Patrice Aknin

• PhD, HDR
• Managing Director at Institut de Recherche Technologique SystemX

Dans le domaine de la mobilité, les travaux de recherche visent à mieux comprendre les pratiques de mobilité et la manière dont elles s'articulent à des territoires urbains. L'ouverture des données pose une question essentielle : Quels apports de connaissances sur nos mobilités et nos villes ces nouvelles données offrent-elles ? Et dans quelle mesu...

Industry-academic research partnerships are mostly considered interesting to increase industrial innovativeness, and its benefits have been discussed in the flourishing open innovation literature. They have therefore been thoroughly studied, but not only controversial results exist on the impact of these partnerships on academic research, but how t...

The maintenance optimization of complex systems is a key question. One important objective is to be able to anticipate future maintenance actions required to optimize the logistic and future investments. That is why, over the past few years, the predictive maintenance approaches have been an expanding area of research. They rely on the concept of p...

System degradation modelling is a key problem when performing any type of reliability study. It is used to determine the quality of the computed reliability indicators and prognostic estimates. However, the mathematical models that are commonly used in reliability studies (Markov chains, gamma process. etc.) make certain assumptions that can lead t...

The problem of temporal data clustering is addressed using a dynamic Gaussian mixture model. In addition to the missing clusters used in the classical Gaussian mixture model, the proposed approach assumes that the means of the Gaussian densities are latent variables distributed according to random walks. The parameters of the proposed algorithm are...

In any reliability analysis, the degradation modeling is a key point. Indeed, the accuracy of all reliability indicators and prognosis estimations will directly depends on the quality of the degradation modeling. Commonly used stochastic models such as Markov chains, Gamma process… are generally based on some strong assumptions on the stochastic pr...

In order to meet the mounting social and economic demands, railway operators and manufacturers are striving for a longer availability and a better reliability of railway transportation systems. Commercial trains are equipped with state-of-the-art onboard intelligent sensors monitoring various subsystems all over the train such as tilt, traction, si...

The optimization of maintenance strategies has become a key issue in the railway industry but also in most industrial fields. To address this challenge, many studies dealt with the estimation of optimal maintenance parameters. But what commonly happens when the degradation process suddenly changes? The operator has to face an unexpected, increasing...

Transport systems are more and more complex and the number of users is increasing. So, it makes necessary both to increase the availability of the material and to guaranty an high level of security keeping reasonable maintenance costs. For this reason, the maintenance optimization of transport systems has become a key issue. Currently, industrial j...

This paper deals with a data mining approach applied on Bike Sharing System Origin-Destination data, but part of the proposed methodology can be used to analyze other modes of transport that similarly generate Dynamic Origin-Destination (OD) matrices. The transportation network investigated in this paper is the Vélib’ Bike Sharing System (BSS) syst...

This paper describes a pattern recognition approach aiming to estimate fuel cell duration time from electrochemical impedance spectroscopy measurements. It consists in first extracting features from both real and imaginary parts of the impedance spectrum. A parametric model is considered in the case of the real part, whereas regression model with l...

This paper proposes a method of segmenting temporal data into ordered classes. It is based on mixture models and a discrete latent process, which enables to successively activates the classes. The classification can be performed by maximizing the likelihood via the EM algorithm or by simultaneously optimizing the model parameters and the partition...

A non linear regression approach which consists of a specific regression model incorporating a latent process, allowing various polynomial regression models to be activated preferentially and smoothly, is introduced in this paper. The model parameters are estimated by maximum likelihood performed via a dedicated expecation-maximization (EM) algorit...

A new approach for signal parametrization, which consists of a specific regression model incorporating a discrete hidden logistic process, is proposed. The model parameters are estimated by the maximum likelihood method performed by a dedicated Expectation Maximization (EM) algorithm. The parameters of the hidden logistic process, in the inner loop...

In order to meet the mounting social and economic demands, railway operators and manufacturers are striving for a longer availability and a better reliability of railway transportation systems. Commercial trains are being equipped with state-of-the-art onboard intelligent sensors monitoring various subsystems all over the train. These sensors provi...

The conventional change-point detection problem aims to detect distribution changes at some unknown time point in a sequence of multivariate observations. Such problem is hardly addressed when the data are functional and both the pre-change and post-change distributions are unknown. In this paper, we propose an online sequential procedure based on...

Considering availability purposes for train transportation, passenger accesses (doors and steps) are often designated as critical systems. To improve global availability of its rolling stock, Bombardier Transportation (BT) aims at reinforcing its maintenance procedure by introducing predictive diagnosis. The SURFER project has been initiated to dev...

In most industrial fields, and particularly in the railway industry, the optimization of maintenance policies has become a key issue. Dynamic Bayesian networks (DBN) have been proved as relevant to perform reliability analysis as they can easily represent complex systems behaviors. Based on this formalism, graphical duration models (GDM) were devel...

The increasing demand on railway transportation due to social, economic and demographic factors has triggered an evolution in preventive maintenance strategies towards more optimized and cost-effective processes. Within this context related to the increasing interest in predictive maintenance strategies, commercial trains are being equipped with bo...

This article introduces a state-space model for the dynamic modeling of curve sequences within the framework of railway switches online monitoring. In this context, each curve has the peculiarity of being subject to multiple changes in regime. The proposed model consists of a specific latent variable regression model whose coefficients are supposed...

This paper addresses the problem of temporal data clustering using a dynamic Gaussian mixture model whose means are considered as latent variables distributed according to random walks. Its final objective is to track the dynamic evolution of some critical railway components using data acquired through embedded sensors. The parameters of the propos...

Assessing the operating state of the railway infrastructure and rolling stock using condition measurements acquired through embedded sensors has become a powerful decision-making support for preventive maintenance strategies. This article introduces a dynamic approach for the online monitoring of railway switch operations. The method is based on mo...

In most industrial fields, and particularly in the railway industry, the optimization of maintenance policies has become a key issue. To address this problem, predictive maintenance seems to be one of the most effective known approaches. It consists of driving the maintenance process anticipating the evolution of the system state. A prediction proc...

This study was motivated by the characterization of the dynamic evolution of some critical railway components (point machines and door systems) using condition measurements acquired through embedded sensors. Its final objective is to build a decision-aided support for their preventive maintenance. One of the difficulties in achieving this goal is t...

Cet article aborde le problème de la classification des données temporelles en utilisant un mélange dynamique de lois gaussiennes dont les moyennes sont considérées comme des variables latentes qui évoluent suivant des marches aléatoires.

The problem of durability of fuel cell technology is central for its spreading and commercialization. There is therefore a growing need to build accurate diagnosis tools which can give the operating state of the fuel cell during their use. When supervised machine learning approaches are used to build such diagnosis tools, they generally require a l...

In most industrial fields, the optimization of maintenance policies has become a key issue. To address this problem, predictive maintenance seems to be one of the most effective known approaches. It con-sists of driving the maintenance process anticipating the evolution of the system state. A prediction process, called "prognostic" could also be in...

Detecting change-points and anomalies on sequential data is common in various domains such as fraud detection for credit cards, intrusion detection for cyber-security or military surveillance [1]. This study is motivated by the predictive maintenance of pneumatic doors in transit buses. For this purpose, buses are instrumented and data are collecte...

The increasing interest in preventive maintenance strategies for railway transportation systems and the emergence of telecommunication technologies have both led to the development of floating train data (FTD) systems. Commercial trains are being equipped with both positioning and communications systems as well as onboard intelligent sensors monito...

Independent Factor Analysis (IFA) is used to recover latent components (or sources) from their linear observed mixtures within an unsupervised learning framework. Both the mixing process and the source densities are learned from the observed data. The sources are assumed to be mutually independent and distributed according to a mixture of Gaussians...

Using a statistical model in a diagnosis task generally requires a large amount of labeled data. When ground truth information is not available, too expensive or difficult to collect, one has to rely on expert knowledge. In this paper, it is proposed to use partial information from domain experts expressed as belief functions. Expert opinions are c...

This paper presents a pattern-recognition-based diagnosis approach for fault-diagnosis of fuel cell stacks, using Electrochemical Impedance Spectroscopy (EIS). It aims at implementing a diagnosis tool able to detect fuel cell degradations from EIS measurements. It consists in different steps. First, measurable features are extracted. Then, in order...

Mixture model-based clustering, usually applied to multidimensional data, has become a popular approach in many data analysis problems, both for its good statistical properties and for the simplicity of implementation of the Expectation–Maximization (EM) algorithm. Within the context of a railway application, this paper introduces a novel mixture m...

Efficient dimensionality reduction can involve generative latent variable models such as probabilistic principal component analysis (PPCA) or independent component analysis (ICA). Such models aim to extract a reduced set of variables (latent variables) from the original ones. In most cases, the learning of these models occur within an unsupervised...

Cet article introduit une approche générique nommée VirMaLab (pour atelier virtuel de maintenance) permettant de développer des modèles de maintenance, basée sur la fiabilité, de systèmes multicomposants et multiétats. S’appuyant sur le formalisme des modèles graphiques probabilistes (MGP) [ou réseaux bayésiens (RB)], cette approche stochastique mo...

Diagnosis of complex systems refers to the problem of identifying a breakdown or a failure based on an inspection, a control or a test. Monitoring such industrial complex systems is essential to schedule relevant maintenance actions. We consider an automotive subsystem to monitor: the brake system, because of its impact on the vehicles availability...

Equipments used in industrial environments such as production lines, engineering or mass transport system, are generally complex, multi-components and multi-states systems (MSS). These equipments are subject to degradation mechanisms caused by operating conditions/environment (temperature, vibrations). In addition to these degradation mechanisms, t...

This paper introduces a novel model-based clustering approach for clustering time series which present changes in regime. It consists of a mixture of polynomial regressions governed by hidden Markov chains. The underlying hidden process for each cluster activates successively several polynomial regimes during time. The parameter estimation is perfo...

Independent factor analysis (IFA) defines a generative model for observed data that are assumed to be linear mixtures of some unknown non-Gaussian, mutually independent latent variables (also called sources or independent components). The probability density function of each individual latent variable is modelled by a mixture of Gaussians. Learning...

The remote monitoring of the railway infrastructure and particularly the switch mechanism is of great interest for railway operators. The problem consists in detecting earlier the presence of defects in order to alert the concerned maintenance service before a breakdown occurs. For this purpose, this paper introduces a new probabilistic-based appro...

Reliability analysis has become an integral part of system design and operating. This is especially true for systems performing critical tasks such as mass transportation systems. This explains the numerous advances in the field of reliability modeling. More recently, some studies involving the use of Bayesian Networks (BN) have been proved relevan...

This paper deals with a pattern-recognition-based diagnosis approach, which aim is to estimate the Fuel Cell (FC) operating time, and consequently its remaining duration life. With the method proposed, both static and dynamic information extracted from the stack (i.e. polarization curve records and Electrochemical Impedance Spectroscopy (EIS) measu...

This paper investigates the use of partially reliable information elicited from multiple experts to improve the diagnosis of a railway infrastructure device. The general statistical model used to perform the diagnosis task is based on a noiseless Independent Factor Analysis handled in a soft-supervised learning framework.

Résumé: Les équipements employés dans les milieux industriels, tels que les chaînes de production, l'ingénierie ou le transport, se présentent généralement sous la forme de systèmes complexes multi-composants et multi-états. L'état de ces systèmes est souvent affecté par les conditions d'usage. La compréhension de l'évolution au cours du temps des...

Cet article est une synthèse de travaux de thèse ayant pour objectif industriel l'élaboration d'un modèle de maintenance visant à améliorer la programmation des opérations d'entretien de la voie ferrée du réseau RATP (Régie Autonome des Transports Parisiens). Cette étude s'intéresse en particulier à la dynamique d'évolution des défauts de fatigue d...

Durability is one of the limiting factors for spreading and commercialization of fuel cell technology. That is why research to extend fuel cell durability is being conducted world wide. A pattern-recognition approach aiming to estimate fuel cell operating time based on electrochemical impedance spectroscopy measurements is presented here. It is bas...

A new approach for functional data description is proposed in this paper. It consists of a regression model with a discrete hidden logistic process which is adapted for modeling curves with abrupt or smooth regime changes. The model parameters are estimated in a maximum likelihood framework through a dedicated expectation maximization (EM) algorith...

This paper addresses the problem of fault detection and isolation in railway track circuits. A track circuit can be considered as a large-scale system composed of a series of trimming capacitors located between a transmitter and a receiver. A defective capacitor affects not only its own inspection data (short circuit current) but also the measureme...

Originally devoted to specific applications such as biology, medicine and demography, duration models are now widely used in economy, finance or reliability. Recent works in various fields of application have shown the relevancy of using Bayesian networks to model complex systems, namely stochastic systems with an underlying distribution that does...

Introduction Representation space: Parametrization and selection Signal classification Data fusion General conclusion Bibliography

Reliability analysis is an integral part of system design and operating. Moreover, it can be an input to optimize maintenance policies. Recently, Bayesian Networks (BN) and Dynamic Bayesian Networks (DBN) have been proved relevant to represent complex systems and perform reliability studies. The major drawback of this approach comes from the constr...

This paper describes a pattern recognition approach aiming to estimate fuel cell duration time from electrochemical impedance spectroscopy measurements. It consists in first extracting features from both real and imaginary parts of the impedance spectrum. A parametric model is considered in the case of the real part, whereas regression model with l...

In Independent Factor Analysis (IFA), latent components (or sources) are recovered from only their linear observed mixtures. Both the mixing process and the source densities (that are assumed to be generated according to mixtures of Gaussians) are learned from observed data. This paper investigates the possibility of estimating the IFA model in its...

Reliability analysis is an integral part of system design and operating. Moreover, it can be an input to optimize maintenance policies. Recently, Bayesian Networks (BN) and Dynamic Bayesian Networks (DBN) have been proved relevant to represent complex systems and perform reliability studies. The major drawback of this approach comes from the constr...

Time series are used in many domains including finance, engineering, economics and bioinformatics generally to represent the change of a measurement over time. Modeling techniques may then be used to give a synthetic representation of such data. A new approach for time series modeling is proposed in this paper. It consists of a regression model inc...

A new approach for feature extraction from time series is proposed in this paper. This approach consists of a specific regression model incorporating a discrete hidden logistic process. The model parameters are estimated by the maximum likelihood method performed by a dedicated expectation maximization (EM) algorithm. The parameters of the hidden l...

This paper addresses classification problems in which the class membership of training data is only partially known. Each learning sample is assumed to consist in a feature vector xi in X and an imprecise and/or uncertain “soft” label mi defined as a Dempster-Shafer basic belief assignment over the set of classes. This framework thus generalizes ma...

This paper presents a study on the potential interest of sparse Independent Component Analysis (ICA) for the diagnosis of a complex railway infrastructure device. This complex system is composed of several spatially related subsystems, i.e. a defective subsystem modifies not only its own inspection data but also those of other subsystems. In this c...

Rail corrugation is an oscillatory wear of rail surface due to the interaction between rail and wheel. Standard signal processing approaches to corrugation monitoring, as devised in the European standards and in use in railway networks, are designed in the mileage or wavelength domain. Time-frequency analysis computes the energy distribution of the...

Rail corrugation is an ondulatory mechanical wear of rail surface raising from the interaction between rail and wheel. Signal processing approaches to corrugation monitoring recommended by the European standards are designed in the distance or wavelength domain. However a joint distance and wavelength domain analysis of the monitoring data can prov...

Cet article présente une approche basée sur l'Analyse en Composantes Indépendantes parcimonieuse pour le diagnostic automatique d'un système réparti de l'infrastructure ferroviaire. Ce système est composé de plusieurs sous-systèmes sur lesquels les observations prélevées sont spatialement liées; la défaillance d'un sous-système influera donc sur to...

Independent factor analysis (IFA) is a well known method used to recover independent components from their linear observed mixtures without any knowledge on the mixing process. Such recovery is possible thanks to the hypothesis that the components are mutually independent and non-Gaussians. The IFA model assumes furthermore that each component is d...

This article describes a hybrid diagnosis system based on the combined use of sensor data (local information) and structural knowledge (global information). The approach is illustrated on an application that involves the detection of broken rail for railway infrastructure. Recently, there have been a large number of attempts to solve diagnosis prob...

This paper addresses classification problems in which the class membership of training data is only partially known. Each learning sample is assumed to consist in a feature vector and an imprecise and/or uncertain “soft” label m i defined as a Dempster-Shafer basic belief assignment over the set of classes. This framework thus generalizes many kind...

A non destructive evaluation system dedicated to rail inspection using a non-contact eddy current sensor embedded in a subway train is presented. An original processing approach borrowed from the MUSIC algorithm is proposed for rail surface defects detection and classification. This approach, based on the eigen decomposition of the signal covariance...

Reliability analysis has become an integral part of system design and operating. This is especially true for systems performing critical tasks such as mass transportation systems. This explains the numerous advances in the field of reliability modelling. More recently, some studies involving the use of Probabilistic Graphical Models (PGMs), a.k.a....

An original pattern recognition approach for the diagnosis of switch mechanisms driven by an electric motor is presented in this paper. Its main advantage is that it does not require a physical model of the system and can easily be adapted to other complex systems. The available data for this task are the signals of the electrical power consumption...

Ultrasonic inspection systems generally use ASCAN mode to detect rail crack. A flight time window is open where a flaw echo is validated when it occurs with a sufficient level. The maintenance operators well know the difficulties to adjust the threshold and the position of the time window : if the threshold is too high, the system misses cracks, if...

Reliability analysis has become an integral part of system design and operating. This is especially true for systems performing critical tasks. Moreover, recent works in reliability involving the use of probabilistic graphical models, also known as Bayesian networks, have been proved relevant. This paper describes a specific dynamic graphical model...

Originally devoted to specific applications such as biology, medicine and demography, the duration models are now widely used in economy, finance or reliability. Some recent works in reliability analysis have been proved relevant the use of bayesian networks. In this paper, we describe a specific dynamic bayesian network, named graphical duration m...

Reliability analysis has become an integral part of system design and operating. This is especially true for systems performing critical tasks such as mass transportation systems. This explains the numerous advances in the field of reliability modelling. More recently, some studies involving the use of Probabilistic Graphical Models (PGMs), a.k.a....

Cet article présente l'implémentation d'un réseau bayésien dynamique avec variable exogène continue, appliquée à la classification d'évènements discrets irrégulièrement espacés, organisés en séquence. La modélisation des tables de probabilités conditionnelles, qui sont fonction d'une variable exogène codant un vecteur de distances entre évènements,...

This paper proposes a new signal denoising methodology for dealing with asymmetrical noises. The adopted strategy is based on a regression model where the noise is supposed to be additive and distributed following a mixture of Gaussian densities. The parameters estimation is performed using a Generalized EM (GEM) algorithm. Experimental studies on...

Hidden Markov Random Fields (HMRF) are widely used in solving various problems. Image segmentation is an example of such HMRF success. This paper presents a post-processing tool based on such a model and designed to increase the relevancy of a diagnosis system for rail defects detection. In this application, the hidden Markov field is not only used...

Cet article présente un outil de diagnostic automatique d'un système de l'infrastructure ferroviaire : le circuit de voie. Ce système est composé de plusieurs sous-systèmes (condensateurs) dont le nombre est variable et sur lesquels les observations prélevées sont spatialement dépendantes ; une défaillance d'un sous-système influera donc sur toute...

This paper addresses the problem of fault detection in a complex system made up of several spatially dependent subsystems. The diagnosis method consists of both detecting and localizing a defect on the system by combining the outputs scores of subclassifiers within the framework of belief function theory. This paper is focused on the coding and the...

This paper introduces a diagnosis scheme of a railway infrastructure component based on a combined use of empirical mode decomposition (EMD) and Hilbert transform. This component is dedicated to track/vehicle transmission referred as track circuit. The aim is to detect its working state from one measurement signal which can be viewed as a superposi...

The ground-to-train transmission system (TVM) is a vital factor in railway safety on high speed lines. It is used to continuously transmit signalling information to the driver in the cab. The ground-based part of the system consists of a compensated UM71-type track circuit including a transmitter, a receiver and trimming capacitors, whose character...

- Cet article introduit une variante de la transformée de Radon qui autorise la mise en oeuvre d'une procédure de détection rapide non pénalisée par la rotation du segment recherché dans une image à horizon glissant. Cette transformée de Radon dite lissée utilise une fonction de voisinage sur lequel s'effectue l'intégration au lieu de la projection...

This work deals with a diagnosis system, based on a combined use of partial least squares regression (PLS) and neural network (NN). An application concerning the French railway track/vehicle transmission system illustrates this approach. It is shown that a reliable selection of a reduced set of relevant descriptors is made by the PLS regression. Mo...

In a classification problem, hard margin SVMs tend to minimize the generalization error by maximizing the margin. Regularization is obtained with soft margin SVMs which improve performances by relaxing the constraints on the margin maximization. This article shows that comparable performances can be obtained in the linearly separable case with the...