Publications

Fact sheets

For a selection of publications and student projects, we have concise fact sheets available. These fact sheets are 2-page summaries that provide an accessible, high-level overview of the performed research.

Click here to navigate to the fact sheet overview.

List of peer-reviewed publications

D. Kerkkamp, Z. A. Bukhsh, Y. Zhang, and N. Jansen, “Grouping of maintenance actions with deep reinforcement learning and graph convolutional networks,” in International Conference on Agents and Artificial Intelligence (ICAART), 2022.
[BibTeX]

@inproceedings{Kerkkamp2021,
author = {David Kerkkamp and
Zaharah A. Bukhsh and
Yingqian Zhang and
Nils Jansen},
title = {Grouping of Maintenance Actions with Deep Reinforcement Learning and Graph Convolutional Networks},
booktitle = {{International Conference on Agents and Artificial Intelligence (ICAART)}},
year = {2022}
}

T. S. Badings, A. Abate, N. Jansen, D. Parker, H. A. Poonawala, and M. Stoelinga, “Sampling-based robust control of autonomous systems with non-gaussian noise,” in 36th AAAI Conference on Artificial Intelligence, 2022.
[BibTeX] [Download PDF]

@inproceedings{DBLP:journals/corr/abs-2110-12662,
author = {Thom S. Badings and
Alessandro Abate and
Nils Jansen and
David Parker and
Hasan A. Poonawala and
Mari{\"{e}}lle Stoelinga},
title = {Sampling-Based Robust Control of Autonomous Systems with Non-Gaussian
Noise},
booktitle = {{36th AAAI Conference on Artificial Intelligence}},
year = {2022},
url = {https://arxiv.org/pdf/2110.12662.pdf}
}

L. S. Keizers, R. Loendersloot, and T. Tinga, “Unscented kalman filtering for prognostics under varying operational and environmental conditions,” International journal of prognostics and health management, vol. 12, iss. 2, 2021.
[BibTeX] [Download PDF]

@article{keizers2021unscented,
title={Unscented Kalman Filtering for Prognostics Under Varying Operational and Environmental Conditions},
author={Keizers, Luc S. and Loendersloot, Richard and Tinga, Tiedo},
journal={International Journal of Prognostics and Health Management},
volume={12},
number={2},
year={2021},
url={http://www.papers.phmsociety.org/index.php/ijphm/article/view/2943}
}

Z. A. Bukhsh, N. Jansen, and A. Saeed, “Damage detection using in-domain and cross-domain transfer learning,” Neural computing and applications, 2021.
[BibTeX] [Download PDF]

@article{BukshJansenSaeed2021,
author = {Zaharah Allah Bukhsh and
Nils Jansen and
Aaqib Saeed},
title = {Damage detection using in-domain and cross-domain transfer learning},
journal = {Neural Computing and Applications},
year = {2021},
url = {https://doi.org/10.1007/s00521-021-06279-x}
}

T. S. Badings, A. Hartmanns, N. Jansen, and M. Suilen, “Balancing wind and batteries: towards predictive verification of smart grids,” in 13th NASA Formal Methods Symposium, 2021.
[BibTeX] [Download PDF]

@inproceedings{Badings2021NFM,
author = {Thom S. Badings and
Arnd Hartmanns and
Nils Jansen and
Marnix Suilen},
title = {Balancing Wind and Batteries: Towards Predictive Verification of Smart
Grids},
booktitle = {{13th NASA Formal Methods Symposium}},
year = {2021},
url={https://arxiv.org/abs/2101.12496}
}

T. D. Simão, N. Jansen, and M. T. J. Spaan, “Alwayssafe: reinforcement learning without safety constraint violations,” in 20th International Conference on Autonomous Agents and Multiagent Systems, 2021.
[BibTeX]

@inproceedings{simao-et-al-aamas-2021,
author = {Thiago D. Sim{\~{a}}o and Nils Jansen and Matthijs T. J. Spaan},
title = {AlwaysSafe: Reinforcement Learning without Safety Constraint Violations},
booktitle  = {{20th International Conference on Autonomous Agents and Multiagent Systems}},
year = {2021},
note = {to appear},
keywords = {reviewed}
}

M. Cubuktepe, N. Jansen, S. Junges, A. Marandi, M. Suilen, and U. Topcu, “Robust finite-state controllers for uncertain pomdps,” in 35th AAAI Conference on Artificial Intelligence, 2021.
[BibTeX]

@inproceedings{cubuktepe-et-al-aaai-2021,
author = {Murat Cubuktepe and
Nils Jansen and
Sebastian Junges and
Ahmadreza Marandi and
Marnix Suilen and
Ufuk Topcu},
title = {Robust Finite-State Controllers for Uncertain POMDPs},
booktitle  = {{35th AAAI Conference on Artificial Intelligence}},
year = {2021},
keywords = {reviewed}
}

B. Ton, R. Basten, J. Bolte, J. Braaksma, A. Di Bucchianico, P. van de Calseyde, F. Grooteman, T. Heskes, N. Jansen, W. Teeuw, T. Tinga, and M. Stoelinga, “Primavera: synergising predictive maintenance,” Applied sciences, vol. 10, iss. 23, 2020. doi:10.3390/app10238348
[BibTeX] [Abstract] [Download PDF]

The full potential of predictive maintenance has not yet been utilised. Current solutions focus on individual steps of the predictive maintenance cycle and only work for very specific settings. The overarching challenge of predictive maintenance is to leverage these individual building blocks to obtain a framework that supports optimal maintenance and asset management. The PrimaVera project has identified four obstacles to tackle in order to utilise predictive maintenance at its full potential: lack of orchestration and automation of the predictive maintenance workflow, inaccurate or incomplete data and the role of human and organisational factors in data-driven decision support tools. Furthermore, an intuitive generic applicable predictive maintenance process model is presented in this paper to provide a structured way of deploying predictive maintenance solutions.

@Article{app10238348,
AUTHOR = {Ton, Bram and Basten, Rob and Bolte, John and Braaksma, Jan and Di Bucchianico, Alessandro and van de Calseyde, Philippe and Grooteman, Frank and Heskes, Tom and Jansen, Nils and Teeuw, Wouter and Tinga, Tiedo and Stoelinga, Mariëlle},
TITLE = {PrimaVera: Synergising Predictive Maintenance},
JOURNAL = {Applied Sciences},
VOLUME = {10},
YEAR = {2020},
NUMBER = {23},
ARTICLE-NUMBER = {8348},
URL = {https://www.mdpi.com/2076-3417/10/23/8348},
ISSN = {2076-3417},
ABSTRACT = {The full potential of predictive maintenance has not yet been utilised. Current solutions focus on individual steps of the predictive maintenance cycle and only work for very specific settings. The overarching challenge of predictive maintenance is to leverage these individual building blocks to obtain a framework that supports optimal maintenance and asset management. The PrimaVera project has identified four obstacles to tackle in order to utilise predictive maintenance at its full potential: lack of orchestration and automation of the predictive maintenance workflow, inaccurate or incomplete data and the role of human and organisational factors in data-driven decision support tools. Furthermore, an intuitive generic applicable predictive maintenance process model is presented in this paper to provide a structured way of deploying predictive maintenance solutions.},
DOI = {10.3390/app10238348}
}

{. A. }. Jimenez-Roa, T. Heskes, and M. Stoelinga, “Fault trees, decision trees, and binary decision diagrams: a systematic comparison,” in Proceedings of the 31st european safety and reliability conference (esrel 2021), 2021, p. 673–680. doi:10.3850/978-981-18-2016-8_241-cd
[BibTeX] [Abstract] [Download PDF]

In reliability engineering, we need to understand system dependencies, cause-effect relations, identify critical components, and analyze how they trigger failures. Three prominent graph models commonly used for these purposes are fault trees (FTs), decision trees (DTs), and binary decision diagrams (BDDs). These models are popular because they are easy to interpret, serve as a communication tool between stakeholders of various backgrounds, and support decision-making processes. Moreover, these models help to understand real-world problems by computing reliability metrics, minimum cut sets, logic rules, and displaying dependencies. Nevertheless, it is unclear how these graph models compare. Thus, the goal of this paper is to understand the similarities and differences through a systematic comparison based on their (i) purpose and application, (ii) structural representation, (iii) analysis methods, (iv) construction, and (v) benefits & limitations. Furthermore, we use a running example based on a Container Seal Design to showcase the models in practice. Our results show that, given that FTs, DTs, and BDDs have different purposes and application domains, they adopt different structural representations and analysis methodologies that entail a variety of benefits and limitations, the latter can be addressed via conversion methods or extensions. Specific remarks are that BDDs can be considered as a compact representation of binary DTs, since the former allows sub-node sharing, which makes BDDs more efficient at representing logical rules than binary DTs. It is possible to obtain cut sets from BDDs and DTs and construct a FT using the (con/dis)junctive normal form, although this may result in a sub-optimal FT structure.

@inproceedings{JimenezRoa2021_ESREL,
title = "Fault Trees, Decision Trees, and Binary Decision Diagrams: A systematic comparison",
abstract = "In reliability engineering, we need to understand system dependencies, cause-effect relations, identify critical components, and analyze how they trigger failures. Three prominent graph models commonly used for these purposes are fault trees (FTs), decision trees (DTs), and binary decision diagrams (BDDs). These models are popular because they are easy to interpret, serve as a communication tool between stakeholders of various backgrounds, and support decision-making processes. Moreover, these models help to understand real-world problems by computing reliability metrics, minimum cut sets, logic rules, and displaying dependencies. Nevertheless, it is unclear how these graph models compare. Thus, the goal of this paper is to understand the similarities and differences through a systematic comparison based on their (i) purpose and application, (ii) structural representation, (iii) analysis methods, (iv) construction, and (v) benefits & limitations. Furthermore, we use a running example based on a Container Seal Design to showcase the models in practice. Our results show that, given that FTs, DTs, and BDDs have different purposes and application domains, they adopt different structural representations and analysis methodologies that entail a variety of benefits and limitations, the latter can be addressed via conversion methods or extensions. Specific remarks are that BDDs can be considered as a compact representation of binary DTs, since the former allows sub-node sharing, which makes BDDs more efficient at representing logical rules than binary DTs. It is possible to obtain cut sets from BDDs and DTs and construct a FT using the (con/dis)junctive normal form, although this may result in a sub-optimal FT structure.",
keywords = "Fault tree analysis, Decision tree, Binary decision diagrams, Systematic comparison, Reliability engineering, Decision making, Graph models",
author = "Jimenez-Roa, {Lisandro A.} and Tom Heskes and Mari{\"e}lle Stoelinga",
year = "2021",
month = sep,
day = "21",
doi = "10.3850/978-981-18-2016-8_241-cd",
language = "English",
pages = "673--680",
editor = "Bruno Castanier and Marko Cepin and David Bigaud and Christophe B{\'e}renguer",
booktitle = "Proceedings of the 31st European Safety and Reliability Conference (ESREL 2021)",
publisher = "Research Publishing",
note = "European Safety and Reliability Conference 2021, ESREL 2021 ; Conference date: 19-09-2021 Through 23-09-2021",
url = "http://esrel2021.org/en/index.html",
}

{. N. {Alves da Silveira}, R. Loendersloot, {. A. Meghoe, and T. Tinga, “Data selection criteria for the application of predictive maintenance to centrifugal pumps,” in Proceedings of the 6th european conference of the prognostics and health management societ, 2021, p. 372–380.
[BibTeX] [Abstract] [Download PDF]

The maintenance of vehicles and components is present in most people’s daily lives, ranging from changing a private vehicle’s oil to the failure prediction of an aircraft component during flight. Usually, the manufacturer’s maintenance recommendation is a good solution when the cost is not too high, and the real application is used as indicated by the manufacturer. However, this recommendation can turn unfeasible when there is a significant variation in operational conditions or high maintenance costs. In these cases, the manufacturer’s suggestion is typically conservative, leading to unnecessarily high costs. Therefore, the challenge is to find the best approach for optimizing a component’s maintenance, given the system in which it is integrated and the associated operational and environmental conditions. Nevertheless, the available information on the loads on the component also plays a role in that choice. This paper proposes to combine case-specific information with generic degradation prediction models to obtain an acceptable but also affordable approach. The objective is to develop data selection criteria to indicate the parameters that have a high impact on the failure prediction, in this case, of a generic impeller pump. Subsequently, the approach delivers to the user an indication of the component remaining useful life using different operational scenarios.

@inproceedings{daSilveria2021_PHM,
title = "Data Selection Criteria for the Application of Predictive Maintenance to Centrifugal Pumps",
abstract = "The maintenance of vehicles and components is present in most people's daily lives, ranging from changing a private vehicle's oil to the failure prediction of an aircraft component during flight. Usually, the manufacturer's maintenance recommendation is a good solution when the cost is not too high, and the real application is used as indicated by the manufacturer. However, this recommendation can turn unfeasible when there is a significant variation in operational conditions or high maintenance costs. In these cases, the manufacturer's suggestion is typically conservative, leading to unnecessarily high costs. Therefore, the challenge is to find the best approach for optimizing a component's maintenance, given the system in which it is integrated and the associated operational and environmental conditions. Nevertheless, the available information on the loads on the component also plays a role in that choice. This paper proposes to combine case-specific information with generic degradation prediction models to obtain an acceptable but also affordable approach. The objective is to develop data selection criteria to indicate the parameters that have a high impact on the failure prediction, in this case, of a generic impeller pump. Subsequently, the approach delivers to the user an indication of the component remaining useful life using different operational scenarios. ",
author = "{Alves da Silveira}, {Nubia Nale} and Richard Loendersloot and Meghoe, {Annemieke Angelique} and Tiedo Tinga",
year = "2021",
month = jun,
language = "English",
series = "Archives of the PHM Society European Conference",
number = "1",
pages = "372--380",
booktitle = "Proceedings of the 6th European Conference of the Prognostics and Health Management Societ",
note = "6th European Conference of the Prognostics and Health Management Society, PHME 2021, PHME 2021 ; Conference date: 28-06-2021 Through 02-07-2021",
url = "https://phm-europe.org/",
}

List of student theses and internship reports

M. Crooymans, “Investigating the application of condition-based maintenance at the rnln using partially observable markov decision processes,” Eindhoven University of Technology. Internship performed at the Royal Netherlands Navy, 2020.
[BibTeX] [Download PDF]

@Article{Crooymans2020,
AUTHOR = {Crooymans, Maud},
TITLE = {Investigating the Application of Condition-Based Maintenance at the RNLN using Partially Observable Markov Decision Processes},
JOURNAL = {{Eindhoven University of Technology. Internship performed at the Royal Netherlands Navy}},
YEAR = {2020},
URL = {https://primavera-project.com/wp-content/uploads/2020/12/Crooymans2020_1MSE15_RNLN.pdf}
}