In a sustainable development context, the stakes of the last stage of system life cycle, the end-of-life stage, have increased over recent years. End-of-life systems have to be de-manufactured in order to be valued so as to respond to environmental concerns. The aim of a disassembly strategy consists in issuing a solution to the whole decision problem raised during the end-of-life stage of systems. Indeed, decision makers have to select valuable components according to technical, economical and environmental criteria and then design and optimise a disassembly support system that will generate these products. The solution obtained is what we refer to in this article as a disassembly trajectory. The work presented in this article is about planning these trajectories on different horizons integrating several arrivals of end-of-life systems. The proposed approach, with Bayesian networks and influence diagrams as the underlying mathematical tools, enables dynamically defined uncertainties to be taken into account.

From reflectometry methods, this work aims at locating accurately electrical faults in complex wiring networks. Increasing demand for online diagnosis has imposed serious challenges on interference mitigation. In particular, diagnosis has to be carried out while the target system is operating. The interference becomes more even critical in the case of complex networks where distributed sensors inject their signals simultaneously. The objective of this paper is to develop a new embedded diagnosis strategy in complex wired networks that would resolve interference problems and eliminate ambiguities related to fault location. To do so, OMTDR (Orthogonal Multi-tone Time Domain Reflectometry) method is used. For better coverage of the network, communication between sensors is integrated using the transmitted part of the OMTDR signal. It enables data control and transmission for fusion to facilitate fault location. In order to overcome degradation of diagnosis reliability and communication quality, we propose a new sensor clustering strategy based on network topology in terms of distance and number of junctions. Based on CAN bus network, we prove that data fusion using sensor clustering strategy permits to improve the diagnosis performance.

The evaluation of maintenance strategies is a problem which has been widely tackled. However, it remains a topic worthy of consideration since the economic stakes peculiar to this activity are huge. The work presented in this paper deals with this problem under the point of view of data feedback. A statistical analysis is made from the examination of the forms of the failure rate distribution. The laws resulting from the implementation of a given maintenance strategy inform us about the relevance of the choices representative of this strategy. The construction of these laws from a raw time of repair is a global approach which tends to hide the induced phenomenon. A decomposition of the repair process into phases allows us to better identify the influence of the strategy parameters. The analysis of each elementary phase distribution is interesting since it givesinformation about the performance of the implemented maintenance strategy. Copyright © 2003 John Wiley & Sons, Ltd.

Nowadays, increasing demands for on-line wire diagnosis using reflectometry have imposed serious challenges on signals processing, bandwidth control and interference mitigation. On-line diagnosis aims at detecting and locating faults accurately while the target system is running. In this work, a new reflectometry method, named “Orthogonal Multi-Tone Time Domain Reflectometry” (OMTDR), is proposed. OMTDR, based on Orthogonal Frequency Division Multiplexing (OFDM), is a suitable candidate for on-line diagnosis as it permits interference avoidance, bandwidth control and data rate increase thanks to the use of orthogonal tones and guard intervals. Over the diagnosis function, OMTDR adds communication between sensors to more accurately determine faults position in a multi-branch network using a distributed strategy. OMTDR was tested on a branched network consisting of three cables with different lengths, with sensors at each cable end. Here, the sensors signals are carefully constructed using a resource allocation scheme to use frequencies below and above the prohibited bandwidth, used by the target system, for communication and diagnosis. Simulation results show that the proposed method performs well in a branched wiring network as it permits to detect and locate faults accurately even when the target system is operating.