.

This position will be supervised by Dr Christie Maddock. There will be 3 co-supervisors: Dr Akartunali, Dr Revie and Dr Minisci.

Project title: Towards a New Paradigm for the Control and the Analysis of Experiment Data.

Giulio Gori
Objectives: To develop a unified computational framework for considering at the same time systematic, model and experimental uncertainty based on a Bayesian framework; To understand how using experiments for improving prediction of the numerical simulation; To use numerical simulation for improving the experiment in terms of physical modelling and optimal operating conditions; To perform UQ experiments by imposing some controlled experimental variability for checking accuracy of the computational framework.
Expected Results: Improve drastically the interaction between experiment and numerical simulation, in order to merge these different sources of information for a better understanding of a given phenomenon. Use of an efficient procedure for determining physical model efficiency with respect to the experimental data. Formulation of an error estimation associated to both experimental measurements and numerical simulation. This could provide an indication about the necessity of new experiments or more accurate simulation.
Planned Secondments: VKI (M15-20) to apply the developed methods to experiments of Longshot within WP2.3, Po-liMi (M27-29) to apply the UQ framework to anti icing within WP2.3.

This position will be supervised by Dr Pietro Congedo and Dr Beaugendre.

Project title: Uncertainty Quantification Framework for Large-scale Unsteady Problems.

Joao Reis
Objectives: To improve Anchored-ANOVA methods for very large dimension problems (up to 500 uncertainties); To formulate and develop highly scalable algorithms that can be implemented on multiprocessor supercomputers architectures; To develop a strategy control for taking into account the unsteady character; To include the model discrepancy by using a Bayesian framework; To build a HPC framework for treating large-dimension unsteady problems at a moderate computational cost; To develop robust optimisation methods with an adaptive refinement on the coupled optimisation/stochastic space.
Expected Results: Development of an efficient parallel strategy permitting the treatment problems featuring high non-linearities where models with different fidelity can provide very large differences in the stochastic space. An innovative procedure for optimizing unsteady systems where some uncertainties can vary in time. Application to anti icing and morphing systems.
Planned Secondments: UoS (M15-17) to work on the HPC framework for treating large-dimension unsteady problems, PoliMi (M32-34) to work on the application of UQ to morphing within 2.2. .

This position will be supervised by Dr Pietro Congedo and Dr Beaugendre. ESR4 will be co-supervised by Prof Luc Giraud.

Project title: Optimal Energy-Driven Aircraft Design Under Uncertainty

Elisa Morales
Objectives: To build a multi-fidelity model for energy driven aircraft design; To develop an MDO loop for robust reliable design of aircraft with reduced drag and minimal fuel consumption; To develop an integrated design tool capable of integrating uncertainty quantification, multi-criteria optimisation and advanced CFD simulations; To develop reliability-based design optimisation algorithms based on the Generalized Inverse Distribution Functions.
Expected Results: Multi-fidelity model and model reduction specialised for energy-driven aircraft design. Extension of the SPOT framework to handle the optimal design of energy-driven aircraft under uncertainty. Reliability design optimisation of energy-driven aircraft through the Generalised Inverse Distribution Function approach.
Planned Secondments: DLR (M25-27) to work on reliability design of energy-driven aircraft within WP3.5, THK (M31-33) to work on many-objective optimisation techniques for expensive problems within WP3.2.

This position will be supervised by two senior scientists belonging to the Fluid Dynamics Laboratory: Dr Quagliarella and Dr Catalano.
The ESR will be given the opportunity to enrol in the PhD programme at the University Federico II in Naples, with whom CIRA has a permanent agreement, and will be supervised there by Prof Tognaccini and Prof De Nicola

ESR6 Von Karman Institute
Project title: Uncertainty treatment in aerothermodynamic experimental databases

Anabel del Val
Objectives: To study and develop uncertainty quantification methodologies for the assessment of the aerothermodynamic database of an ablative TPS; To develop surrogate models for uncertainty quantification applicable to both experimental and computational data sources for the development of reliable aerothermodynamic databases for ablative TPS; To understand how to use experiments for improving the predictions of numerical simulations.
Expected Results: Evaluation of different approaches to uncertainty quantification for aerothermodynamic experimental and numerical databases. A methodology for uncertainty quantification of databases based on the previous evaluation work. An experimental validation benchmark test case set and related experimental prescriptions. An flight prediction scenario to rebuild free stream conditions of spacecraft from measurement sensors mounted in TPS.
Planned Secondments: INRIA (M12-21) to work on surrogate models and Bayesian methods for model validation and prediction, Durham (M31-34) to work on robust Bayesian methods for experimental analysis.

This positions will be supervised by Prof Thierry Magin and co-supervised by Prof Olivier Chazot.

Project title: Reliability-Based Optimisation for Multidisciplinary Model-based Collaborative System Engineering

Péter Zénó Korondi
Objectives: To define and improve collaborative engineering environments and multidisciplinary collaborative optimisation strategies, for distributed systems; To study and implement strategies for multidisciplinary collaborative P2P System Engineering and optimisation under uncertainty; To find an effective treatment of large-scale problems, by finding a suitable way to handle the interactions among different subsystems; To define proper reliability measures for the multi-disciplinary optimisation problems must be defined; To extend modelling languages beyond the state-of-the-art , in order to match the situations of interest for this project; To define a methodology for reliability-based design optimisation in a multidisciplinary model-based collaborative framework and with man-in-the-loop activities.
Expected Results: An integrated multidisciplinary model-based collaborative distributed environment. The extension of MMBCSE paradigms to large scale problems under uncertainty.
Planned Secondments: CIRA (M37-39) to work on optimal energy-driven aircraft design under uncertainty within WP1.1, JSI (M21-23) to work on multi-level optimisation techniques for reliability based optimisation within WP3.1.

This position will be supervised by Prof Poloni, and co-supervised by Dr Marchi and Dr Kavka from ESTECO and by Prof Pediroda from the University of Trieste, where the PhD will be awarded.

ESR8 University of Durham
Project title: Prediction of System Reliability during Design Phases

Daniel Krpelik
Objectives: To develop suitable theory of system reliability quantification, using imprecise probabilities, in order to reflect carefully the uncertainties involved in this process at different stages; To derive an approximation of the lower and upper probabilities of system functionalities; To upscale to the propagation of upper and lower previsions to large systems; To study a representation of uncertainty in multi-phased design of aerospace systems.
Expected Results: A theoretical and computational framework for system reliability quantification in multi-phase processes using Imprecise Probability Theory. A theoretical and computational framework for robust optimisation and decision making in multi-phase processes. A demonstrative example of application to the life cycle assessment of a launcher. New methods for system reliability quantification at different stages of system design, reflecting indeterminacy in the specification of the required functionality and providing the opportunity to focus on robustness with regard to resilience of the system. New computational methods, including the use of approximations, to enable upscaling of recently present-ed theory of imprecise probabilities for system reliability to large real-world multi-phase processes.
Planned Secondments: SU (M27-29) to work on the application of imprecise probabilities and expert elicitation to the end-to-end design of space systems within WP3.3 and WP3.5, ESTECO (M37-39) to work on the application of the pro-posed methodology to multidisciplinary model-based collaborative system engineering within WP3.4 and 3.5.

This position will be supervised by Prof Frank Coolen and co-supervised by Dr Louis Aslett.

ESR9 University of Durham
Project title: Large Scale Simulation for Quantifying Severe Uncertainty with Imprecise Probabilities

Tathagata Basu
Objectives: To investigate algorithms and methods for simulation with Robust Bayesian models; To investigate how standard statistical simulation approaches, such as for instance Markov chain Monte Carlo, can be extended to do inference from sets of prior distributions, in a way that leads to computationally efficient yet still reliable inference about the actual risks in the system; To derive theoretical results from small scale tests; To upscale the dimensionality of the application, depending on the results of the small scale tests.
Expected Results: Enabling imprecise methods to be applied to much larger problems than is currently possible. An efficient computational framework to deal with sets of prior distributions in Robust Bayesian inference. New efficient statistical simulation techniques for imprecise probability. Improved representation of model uncertainty in simulation models, leading to better risk-informed decisions.
Planned Secondments: NPL (M19-21) to work on the treatment of experimental data and model validation within WP2.3, VKI (M28-30) to work on high-dimensional uncertainty propagation and the treatment of experimental data within WP2.2 and 2.3.

This position will be supervised by Dr Jochen Einbeck and co-supervised Dr Matthias Troffaes.

ESR10 Politecnico di Milano
Project title: Uncertainty Characterisation in Multi-fidelity Anti-ice System and Design

Bárbara Arizmendi
Objectives: To develop and analyse diverse models of variable complexity to determine the aerodynamic flow-field, the droplet trajectories and the ice accretion over an aircraft equipped with an anti-ice system; To characterise both aleatory and epistemic uncertainties in the above multi-fidelity models; To produce the optimised design of a robust anti-ice system for fixed- and rotary-wing aircraft;
Expected Results: A set of multi-fidelity icing models verified and validated against experimental data. A Bayesian-based techniques for model and experimental uncertainty of an aircraft anti-icing systems. The optimisation under uncertainty of an anti-ice systems on complex aircraft configurations. The assessment, against experimental data, of in-flight icing accretion model predictions.
Planned Secondments: Alenia (M15-21) to work on multifidelity models of anti-ice systems, INRIA (M25-27) to de-velop and implement Bayesian-based techniques for handling uncertainty in model and experimental measurements.

This position will be supervised by Prof Guardone and co-supervised by Dr Quaranta.

ESR11 Jožef Stefan Institute
Project title: Efficient Computational Methods for Worst-case and Multi-level Optimisation

Margarita Antoniou
Objectives: To investigate efficient methods and algorithms for worst-case and multi-level optimisation; To develop algorithms that are optimal on expensive problems in Optimisation Under Uncertainty; To test these methods in the applications developed in WP3.4 and 3.5.
Expected Results: A new set of optimisation algorithms, for the efficient solution of multi-level problems. Implementation of a new set of problem instances of multi-level optimisation. Evaluation framework with defined measures and classifying definitions targeting problems of robust optimisation and reliability-based optimisation.
Planned Secondments: ESTECO (M21-23) to work on the use of the proposed optimisation techniques to reliability based optimisation within WP3.4, Ghent (M33-35) to work on the coupling of the optimisation techniques with uncertainty propagation through dynamical systems within WP3.5.

This position will be supervised Dr Gregor Papa and Dr Korošec.

ESR12 Technische Hochschule Köln
Project title: Robust Mixed-integer Optimisation in Uncertain Environments

Lorenzo Gentile
Objectives: To develop and study evolutionary optimisation methods with regard to mixed-integer problems involving uncertainty; To define a procedure that can efficiently explore large parameter spaces in robust design; To integrate the identified methods in the sequential parameter optimisation toolbox (SPOT); To analyse and develop suitable optimisation methods and parameter settings for efficient optimisation of constraint robust mixed-integer design problems.
Expected Results: Evaluation of different approaches to uncertainty quantification for mixed-integer optimisation problems. Algorithms for the solution of nonlinear mixed-integer problems under uncertainty. Integration into SPOT frame-work of a simplified simulator, a constraint nonlinear mixed-integer optimisation problem, is one expected result. Experimental and theoretical analysis of performance.
Planned Secondments: SU (M31-33) to apply MINLP to optimal multi-target tracking with FSS within WP3.2, DLR (M37-39) to apply MINLP to a drag reduction problem within WP3.2.

This position will be supervised by Prof. Dr. Thomas Beielstein and Prof Naujoks.

ESR13 Technische Hochschule Köln
Project title: High Efficiency Many Objective Evolutionary Optimisation Methods

Dani Irawan
Objectives: To build an effective many-objective optimisation loop for robust optimisation; To identify promising techniques for handling many-objective optimisation tasks for aerospace applications; To identify a framework for integration of techniques for many-objective optimisation of large scale expensive problems; To integrate such methods for large scale robust design optimisation problems from WP3.2 with UQ techniques; To develop a simplified simulator to accelerate the testing/tuning loop in collaboration with CIRA providing simplified simulators based on RANS CFD solvers but using coarser grids etc.; To test the considered techniques on the design of aerospace transportation systems, optimal energy-driven aircraft design and RLV design; Identify most promising approaches capable of handling alternative applications from the project‟s field.
Expected Results: Algorithms for the solution of many-objective expensive problems under uncertainty. Theoretical and experimental analysis of the structure of the problem and performance of the algorithms. Extension of the SPOT frame-work to handle such many-objective, expensive, constrained nonlinear problems. Development and incorporation of a simplified (low-fidelity) simulator as an ideal test case for statistical analysis.
Planned Secondments: Airbus Gmbh (M15-17) and CIRA (M31-33) to work on the application of many-objective evolutionary optimisation techniques to the optimal energy-driven aircraft design under uncertainty within WP3.2.

This position will be supervised by Prof. Dr. Thomas Beielstein and Prof Boris Naujoks.

ESR14 Ghent University
Project title: Imprecise (Hidden) Markov Chains, Dynamical Properties and Inference Algorithm

Thomas Krak
Objectives: To develop imprecise probability methods for estimating imprecise transition (and emission) probabilities for (hidden) Markov models; To study the dynamical properties of (hidden) Markov models; To study and development of algorithms for making inferences in imprecise hidden Markov models; To apply the proposed methodology to the estimation of states in a process or dynamical system from outputs, and on estimating state sequences from outputs sequences; To incorporate the proposed inference algorithm to the optimal control of a process in aerospace transportation.
Expected Results: An extension of the analysis of the ergodicity for imprecise Markov chains to deal with birth-death processes. A solution to the explicit form of the stationary distribution for (ergodic) imprecise Markov chains. An imprecise probabilistic counterparts for the many dynamic parameters of precise Markov chains, such as for instance mean transition and recurrence times, has immediate practical applicability.
Planned Secondments: SU (M15-17) to test the proposed inference algorithm on multisensory tracking and vehicle tracking with FSS, Durham (M25-27) to work on Evolvable Optimisation Under Uncertainty within WP3.5.

This position will be supervised by Prof Dr Gert de Cooman and Dr Jasper De Bock.

Project title: Robust design of a shock bump for natural laminar flow over a wing

Christian Sabater
Objectives: To use robust and reliability-based optimisation to design a shock bump which is less sensitive, to small random changes in onflow conditions (operational conditions) and geometrical uncertainties, like manufacturing tolerances or degradation, in order to maintain natural laminar flow (NLF); To develop a suitable parameterization of the bump in 2D and 3D; To investigate different measures of robustness suitable in the context of NLF; To define input PDFs for the uncertain parameters in shock bump and NLF models; To make use of efficient nonintrusive UQ methods for many uncertainties, including gradient-enhanced surrogate models and reduced-order models;
Expected Results: A demonstration of the efficiency and effectiveness of the robust design methodology on 2D (airfoil) cases and 3D wings in the transonic flow regime; A validated software framework for robust design;
Planned Secondments: INRIA (M9-12) on robust and reliability-based design optimisation algorithms; Airbus GmbH (M30-33) to apply the developed process.

This position will be supervised by Prof Dr Cord-Christian Rossow and co-supervised by Prof Dr Norbert Kroll and Dr Stefan Görtz.