Abstract

The manufacturers and the operators of the fleets of cyber-physical systems (CPSs) are subjected to huge expectations expressed in terms of the availability and reliability of the provided products and services during the exploitation of these fleets in dynamic environments. These expectations foster the fleet manufacturers, particularly in the transportation sector, to develop effective mechanisms as far as the reactive planning of the maintenance operations at the fleet level is concerned. In this research work, a multi-agent system (MAS) for the reactive maintenance planning of a fleet of CPSs is proposed. The proposed MAS is conceived by using the ANEMONA design methodology and it aims at optimizing the fleet maintenance planning decisions to meet the specified objectives. The experiments carried out in the course of this work demonstrate the ability of the proposed MAS in planning the fleet maintenance effectively (i.e. satisfying the fleet’s availability and reliability requirements in a static environment) and reactively (i.e. being able to adapt/modify the fleet maintenance planning decisions following perturbations). The effectiveness of the MAS model is validated by a mathematical programming model and its reactivity is tested by using simulated perturbations. An application in rail transport industry to the fleet of trains at Bombardier Transportation France is proposed. The proposed MAS is integrated in a decision support system called "MainFleet". The development of Main-Fleet at Bombardier is ongoing.; Les industriels et les opérateurs des flottes de systèmes cyber-physiques (CPS) sont soumis à de fortes exigences exprimées en termes de disponibilité, fiabilité des produits et des services fournis lors de l’exploitation de ces flottes dans des environnements dynamiques. Ces attentes incitent les industriels, et notamment dans le secteur du transport, à développer des mécanismes efficaces de planification réactive des opérations de maintenance au niveau de la flotte. Dans cette thèse, un système multi-agent (SMA) pour la planification réactive de la maintenance d’une flotte de CPS est proposé. Ce SMA est construit en utilisant la méthode de conception ANEMONA et a pour objectif d’optimiser la planification de la maintenance au niveau flotte afin de répondre aux exigences spécifiées. Les expériences réalisées au cours de ces travaux démontrent la capacité de ce SMA à planifier la maintenance de la flotte de manière efficace (c'est-à-dire satisfaire les exigences de disponibilité et de fiabilité de la flotte dans un environnement statique) et de manière réactive (c'est-à-dire être capable d'adapter/de modifier les décisions de planification de la maintenance à la suite des perturbations). L'efficacité de ce modèle SMA est validée par un modèle mathématique et sa réactivité est testée par simulation de perturbations. Une application dans le domaine ferroviaire au sein de Bombardier Transport France est proposée. Le SMA est intégré à un système d’aide à la décision dénommé « MainFleet ». Le développement de MainFleet est en cours.

Abstract

In recent years, more and more manufacturers and operators of fleets of mobile systems have been focusing their efforts on studying and developing conditional maintenance, monitoring, and diagnostic strategies to cope with an increasingly competitive, unstable, costly, and unpredictable environment. This paper proposes a case study concerning the application of a novel event management architecture, called EMH2, to a fleet of trains. This EMH2 architecture, which applies the holonic paradigm, aims to facilitate the monitoring and diagnosis of a fleet of mobile systems. It is based on a recursive decomposition of cooperative monitoring holons. The definition of a generic event modeling, called SurfEvent, is the second key element of the contribution. EMH2 has been designed to be applicable to any kind of system or equipment up to fleet level. The edge computing paradigm has been adopted for implementation purpose. The EMH2 architecture is designed to facilitate asynchronous and progressive onboard and off-board deployments. A real-world application of EMH2 to a fleet of ten trains currently in use, in collaboration with our industrial partner, Bombardier Transport, is presented. Three key performances indicators have been estimated by comparing EMH2 with the current industrial situation. These indicators are (1) the number of fleet maintenance visits, (2) the time needed by a maintenance operator to investigate and diagnose, and (3) the time needed by the system to update data regarding the health status and monitoring of trains. Results obtained outperformed industrial expectations. The paper finally discusses feedbacks from experience and limitations of the work. Document type: Article