Systèmes complexes et la théorie (science) de la complexité

MÉTHODOLOGIE DE RECHERCHE

Introduction à la méthodologie

Ce chapitre présente la méthodologie de recherche et définit la structure des prochains chapitres de la thèse. Son développement se base sur les questions, but et objectifs de recherche.
Elle vise à développer de manière originale les éléments constitutifs de la méthodologie et du modèle global de prise de décision en GDA en passant par les étapes principales suivantes :
 élaborer un modèle intégral de la gestion des actifs en tenant compte de la complexité de l’environnement opérationnel et d’affaires et des risques;
 effectuer des analyses approfondies pour identifier les parties constitutives et leurs relations au sein du modèle intégral; déterminer les niveaux de complexité et les types d’incertitude des éléments constitutifs du modèle;
 identifier et caractériser événements extrêmes et rares pouvant affecter la performance du processus de la GDA;
 définir des principes pour atteindre la robustesse, la résilience et la flexibilité de l’organisation en GDA;
 élaborer un modèle intégral de prise de décision en gestion des actifs.
La recherche fournira un cadre de référence amélioré pour des stratégies de la GDA optimales. Le processus global de prise de décision en tenant compte des risques en GDA doit également intégrer les intrants de l’orientation stratégique d’une organisation, les incertitudes associées (aléatoires et épistémiques) et les contraintes liées à la gestion globale du risque. En outre, le modèle devrait intégrer intrinsèquement un processus d’amélioration continue basé sur divers retours d’expérience interne et externe pertinents au cours du cycle de vie de l’organisation et des actifs (par exemple, audits internes et externes et/ou expérience d’exploitation).
La recherche actuelle n’a pas l’intention de développer tous les sous-modèles en GDA étant donné que certains d’entre eux sont déjà partiellement ou entièrement disponibles ou peuvent être développés et améliorés dans de futurs travaux de recherche.
Le travail portera principalement sur la gestion des actifs physiques. Dans le même temps, il considérera également d’autres types d’actifs (financiers, humains, information, autres) lorsqu’ils ont un impact sur la gestion optimisée des actifs physiques et pour assurer une cohérence globale de la GDA.
La thèse est constituée sept chapitres. Les trois chapitres suivants présentent les articles de revues avec comité de lecture qui sont directement liés au sujet de cette recherche. En effet, quatre catégories d’articles ont été produites durant les travaux de recherche. Elles sont présentées à la figure 3.1.
Les articles de la catégorie 1 (articles directement liés au sujet de la recherche et publiés dans des revues avec comité de lecture) représentent le résultat principal du développement de la méthodologie et de son application.
Les articles de la catégorie 2 (articles directement liés au sujet de la recherche et publiés dans des conférences internationales avec un comité de lecture) ont précédé les articles correspondants de la catégorie 1 et montrent une phase essentielle du développement de la méthodologie de recherche. Les études de cas des articles des catégories 1 et 2 sont réalisées en industrie minière, nucléaire ainsi qu’une utilité électrique d’envergure (Hydro-Québec). Ce fait démontre l’universalité de la méthodologie développée et son potentiel de l’application dans diverses industries (figure 3.2).
Les articles de la catégorie 3 et 4 représentent du développement et des applications spécifiques d’un des éléments constitutifs du modèle intégral de la gestion des actifs : a) la méthodologie améliorée de la prise de décision dans le contexte particulier des énergies renouvelables, b) l’impact de la complexité sur la gestion des actifs en industrie minière (travail exploratoire) et c) l’analyse des risques systémiques et organisationnels en santé et sécurité au travail (SST) en industrie minière en tenant compte de la complexité (figure 3.2). Ce développement est parfois réalisé dans le cadre d’un autre projet de recherche mais en lien avec les recherches dans la présente thèse.
L’indication des articles des catégories 2, 3 et 4 sert à illustrer l’envergure des travaux de recherche réalisés. Les résultats de ces travaux ont permis de démontrer le potentiel d’application de la méthodologie intégrale en GDA dans différentes disciplines ou industries.
Le tableau 3.1 présente la liste des articles de la catégorie 1 produits durant les travaux de recherche. La description de la méthodologie ci-dessous se base sur ces articles. Les articles des autres catégories démontrent l’applicabilité de la méthodologie développée dans divers domaines et activités industrielles. Un total de 18 articles a été préparé et/ou publié depuis le début de la recherche en 2013 dont neuf dans deux premières catégories. Tous les articles sont énumérés dans l’annexe 1.

Processus de la gestion des actifs comme un système complexe adaptatif et la création du modèle intégral initial de la GDA

Komljenovic, D., Abdul-Nour, G., & Popovic, N. (2015). Approach for strategic planning and asset management in mining industry in the context of business and operational complexity. International Journal of Mining and Mineral Engineering, 6(4), 338-360.
Cet article identifie la planification stratégique et le système de gestion des actifs en industrie minière comme un système complexe adaptatif. Il démontre que l’environnement opérationnel et d’affaires est aussi un système complexe en statuant que l’utilisation singulière des méthodes traditionnelles de modélisation et d’analyse n’est plus entièrement appropriée pour capturer et modéliser adéquatement cette complexité. Par conséquent, l’article propose la structure principale initiale des éléments constitutifs d’un modèle intégral de la gestion des actifs dans le contexte de la complexité. Il introduit la théorie de complexité pour complémenter les méthodes traditionnelles d’analyse et démontre la pertinence de cette approche. L’article montre aussi que les principes développés pour l’industrie minière sont facilement adaptables et applicables dans d’autres types d’industrie. Il discute aussi les bénéfices potentiels de la nouvelle approche ainsi que les défis
rencontrés pour y arriver et énumère les recherches futures. Cette manière d’aborder et analyser la GDA représente une nouveauté scientifique.

Amélioration du modèle intégral de la GDA et l’introduction des risques des événements extrêmes et rares en GDA

Komljenovic, D., Gaha, M., Abdul-Nour, G., Langheit, C., & Bourgeois, M. (2016).
Risks of extreme and rare events in asset management. Safety Science, 88, 129-145.
Le travail de recherche améliore le modèle initial. Il identifie et caractérise les facteurs d’influence critiques en GDA. L’article démontre à nouveau que la GDA est un système complexe adaptatif. Il développe spécifiquement pour la GDA le concept de la prise de décision en tenant compte des risques (RIDM). Le modèle amélioré intègre sept sous-modèles spécifiques et identifie les types d’incertitudes dominants dans chacun (aléatoire ou épistémique). Il identifie aussi les types de connections et liens entre les sous-modèles ainsi que leurs caractéristiques et leur niveau de complexité. L’approche pour caractériser les risques des événements extrêmes et rares en GDA est aussi élaborée. La recherche a démontré qu’il n’est pas approprié de calculer les probabilités extrêmement faibles des événements rares à cause des incertitudes et la possibilité de se tromper de plusieurs ordres de grandeur. Dans ce contexte, il est important d’appliquer le concept de robustesse et de résilience pour se protéger contre ce type de risques et assurer la continuité des
opérations de l’entreprise sans détruire sa viabilité économique. Deux études de cas (l’utilité électrique et l’industrie nucléaire) démontrent l’applicabilité et la validation de la méthodologie développée. L’article définit aussi les recherches futures d’intérêt pour approfondir le sujet. L’ensemble des éléments élaborés dans l’article représentent une contribution scientifique.

Prise de décision en gestion des actifs en tenant compte des risques et bde la complexité – modèle intégral

Komljenovic, D., Abdul-Nour, G., & Boudreau, J.F. (2018). Risk-informed decision-making in asset management as a complex adaptive system of systems.
International Journal of Strategic Engineering Asset Management (soumis pour la revue des pairs en janvier 2018).
L’article présente l’étape finale du développement du modèle intégral en gestion des actifs en élaborant une approche novatrice de la prise de décision en tenant compte des risques et de la complexité. La méthodologie développée se base sur les deux articles précédents et introduit la définition et séparation des rôles des analystes, experts et décideurs. La considération du niveau de connaissances des intervenants et son impact sur le résultat final est introduit à la méthodologie comme un facteur important. La délibération dans le processus décisionnel est introduite dans la méthodologie et son importance est démontrée. Le modèle décisionnel intègre aussi les risques d’événements extrêmes et rares dans l’évaluation globale des risques dans la prise de décision en GDA. La méthodologie propose l’introduction du concept de gouvernance de système complexe comme moyen de faire face à la complexité en GDA. L’approche développée montre aussi les limites des modèles quantitatifs et du danger de leur utilisation singulière comme base de la prise de décision en GDA.
Les recherches futures sont identifiées pour continuer à améliorer le modèle proposé. Le concept développé représente une contribution scientifique dans le domaine.

Case Studies

The risks of extreme and rare events in asset management will be analyzed through two cases of some Hydro-Québec’s assets. Hydro-Québec is one of the largest North American companies which generates, transmits and distributes electricity. Its sole shareholder is the government of Quebec. It uses mainly renewable generating sources, in particular large hydro units, and supports the development of other technologies such as wind energy and biomass (Hydro-Quebec, 2015a). Firstly, we analyze extreme interruptions in its power grid where reliable historical data are available (case a) of the risk analysis step IIb) presented above). They will be statistically characterizedthrough the extreme value theory. The risks and impacts of such large interruptions will be discussed in terms of operational and asset management challenges.
Secondly, we conduct an analysis with regard to the impact of rare, surprising events which led to the abandoning of the Gentilly-2 Nuclear Power Plant (NPP) Refurbishment Project, where no data were available (case b) in the analysis of risks step IIb) previously depicted).

Risks of Extreme Power Interruptions in the Grid

The Hydro-Quebec’s distribution grid is generally composed of overhead lines. Its underground grid is mainly installed in large urban areas such as Montreal and Quebec City, and represents a smaller part of the overall installations. Overhead lines are exposed to external events. Those events are typically weather generated ones, and usually cause tree or tree branch falls on power grid lines which trigger unplanned power interruptions. Furthermore, there are accidental animal, bird or human made interruptions which are comparatively low. Some other unplanned interruptions may come from the transmission grid disturbances which are small in numbers, but affect a large number of customers, and may last for a long time (Table and Table 5.16).
There are also interruptions caused by equipment failures, but they are relatively minor. All the power interruptions are recorded as CHI (Customer Hours ofInterruption) on a daily basis in an enterprise database. Other performance indicators are also calculated and used as per common practices (IEEE, 2012)1.
The enterprise records data upon all the interruption events, including extreme ones.
In this case, we may apply the theory of extreme value to characterize them (case a) presented above). Such external perturbing events usually originate from the submodel 6 (Figure 5.2, Tables 5.1 and 5.13). However, they primarily fell within investigations related to the sub-model 2 of the global RIDM model. Those extreme events affect the assets’ ability to fulfill their intended function. Obviously, perturbations within the sub-model 2 also touch other sub-models as per interdependencies and attributes shown in Tables 5.1 and 5.11. In such context, it is important to characterize the risks of extreme interruptions in the power grid since they negatively affect expected customer service, and mobilize important enterprise resources (human, material and financial) in order to get the service restored.
Knowledge upon the risks of the extreme grid events is of key importance from the asset management point of view regarding the continuity of service after a major interruption. The enterprise management team has to adequately plan the necessary resources in order to successfully handle such interruptions (contingency planning) to ensure the continuity of the service.
The analysis presented here aims at characterizing the risks of such extreme events, and discusses their impact on the overall asset management. It covers the timeframe from 1987 to 2015 inclusively (29 years in total), and takes into consideration unplanned power interruptions only.
The investigation of the interruptions indicates that approximately 93 % of the power interruption occurrences are inferior to 100 kCHI/day. However, the interruptions of 100 kCHI/day and over contributed to 65 % of the total interruption duration over the analyzed period of time. It gives the following ratios:
 35 % (durations)/93 % (number of days with less than 100kCHI/day) = 0.38;
 65 % (durations)/7 % (number of days with 100kCHI+/day) = 9.28.
The interruptions of 1 million CHI/day (MCHI+/day) and more represent approximately 0.6 % of the total number of interruptions. Meanwhile, they contributed to 42 % of the total interruption durations. Those values highlight the ratios:
 58 % (durations)/99.4 % (number of days with less than 1MCHI/day) = 0.58;
 42 % (durations)/0.6 % (number of days with 1MCHI+/day) = 66.86.
These values mean that each percentage in the number of interruptions of less than 100kCHI/day contributes 0.38 % of the total interruption duration (0.58 % for less than 1MCHI/day). On the other hand, each percentage in the number of interruptions for 100kCHI+/day contributes to 9.28 % regarding the total interruption duration (66.86 % for 1MCHI+/day).
The above numbers clearly demonstrate that the power interruptions of large magnitudes are less frequent, but their impact to the interruption durations is substantial. Thus, the above facts show the importance of adequate preparations in handling extreme interruption events. These findings will be analyzed more in depth below.

Statistical Distribution of Extreme Power Interruptions

Table 5.15 displays the maximal daily interruptions in (CHI/day)max on an annual basis (one maximal value per year is selected for the calculations). The maximal observed value is more than 46.35MCHI/day in 1998, and the minimal one is 380,606 CHI/day in 2004 (ratio of 122). It represents a range of almost 46MCHI/day. Table 5.16 provides selected causes that have been at the origin of some extreme interruptions based on HQ’s records.

Other characteristics and parameters of this distribution may be found in relevant statistical references.

The characterization of the parameters of a Gumbel distribution regarding the maximum power interruptions has been performed as per the approach described by the U.S. National Institute of Standard and Technology, Statistical Engineering Division (2015).
An initial analysis has indicated that the maximal daily values of (CHI/day)max have rather an exponential relationship with the values [-ln(-ln(PV))] used in the method with a high correlation coefficient (R2 = 0.9931). In order to obtain a linear relationship, the natural logarithm of (CHI/day)max values has been calculated. Thus, the CDF described in equation (1) is modified by expressing it as the natural  logarithm.
Based on the enterprise’s records, the duration of extreme interruptions may vary between 1-2 days up to several weeks, as it was the case of the ice storm in 1998.
The consequence categories as per Table 5.14 encompass mainly the category G (loss of production), but other categories are also implicated: A, C, D, E, H and I. Thus, the severity of the consequences varies between major and catastrophic (see Table 5.14). Considering the results of this analysis, we proposed a risk matrix for E&RE with regard to the loss of production, i.e. the category G as a dominant category (Figure 5.5). The probability levels are defined for 5 or less, 15, 40 and 60+ MCHI/day calculated through Equation (4). Theoretically, a maximum daily CHI is around 100 million for the current number of customers in Quebec (Hydro- Quebec, 2015a). It represents a total blackout in Quebec for 24 hours. Consequently, the levels of interruptions considered in the risk matrix reflect the percentages of the loss production for the category G in Table 5.14 (less than 15 %, 40 %, 60 % and more). The risk matrix has four levels of risk: low, moderate, high and very high.
For example, an interruption of 60MCHI+/day has a probability of 1.6 % (Table 5.17), and a catastrophic (CT) severity category for the loss of production (G) (Table 5.14). It should be seen as a moderate risk event (MOR) (Figure 5.5).
The obtained results can serve as risk insights and input into the holistic RIDM model (Figure 5.2, Table 5.1). They are beneficial in refining the enterprise’s approach concerning its contingency planning with regards to the levels of risks, and based on the BCM requirements.

Refurbishment of the Gentilly-2 Nuclear Power Plant

In the second case study, we have performed an analysis with regard to the impact of a series of rare events leading to the abandoning of the Gentilly-2 Nuclear Power Plant (NPP) Refurbishment Project in 2012 (Komljenovic and Abdul-Nour, 2015).
The G2 NPP was the sole Hydro-Quebec’s nuclear generating utility. It was a CANDU6 nuclear power plant and had been designed for a 30-year service life, or more accurately 210 000 Equivalent Full Power Hours (EFPH) as it is the case of all CANDU nuclear generating stations (Canteach, 2016). It started its commercial operation in 1983, and should have reached this limit somewhere in 2013. For that reason, Hydro-Quebec has initiated prefeasibility studies in the early 2000s in order to examine the possibility to refurbish the station and to extend its useful life for another 30 years. In these years, a general trend in the nuclear power industry was oriented toward an extension of the operations beyond the initial service life. It was based on an accumulated operational experience and new scientific insights which showed that it was possible. Following these studies, Hydro-Quebec made a positive decision in 2008 to refurbish the station. Required engineering and field works started under the Refurbishment Project. The costs of those activities had been estimated at 1.9B$, and the refurbishment work should have started in March 2011.
The restart was foreseen for November 2012 (Hydro-Quebec, 2015b).
However, after initial works began, the Refurbishment Project has been delayed several times. Finally, Hydro-Quebec announced on October 3, 2012 the closure of the Gentilly-2 NPP at the end of 2012, and its decommissioning. It ended its commercial operation on December 28, 2012 (Government of Quebec, 2012; Hydro-Quebec, 2015b,c). What happened within the four years from the initial positive announcement to the abandoning of a project of such magnitude?
A series of unfavorable and rare events occurred over a short period of time which definitely contributed to overturning the initial decision – a “Perfect Storm” as Paté- Cornell (2012) called it. Such events and their combination represent significant epistemic uncertainties for a decision-making process. They are almost impossible to predict, or to mathematically characterize in a complex operational and business environment. Although numerous warning signals and precursors indicating that certain key influential factors went wrong were available, the enterprise’s capacity to react was rather limited. Table 5.18 provides a summary overview of the analysis.
The coincidence and the combination of these six major rare events resulted in a non-linear amplification of their aggregate risk. These events are not all independent between them, which adds to overall complexity and opacity of the context (also see Table 5.1). The aggregate risk is considerably superior to the simple sum of their individual risks and effects. The above analysis shows that the overall operational and business context of the G2 Refurbishment Project as an AsM activity behaved as a CAS, and the proposed model capture this feature. It was practically impossible to mathematically model such risks by using traditional approaches. Consequently, they do not enable to accurately forecast both the occurrence and the gravity of the consequences of such events, and their coincidence. Thus, a major AsM project, initially approved in 2008, was abandoned four years later. This case study illustrates how rare events and their unfavorable combination may generate risks that have the capability to entirely change or disrupt major decisions regarding asset management within a relatively short timeframe.
Even, in having the proposed model for the purpose of the analysis, Hydro-Quebec would have probably not avoided the closure of the NPP, and such scenarios are sometimes inevitable. However, it is worth mentioning that the enterprise has not used a prospective option to submit to the Canadian Nuclear Safety Commission (CNSC) a demand to extend the operation of G2 NPP beyond 210,000 EFPH without refurbishment. This alternative might have enabled an increased resilience and robustness of the overall enterprise through an extended exploitation period of the NPP allowing more time to prepare its closure, and manage organizational and technical challenges involved. The recent CANDU industry experience from Ontario Power Generation (OPG) and Bruce Power (BP) has exposed this possibility. Following detailed studies and clear demonstration of the safety case, CNSC granted an extension to 247,000 EFPH to OPG’s Pickering Nuclear Generating Station, and to 245,000 EFPH to the Bruce Power NPPs Bruce B Units.

Conclusion

Enterprises worldwide are constantly forced to produce more at lower costs. They are also confronted with a highly complex business and operational environment, and this complexity keeps growing. As per recent industry-wide development, asset management plays a key role in this context.
In such circumstances, industries tend to develop various processes and approaches, which may enable to efficiently address these issues and manage associated risks. They are often based on traditional methods which are generally unable to adequately grasp and tackle the complexities and uncertainties. It is particularly true when considering extreme and rare events which have capabilities to disrupt strategic activities or even jeopardize the survival of enterprises.
We claim that the modern enterprises and their asset management strategy should be considered as Complex Adaptive Systems (CAS), which should be modeled through methods and tools of the complexity science. In such systems, the occurrence of extreme and rare events is very plausible since we do not entirely grasp their scope, nature of associated epistemic uncertainties and risks, and connections between their constituent elements.
This study presents a holistic high level Risk-Informed Decision-Making (RIDM) model (framework) in asset management and initial results on how to tackle the risks of E&RE within this context. Such a methodology may assist decision-makers in key decision-makings by providing more realistic insights. The proposed method may positively complement existing traditional approaches.
The two case studies related to Hydro-Quebec’s assets demonstrate the relevance of considering more systematically the complexity and risks of extreme and rare events in asset management.
Future research works should be directed to a deeper understanding of the complexity in AsM, and the development of AsM models using modeling and simulation techniques of the complexity science. This research should also include: the development of adequate stress tests and models for risk exposure from E&RE in AsM and their validation, enhanced risk assessment methods for E&RE in AsM, improved characterization of associated uncertainties, development of algorithms for efficiently generating E&RE in simulation models, efficient ways of improving resilience and robustness in AsM while remaining economically viable, modeling the role of organizational, and human performance, biases and behavior in generating risks from E&RE. Future research works also ought to investigate how to better capture opportunities from E&RE. It is necessary to analyze more in detail the links and complementarities between AsM and BCM as well.
Furthermore, it is indispensable to highlight challenges that the development and application of the proposed approach may encounter. They include, but are not limited to:
 Lack of appropriate analysis and modeling methods, scientific understanding of the complexity and the risks from extreme and rare events in AsM; continuous increase in the overall complexity makes this task more difficult;
 Availability of pertinent data in order to perform the required analyses. Further investigations are needed to determine which data are really needed and whether the quality of the available data is satisfactory. Collecting and preparing them could imply considerable efforts;
 Availability of decision-making support models and tools: they have to be developed and tailored according to the needs of a specific organization/industry. This research may also require an adaptation of existing traditional tools to better fit novel methods and approaches;
 Costs of integrating the complexity framework and new risk assessment methods regarding extreme and rare events in AsM may be consequential (new research, data collection, development of methods and tools, their implementation, training, maintenance, etc.);
 Acceptability of novel approaches by the industry: introduction of new ways of performing analyses or decision-making may face resistance and unwillingness to embrace them.

Abstract

Decision-making is an essential activity in Asset Management. It is influenced by various factors (strategic, technical/technological, economic, organisational, regulatory, safety, markets, etc.). Sound decision-making in AsM ought to take into account relevant factors in order to balance risks, opportunities, performance, costs, and benefits. Additionally, modern organisations evolve in complex operational and business environments and are exposed to significant uncertainties. In such a context, decision-making in AsM becomes more challenging. This study proposes a holistic three-step Risk-Informed Decision-Making (RIDM) methodology developed for AsM, considering it as a Complex Adaptive System of Systems. The methodology is applied in a case study to analyse possible modification strategies for a nuclear power plant’s emergency core cooling system. Through the RIDM process, quantitative models and other factors have been taken into account in order to obtain the necessary comprehensive insights regarding the decision to be made. Keywords: asset management, complex adaptive systems, uncertainties, riskinformed decision-making.

Introduction

Asset Management (AsM) has become widespread among contemporary enterprises and organisations as an effective approach allowing to deliver value from assets and to ensure the sustainability of the business and its operations (The Institute of Asset Management, 2015a; Komljenovic et al., 2016; Hastings, 2010). This concept becomes particularly relevant considering the globalisation and increased competition which characterises markets worldwide.
The advances of AsM experience and the accumulated knowledge across various industries resulted in a new International Standard on AsM, the standard ISO 55000 (ISO, 2014a,b,c).
In practice, AsM is sometimes depicted as being essentially related to maintenance and reliability (The Institute of Asset Management, 2015a). However, AsM covers much more than these two fields. It is defined in this Standard as a coordinated activity of an organisation to realise the value of assets. The ISO Technical Committee for Asset Management Systems, ISO/TC251 clarifies the difference between the concepts of Managing Assets and Asset Management (ISO, 2017). It highlights that over the years, people, organisations and enterprises have developed whole disciplines to help define the best ways to care for assets throughout their useful lives. As such, they have been Managing Assets for a long time. Meanwhile, with the introduction of the formal discipline of Asset Management roughly 20 years ago, structured approaches have been developed, which assure stakeholders that those care activities are focused on deriving value for the organisation and not just promoting best asset care arrangements. In this regard, Asset Management and Managing Assets are not alternatives. Contemporary enterprises operate in a market, natural, technical, technological, organisational, regulatory, legal, political and financial environment (hereafter called “business and operational environment”), which is complex and characterised by significant risks and uncertainties. Furthermore, modern enterprises themselves are complex because of their organisational, management and operational structure (particularly the larger ones), which also increases the overall complexity and uncertainties (Komljenovic et al., 2015; El-Thalji and Liyanage, 2015; Beer and Liyanage, 2014; Harvey and Stanton, 2014; Komljenovic et al., 2016; Stacey and Mowles, 2016; Rzevski and Skobelev, 2014).
In contemporary organizations, assets and their systems, exhibit characteristics of complexity, interdependence, and dynamic emerging behaviour (Chopra and Khanna, 2015; The Institute of Asset Management, 2015a,b; EPRI, 2004). Zio (2016) introduces the notions of the structural and dynamic complexity of modern critical infrastructure such as energy transmission and distribution networks, telecommunication networks, transportation systems, water and gas distribution
systems, etc.
Consequently, modern organisations are fairly complex socio-technologicaleconomic entities involving many interacting and interdependent elements with hardly predictable long term behaviours at micro and macro levels. Anticipating and assessing such behaviours and dynamics requires extensive knowledge from multiple disciplines in engineering and beyond.
In this context, the decision-making process related to AsM may reveal very challenging due to significant uncertainties related to the nature and complexity of often conflicting influence factors. There are two types of uncertainties that should be taken into account in engineering and AsM: aleatory (arises when an event occurs randomly) and epistemic (has been referred to as a state-of-knowledge uncertainty).
More details may be found in relevant references (Kumamoto, 2007; US NRC, 2013; Komljenovic et al., 2016; EPRI, 2015; ISO, 2009a).
Modern organisations attempt to address these issues by using various models and tools that help decrease uncertainties and better quantify risks within their asset management decision-making process.

Le rapport de stage ou le pfe est un document d’analyse, de synthèse et d’évaluation de votre apprentissage, c’est pour cela rapport-gratuit.com propose le téléchargement des modèles complet de projet de fin d’étude, rapport de stage, mémoire, pfe, thèse, pour connaître la méthodologie à avoir et savoir comment construire les parties d’un projet de fin d’étude.

Table des matières

CHAPITRE 1  INTRODUCTION
1.1 Mise en contexte
1.2 Problématique de la recherche
1.3 Questions, but et objectifs de la recherche
CHAPITRE 2  REVUE DE LITTÉRATURE
2.1 Systèmes complexes et la théorie (science) de la complexité
2.2 Prise de décisions en tenant compte des risques (RIDM)
2.3 Gestion des actifs et processus décisionnel en GDA
2.4 Incertitudes en ingénierie et leur impact
2.5 Prévisions et risques des événements extrêmes et rares
CHAPITRE 3  MÉTHODOLOGIE DE RECHERCHE
3.1 Introduction à la méthodologie
3.2 Structure et contribution scientifique des articles
CHAPITRE 4  ARTICLE 1
4.1 Introduction
4.2 Overview of the Complexity Science and Complex Adaptive Systems
4.3 Literature review
4.4 Global Model of Strategic Planning and Asset Management in Mining in the Context of Complexity
4.4.1 Global Model
4.4.2 Modelling Methods
4.5 Discussion
4.5.1 New insights through complexity science
4.5.2 Anticipated benefits
4.5.3 Challenges
4.5.4 Future research
4.6 Conclusion
4.7 References
CHAPITRE 5  ARTICLE 2 
5.1 Introduction
5.2 Literature review
5.2.1 Asset Management
5.2.2 Extreme and Rare Events
5.2.3 Complex Adaptive Systems and the Complexity Theory
5.3 Model for characterizing the risks of extreme and rare vents in assetnmanagement
5.3.1 Global RIDM Model in Asset Management
5.3.2 Model for Assessing Risks of Extreme and Rare Events in Asset Management
5.4 Case Studies
5.4.1 Risks of Extreme Power Interruptions in the Grid
5.4.1 Refurbishment of the Gentilly-2 Nuclear Power Plant
5.5 Conclusion
5.6 References
CHAPITRE 6  ARTICLE 3 
6.1 Introduction
6.2 Asset Management in different industries – a literature review
6.3 Risk-Informed Decision-Making Model in Asset Management
6.3.1 Decision-Making in AsM: General Considerations
6.3.2 Decision diamond
6.3.3 Description of the Decision-Making Steps in Asset Management
6.4 Case Study
6.4.1 Step 1: Setting the framework
6.4.2 Step 2: Detailed analyses
6.4.3 Step 3: Deliberations, decision-making, communication, and implementation
6.5 Conclusions
6.6 References
Appendix 6.A
CHAPITRE 7  CONCLUSIONS
7.1 Discussion générale
7.2 Contributions scientifiques
7.3 Bénéfices anticipés
7.4 Défis futurs et limites de la méthodologie
7.5 Recherches futures
RÉFÉRENCES 
ANNEXE 1

Télécharger aussi :

Laisser un commentaire

Votre adresse e-mail ne sera pas publiée. Les champs obligatoires sont indiqués avec *