Excellent papers for the 1st RECI workshop demonstrated strong synergy between Reliability Engineering and Computational Intelligence but scientific differences remain. The second RECI workshop attempts to address further integration by a) presenting papers on RECI topics, b) panel discussions about future RECI research and c) a curriculum discussion for an international RECI curriculum. The workshop is organized by University of Žilina, TNO, Delft University of Technology and the University of Huddersfield. Call for papers can be downloaded here.
The workshop will be supported by the following projects:
The workshop RECI 2022 will be a hybrid event: we facilitate contributions in-person and online from Delft University of Technology. The organizers believe that the 55 Euro fee for in-person should not be prohibitive for in-person participation. On-line participation will be free of charge. TUDelft hosts the event at the event location ‘X’ on Mekelweg 8, 2628 CD, Delft, the Netherlands
The special section “Reliability Engineering and Computational Intelligence application in healthcare” will be organized under the workshop. The goal of this section is discussion relevant topics and trends of reliability engineering, data mining, and machine learning in healthcare, medicine, and biotechnologies. The possible topics of this section will be but not limited are:
The section organization is supported by the project “University-Industry Educational Centre in Advanced Biomedical and Medical Informatics” (612462-EPP-1-2019-1-SK-EPPKA2-KA)
The abstracts of the presentations will be published in the book of abstracts, which will be available on the website of the workshop. Also, authors of all presentations will be able to publish their own research as papers in the journal CERES.
Authors of the selected presentations will be invited to prepare contributions to the book “Reliability Engineering and Computational Intelligence” (working title), which will be published by Springer in a series of Studies in Computational Intelligence.
The presented contributions at the workshop RECI 2020 after review by the program committee were published in the book "Reliability Engineering and Computational Intelligence"
The workshop develops RECI as a research domain by calling papers on current research. Topics of interest include, but are not limited to:
Please enlist as a participant through the registration form by November 7, 2022. If your contribution is aimed to the RECI workshop, please select “Attendance Type: RECI”. If you participate as a member of the ACESYRI project, please select “Attendance Type: ACESYRI”. If you are travelling to Delft to attend in person, please make that clear in the submission and send also an e-mail to: firstname.lastname@example.org to confirm your presence in Delft.
Department of Energy, Politecnico di Milano, Italy
Reliability: the Contribution of Computational Intelligence to Maintenance
Predictive Maintenance makes use of condition monitoring data to detect anomalies (i.e., recognize deviations from normal operating conditions) in production processes, manufacturing equipment and products, diagnose (i.e., characterize the occurring abnormal state) and prognose (i.e., predict the future evolution of the abnormal state up to failure). The capability of performing these tasks can bring significant economic, reliability and safety benefits. This presentation will illustrate the main challenges towards the deployment of predictive maintenance and its effective integration in the operation and maintenance of industrial systems. By way of examples of application, the opportunities provided by the recent advancements of computational intelligence, such as deep learning, with respect to predictive maintenance will be discussed.
Frank P.A. Coolen and Jinlei Qin
Department of Mathematical Sciences, Durham University, United Kingdom
Reliability Evaluation of Multi-State Systems with Multi-State Components using the Survival Signature
The presentation starts with a brief introductory overview of the survival signature for quantification of system reliability, where both the components and the systems only have two states: functioning or failed. The main contribution presented consists of generalization of the survival signature methodology to multi-state scenarios, where both the components and the system can have multiple possible states. Whilst the computation of such generalized survival signatures is non-trivial, an easy algorithm is presented for systems made up of subsystems, in case the overall system state is a function of only the states of the subsystems. Challenges for practical application to large-scale systems are discussed.
Leeds Beckett University, United Kingdom
Expressive Cardinal Direction Representation and Reasoning
In the lecture, I shall demonstrate how to group the nine cardinal directions into sets (known as Horizontal Vertical Constraints Model, HVCM). Mereology (‘Whole and Part’) and cardinal direction relations are combined to create a hybrid model which is more expressive. Lastly, the 2-D HVCM is adapted and extended into a 3-D Horizontal and Vertical Constraints Block model so that it facilitates easy reasoning with 3-D volumetric regions (i.e. without holes and single-pieced) in the real physical world (e.g. intelligent robotics, building construction).