Year 2011

#1. Modelling of ageing components for system reliability analysis and risk assessment. Collaborative project with BUAA. Funded by LIA 2MCSI.

This project aims at developing efficient methods for modeling the process of degradation to failure of industrial components. The mainstream research on aging process modelling is largely dependent on data-driven methods and statistical modelling. The related methods need considerable amount of samples for model building. However, the degradation-to-failure processes in systems like critical infrastructures and technological, high-risk plants, like nuclear, are rare and hardly repeatable in terms of influencing conditions. For this reason, Multi-state physics modeling (MSPM) is considered as a feasible approach for estimating the failure probability of critical components and systems undergoing degradation. This approach integrates multi-state modeling, which describes the degradation process by transitions among discrete states (e.g. initial, micro-crack, rupture, etc), and physics modeling by (physics) equations that describe the degradation process among the states. In reality, the degradation process is non-Markovian, and its transition rates are time-dependent and possibly influenced by uncertain external factors such as temperature and stress. Under these conditions, it is in general difficult to derive the degradation state probabilities analytically. This project aims in particular at studying simulation methods for solving MSPM and implementing the developed methods onto real-world application cases.

Year 2012

#1. Development of methods for modeling degradation and maintenance of critical components and of a framework for integrating information and data of different nature.

Critical plant components are in general highly reliable, and this leads to the fact that there is a very limited amount of data available for building statistical models of the degradation and failure processes, which can be used to guide maintenance decisions. Therefore, it becomes necessary to use also other types of information, e.g. physical models and expert judgment, for completing and supporting the modeling of the degradation and failure behaviors. In addition, components are often subject to multiple degradation processes and failure modes, each of which can be influenced by a number of interdependent external factors (e.g. temperature, stress, overload and/or overstress due to failure of other components, etc).

The purpose of this research, object of the contract, is to develop an effective framework of computational methods for modeling the degradation processes of critical components and inform the successive maintenance decisions. The framework is expected to include methods of degradation process and maintenance system representation (e.g. Markov diagrams, Petri nets, Bayesian networks, etc.), methods for the Monte Carlo simulation of component degradation and related maintenance activities, and methods of information representation/processing (e.g. Bayesian networks, fuzzy logic systems) to handle various types of influential information with their associated uncertainties.

#2. Modelling of ageing components for system reliability analysis and risk assessment (Renewed). Funded by LIA 2MCSI.

The objectives of this project are set in continuation of the outcomes of the previous joint-project, with the same previous vision of sharing a research journey mutually beneficial to both teams, aimed at strengthening the quality and recognition of ECP and BUAA in the areas of component/system aging modeling and reliability engineering in general.

#3. Méthodes pour le pronostic de défaillances des matériels des centrales de production d’énergie. Funded by EDF.

Maintaining the high reliability of nuclear plants requires efficient and timely access to information on plant asset performance and condition as well as processing of this information to make cost-effective decisions regarding maintenance priorities. The expectation of continued constraints on maintenance resources and staffing within the power industry calls for the establishment of a strategy for the effective condition-based maintenance of components. In general, this entails:
i) the timely detection of abnormal operation conditions (condition monitoring);
ii) the identification of the causes of abnormality (fault diagnosis);
iii) the prediction of the remaining useful life in the given abnormal conditions (fault prognosis).
Significant efforts to increase the power plants competitiveness have been exerted recently, and good results can be achieved today through different approaches, mainly:
- Preventing damage and reducing downtime and cost achieved due to the timely, efficient repair enabled by the early detection of any anomaly.
- Increasing availability by preventing plant shutdown, and through a reduction of unplanned outage time in case of anomaly.
- Arranging maintenance plans at the most convenient time. Spare parts can be scheduled, considering the current operating environment within which the early warning notice was provided.
In this respect, the project aims at a successful resolution of the technical challenges associated to condition monitoring, fault diagnosis and prognosis for establishing effective condition-based preventive maintenance strategies capable of optimally coping with the production availability targets and the safety requirements of nuclear power plants.

#4. Research NEtwork on FlExible Risk AssEssmeNt and DeCision SciencE (REFERENCE). Funded by EU.

The overall aim of this proposed exchange programme is to bring together an international team of researchers, with a wide variety of skills in order to develop a formal safety assessment framework with appropriate support models for application to marine, oil and gas, supply chain management, nuclear and transport areas. Life cycle safety of large complex engineering systems will be investigated in terms of the design, manufacture, installation, commissioning, operations and maintenance phases. A variety of appropriate subject topics such as uncertainty modelling, expert knowledge elicitation, human and organizational factor study, risk prediction, software tools, probabilistic risk estimation, cost benefit modelling and multiple criteria decision making will be investigated in the selected industrial sectors using the expertise the partners of the consortium possess. The proposal is for a project of five partners (3 EU members, 1 AS member and 1 ICPC member) with extensive exchange of both experienced researchers (ERs) and early stage researchers (ESRs) during four years to fully explore the complementary strengths and synergies within the consortium. This project will support and reinforce the collaborations amongst the participants and help establish a long-term research co-operation. The research will increase the European research capacity in this vital and rapidly developing field, and also maintain and enhance the EU’s leading position in the area. Moreover, the interdisciplinary nature of the proposed exchange programme offers a link for research and training of the involved ERs and ESRs in a collaborative academic environment.

Participant number

Partner name

short name


1. Coordinator

Liverpool John Moores University



2. Partner 2

The University of Manchester



3. Partner 3

Ecole Centrale Paris



4. Partner 4

Norwegian University of Science and Technology



5. Partner 5

Wuhan University of Technology