English version

Postdoctoral researcher in uncertainty quantification for finite element analysis

(Ref. BAP-2017-68)

Tewerkstelling : Voltijds
Duur : Bepaalde duur verlengbaar
Plaats : Leuven
Solliciteren tot en met : 28/02/2017
For the Department of Mechanical Engineering, we are looking for a postdoctoral researcher in the domain of uncertainty quantification for finite element analysis. The position fits in our on-going activity on virtual product performance assessment under uncertainty, and specifically focuses on the analysis of spatial uncertainty in geometric model properties.

Postdoctoral researcher in uncertainty quantification for finite element analysis

Within the Noise & Vibration research group, the research unit hosting this position focuses on reliability, optimisation and robustness studies through the use of non-deterministic finite element modelling techniques. Based on this approach, we do not limit the design study to a single nominal case, but enable the assessment of the design quality in real-life conditions including all possible variability stemming from manufacturing or functional effects acting on the design under study. We do so by including variability and uncertainty in all modelling aspects that are not precisely known, as e.g. variable loading conditions, uncertain boundary conditions, geometrical tolerances, material inhomogeneity, manufacturing variability… To this extent, we use state-of-the-art reliability assessment software, among which our own in-house developed code for uncertainty inclusion in numerical models.

We have several research lines in our group. Historically, we are very active in the development of new techniques for the assessment of uncertainty and variability in finite element models, with activities on interval methods as well as spatial variability analysis. More recent activity in this respect concerns the identification and quantification of the model variability from experimental observations in a model validation framework. Apart from those more fundamental developments, our research group is also very active in the translation of these methods to engineering practice by integrating them in industrial engineering activities. We do so by ensuring that our methods interact with off-the-shelve FE software in a non-intrusive sense. In order to make sure our research answers actual needs of mechanical designers, we also work on practical engineering problems in close cooperation with several end-users.


The project objective for this position is to study the effect of manufacturing process variability on the product level in a virtual modelling context. Starting from a limited set of experimental observations of actual realisations of the product, the objective is (a) to quantify the spatial geometric variability over the product in a non-deterministic framework, (b) to project this information on the corresponding FE model, and (c) to propagate this variability to non-deterministic analysis results.

As postdoctoral researcher, you will guide and support following tasks within the project:
  • the translation of the experimental data to an appropriate non-deterministic spatial concept (e.g. random fields using Bayesian updating)
  • the integration of this model in a non-deterministic FE analysis tool (own in-house code)
  • the propagation to and validation of the non-deterministic analysis results


A successful candidate should have:
  • a master degree in engineering
  • a PhD in a domain relevant for this position
  • strong experience with finite element modelling

Ideally, the candidate also has good credentials on at least one of the following topics:

  • non-deterministic finite element modelling
  • uncertainty quantification techniques (Bayesian analysis)
  • random fields


We offer a six month contract with possibility of extension after positive evaluation.


For more information please contact Prof. dr. ir. David Moens, tel.: +32 16 37 28 79, mail: david.moens@kuleuven.be.

You can apply for this job no later than February 28, 2017 via the

Downloads : pdf