Les travaux de recherche présentés dans ce document s'inscrivent dans le cadre de la gestion des connaissances appliquée à la déconstruction des avions en fin de vie avec pour objectif l'aide à la décision par l'évaluation des risques. Pour répondre à cet objectif, nous avons développé des mécanismes d'aide à la décision hybridant les retours d'expérience statistique et cognitif pour évaluer les risques sur les zones critiques d'un système. L'approche proposée permet la combinaison des avis d'experts du domaine avec des statistiques issues d'une base de données en utilisant les fonctions de croyance. L'évaluation des risques est réalisée par le traitement des connaissances combinées au moyen d'un modèle utilisant les réseaux évidentiels dirigés. Ce document s'articule en quatre chapitres.Le premier chapitre constitue un état de l'art abordant les notions liées au risque et au retour d'expérience. Il permet de définir les concepts clés concernant l'évaluation du risque, la gestion des connaissances (et en particulier le processus de retour d'expérience) ainsi que les passerelles entre ces deux concepts. Le second chapitre permet d'introduire un modèle d'évaluation des risques basé sur les méthodes bayésiennes. Cependant, les méthodes bayésiennes ont des limites, en particulier pour ce qui concerne la modélisation de l'incertitude épistémique inhérente aux avis d'experts, qui nous ont incité à proposer des alternatives, telles les fonctions de croyance et les réseaux évidentiels dirigés que nous avons finalement choisi d'utiliser. Le troisième chapitre propose une démarche permettant d'évaluer les risques en utilisant les réseaux évidentiels dirigés. L'approche proposée décrit les mécanismes utilisés pour formaliser et fusionner les connaissances expertes et statistiques, puis pour traiter ces connaissances au moyen des réseaux évidentiels dirigés. Pour finir, des indicateurs permettant la restitution des résultats au décideur sont introduits. Le dernier chapitre présente le projet DIAGNOSTAT qui a servi de cadre à ces travaux de recherche et expose un cas d'étude permettant d'appliquer la démarche introduite précédemment à la déconstruction des avions en fin de vie au moyen de deux scénarios / The research work presented in this document relates to knowledge management applied to aircraft deconstruction. The aim of this work is to provide a decision support system for risk assessment. To meet this objective, mechanisms for decision support hybridizing cognitive and statistical experience feedback to perform risk assessment on system critical areas have been developed. The proposed approach allows to combine expert opinion with statistics extracted from a database by using belief functions. The risk assessment is performed by the combined knowledge processing using a model based on directed evidential networks. This document is divided into four chapters. The first chapter is a state of the art addressing concepts related to risk and experience feedback. It defines key concepts for risk assessment, knowledge management (in particular the experience feedback process) and the links between these two concepts. The second chapter allows to introduce a risk assessment model based on Bayesian methods. However, Bayesian methods have some limitations, particularly with respect to epistemic uncertainty modelling. That is why, some alternatives have been proposed, such as belief functions and directed evidential networks that we finally chose to use. The third chapter proposes an approach for assessing the risk using directed evidential networks. The proposed approach describes the mechanisms used to formalize and combine expert and statistical knowledge, and then to process this knowledge with directed evidential networks. Finally, indicators to inform the decision maker about results are introduced. The last chapter presents the DIAGNOSTAT project which provided the framework for this research and a study case to apply the approach introduced earlier for aircraft deconstruction by using two scenarios

La fusion d'informations issues de plusieurs sources cherche à améliorer la prise de décision. Pour réaliser cette fusion, la théorie des fonctions de croyance utilise des règles de combinaison faisant bien souvent l'hypothèse de l'indépendance des sources. Cette forte hypothèse n'est, cependant, ni formalisée ni vérifiée. Elle est supposée pour justifier le choix du type de règles à utiliser sans avoir, pour autant, un moyen de la vérifier. Nous proposons dans ce rapport de thèse un apprentissage de l'indépendance cognitive de sources d'information. Nous détaillons également une approche d'apprentissage de la dépendance positive et négative des sources. Les degrés d'indépendance, de dépendance positive et négative des sources ont principalement trois utilités. Premièrement, ces degrés serviront à choisir le type de règles de combinaison à utiliser lors de la combinaison. Deuxièmement, ces degrés exprimés par une fonction de masse sont intégrés par une approche d'affaiblissement avant de réaliser la combinaison d'information. Une troisième utilisation de cette mesure d'indépendance consiste à l'intégrer dans une nouvelle règle de combinaison. La règle que nous proposons est une moyenne pondérée avec ce degré d'indépendance. / The theory of belief functions manages uncertainty and proposes a set of combination rules to aggregate beliefs of several sources. Some combination rules mix evidential information where sources are independent; other rules are suited to combine evidential information held by dependent sources. Information on sources ' independence is required to justify the choice of the adequate type of combination rules. In this thesis, we suggest a method to quantify sources' degrees of independence that may guide the choice of the appropriate type of combination rules. In fact, we propose a statistical approach to learn sources' degrees of independence from all provided evidential information. There are three main uses of estimating sources' degrees of independence: First, we use sources' degree of independence to guide the choice of combination rules to use when aggregating beliefs of several sources. Second, we propose to integrate sources' degrees of independence into sources' beliefs leading to an operator similar to the discounting. Finally, we define a new combination rule weighted with sources' degree of independence.

Sammanfattning   Revision är ett än mer aktuellt ämne nu än på flera decennier då det debatteras vitt och brett kring vad en revisor skall eller bör göra. I och med de tidigare införda byrårotationskraven begränsar det företag av allmänt intresse att inte under mer än maximalt tio år använda samma revisionsbyrå, med chans till förlängning. Vi har därför valt att undersöka hur längden på revisionsuppdragen kan inverka på stora onoterade bolag i en mindre geografisk zon, i vårt fall Umeå kommun med omnejd genom att intervjua tre revisorer och deras klienter.   Baserat på problemdiskussionen har vi bestämt följande problemformulering för studien.      Hur upplever revisionsbyråerna och deras klienter att oberoendet samt revisionskvalitet samspelar med revisionsprocessen vid korta respektive långa revisionsuppdrag?   Syftet med vår studie är att undersöka hur förhållandet mellan klient och revisor upplevs ha en samverkan med kvaliteten i revisionen samt på revisorernas oberoende till deras klienter över tid. Detta kan ligga till grund för framtida diskussioner gällande byrårotation för huruvida oberoende och revisionskvalitet utvecklas över tid under revisionsuppdraget.   Med en kvalitativ ansats har vi genomfört nio intervjuer med tre revisorer samt sex klienter. Klientförhållandena till respektive revisor är ett långvarigt revisionsuppdrag (fler än fem år) och ett kortvarigt revisionsuppdrag (mindre än 5 år sedan övertagandet). Denna unika inblick i revisor-klientförhållandet kopplas till flertalet faktorer samt teorier vilka kan ha en samverkan och förklaring till förändringen i oberoendet samt revisionskvaliteten.    Vi kommer i vår studie fram till att revisionsprocessen ser väldigt lik ut vid korta samt långa revisionsuppdrag där första året är speciellt omfattande, men år 2 och framåt vanligtvis är lika omfattande år efter år. Vad gäller oberoende finner vi att desto längre relation som skapas mellan klient och revisionsbyrå desto mer omfattande blir rådgivningen till klienten vilket kan påverka revisionsbyråns oberoende mot klienten rent finansiellt. Vidare är revisionskvalitet en term som definieras olika, men i teorin hävdar flertalet att detta kan speglas av oberoendet till klienten. I vår studie har definitionen uppfattats olika mellan revisor och klient vilket visar på ett förväntningsgap mellan de båda. Enligt respondenterna är revisionskvalitet något som skulle kunna förbättras av byrårotation när flera anser att långa revisionsuppdrag kan bli för rutinmässiga och att det skulle vara bra om nya individer kommer in och tittar på uppdraget från ett nytt perspektiv.   Förslag till vidare forskning skulle till exempel vara en enkätundersökning med ett mer omfattande geografiskt område där fler företag av större storlek innefattas av populationen för att kunna jämställa dessa med de företag som redan omfattas av reglerna om obligatorisk byrårotation.   Nyckelord: Audit, big four, big seven, audit quality, audit independence, audit firm rotation, auditor rotation, expectation gap, evidential matter, audit procedures och audit tenure.

Este trabalho explora três possíveis significados para o \"romance histórico\" Amphitryon, de Ignacio Padilla, e, a partir disso, estuda modos de interação entre história e ficção. O primeiro sentido analisa-o como um livro na história, uma obra marcante que se escreveu sob o contexto de um manifesto célebre por tentar reorganizar a tradição literária latinoamericana. O segundo sentido toma-o como um livro de história, um romance que não só se aproxima de um gênero literário afeito aos fatos historiográficos, mas que, sobretudo, articula conteúdos históricos, literários e culturais sob formas narrativas que também derivam do século passado. O terceiro sentido, finalmente, o lê como um livro da história, uma narrativa que, pela disposição de relatos e narradores, simula o próprio mecanismo do fazer histórico. Conduzindo essas argumentações estão os princípios da dialética formaabertura e da interação autor-obra-leitor, derivados de Umberto Eco, a noção de leitura e escritura como forma de conhecer, cara a Jorge Luis Borges, e o paradigma indiciário, de Carlo Ginzburg. A tentativa final é fazer de um exercício de crítica literária uma reflexão sobre a história. Para tanto, insiste-se na analogia entre verificação e interpretação de dados e as mediações livro-leitor-leituras, e adota-se a \"espiral de leituras\" historicizadas como modo operativo que aproxima discursos ficcionais e discursos históricos, esboçando-se, então, paralelos e limites nos percursos da produção historiográfica e da produção ficcional ao longo do século XX. / This work explores three possible meanings for the \"historical novel\" Amphitryon, of Ignacio Padilla, and, from this, it studies ways of interaction between history and fiction. The first meaning analyzes it as a book in history, a remarkable work that was written under the context of a manifest notable for trying to reorganize the Latin American literary tradition. The second meaning takes it as a history book, a novel that not only comes close to the historiographies facts, but, above all, it articulates historical, literary and cultural contents under narrative forms that are also drawn from the last century. The third meaning, finally, reads it as a book of history, a narrative that, for the disposal of stories and narrators, simulates the mechanism of history. As a guide line for these arguments, there are the principles of the dialectic form-opening and the interaction author-work-reader, derived from Umberto Eco, the notion of reading and writing as forms of knowledge, from Jorge Luis Borges, and the evidential paradigm, of Carlo Ginzburg. The final attempt is to make a reflection on history from a literary critical exercise. In such way, one must insist on the analogy between verification and interpretation of data and the relations bookreader- readings, and adopts the historic \"spiral of readings\" as an operative way that approaches fiction and historical speeches, and so, outlining parallels and limits in the course of the historiography and fictional production throughout the 20th century.

ERP system has shaped the way modern organisations design, control, and execute business processes. It has not only paved the way for efficient use of organisational resources but also offered the opportunity to utilise data logged in the system for ensuring internal control. The key contribution of this research is that it has resulted in a method which can practically be employed by internal auditors for measuring internal fraud risk of business organisations with ERP systems, by utilising process mining technique and evidential reasoning in the form of Bayesian theorem, in a much more effective way compared to conventional frequentist method. The other significant contribution is that it has paved the way for combining process mining technique and evidential reasoning in addressing problems prevalent within organisational contexts. This research has contributed in developing IS theories for design and action especially in the area of soft systems methodology as it has relied on business process modelling in addressing the issue of internal fraud risk. The chosen method has contributed in facilitating incorporation of design science method in problem solving. Researchers have focused on applying data mining techniques within organisational contexts for extracting valuable information. Process mining is a comparatively new technique which allows business processes to be analysed based on event logs. Analysis of business processes can be useful for organisations not only for attaining greater efficiency but also for ensuring internal control inside the organisation. Large organisations have various measures in place for ensuring internal control. Measuring the risk of fraud within a business process is an important practice for preventing fraud as accurate measurement of fraud risk provides business experts with the opportunity to comprehend the extent of the problem. Business experts, such as internal auditors, still heavily rely upon conventional methods for measuring internal fraud risk way by of random check of process compliance. Organisations with ERP systems in place can avail themselves of the opportunity to use event logs for extending the scope of assessing process conformance. This has not been put into practice as there is a lack of well researched methods which can allow event logs to be utilised for enhancing internal control. This research can be proved to be useful for practitioners as it has developed a method for measuring internal fraud risk within organisations. This research aimed to utilise process mining technique that allows business experts to exert greater control over business process execution by allowing the internal fraud risk to be measured effectively. A method has been developed for measuring internal fraud risk of business originations with ERP systems by using process mining and Bayesian theorem. In this method, rate of process deviation is calculated by conducting process mining on relevant logs of events and then that process deviation rate is applied in Bayesian theorem along with historic internal fraud risk rate and process deviation rate calculated manually for arriving at a revised internal fraud risk rate. Bayesian theorem has been relied upon for the purpose of developing this new method as it allows evidential reasoning to be incorporated. The method has been developed as a Design Science Research Method (DSRM) artefact by conducting three case-studies. Data has been collected from three case companies, operating in readymade garments manufacturing industry, pharmaceuticals industry, and aviation industry, regarding their procurement process for conducting process mining. The revised internal fraud risk rates were then evaluated by considering the feedback received from respective business experts of each of the case company. The proposed method is beneficial as it has paved the way for practitioners to utilise process mining using a soft system methodology. The developed method is of immense significance as it has contributed in the field of business intelligence and analytics (BI&A) and the big data analytics which have become significantly important to both academics and practitioners over the past couple of decades.

Cette thèse constitue la première étude générale de la modalité en néo-égyptien. Le chapitre introductif (p. 5-43) est consacré [1] à la définition de ce premier état de langue de légyptien de la seconde phase ; cette définition a permis la délimitation dun corpus servant dassise empirique à létude (la répartition du corpus en fonction de critères chronologiques et géographiques, de la nature du support et des "Textsorten" a donné la possibilité de pondérer et dobjectiver les analyses proposées pour chaque expression de la modalité). Ensuite, [2] un cadre théorique général pour létude de la langue est discuté. Le corps du travail se divise en trois parties consacrées respectivement : [1] à une définition générale de la notion de modalité (cela afin de déterminer les media expressifs qui relèvent de son étude en néo-égyptien) ainsi quà létablissement dun modèle sémantique à la fois économique, cohérent et correspondant aux données typologiques (p. 44-115) ; [2] à létude des modalités radicales (i.e. les modalités déontiques et bouliques en envisageant les relations quelles entretiennent avec le domaine axiologique ; p. 116-341) ; [3] à lexamen des modalités assertives (p. 342-446) : [a] analyse des formes de complémentation, en ce compris les liens entre intégration syntaxique, variation de lassertivité et degré de manipulation, [b] étude de limpact des auxiliaires dénonciation sur le degré dassertivité dune proposition, [c] critique des théories existantes concernant les moyens expressifs du discours indirect en néo-égyptien. Les conclusions (p. 447-466) sont accompagnées de propositions prospectives devant permettre [1] de rendre le modèle défendu applicable à létude des complexes conditionnels, [2] dintégrer la dimension énonciative dans lanalyse des relations interpersonnelles, [3] de proposer une approche globale des media expressifs de la causalité et de la finalité.

The aim of this thesis is to explore the adaptability of the Evidential Reasoning (ER) Rule as a method to provide a useful supporting tool for helping investors make decisions on financial investments. Decision making in financial investment often involves conflicting information and subjective judgment of the investors. Accordingly, the ER Rule, extended from the original popular Evidential Reasoning algorithm and developed for MCDM (Multiple Criteria Decision Making), is particularly suited for handling conflicts in information and to allow for judgmental weighting on the sources of evidence. In order to do so, a specific EIA (Efficient Information Assessment) process modeled by the mass function of Dempster-Shafer Theory has been constructed such that the underlying architecture of the model satisfies the requirement of the ER rule. The fundamental concern is to define and assess “efficient information”. For this purpose, a process denoted the Efficient Information Assessment (EIA) is defined which applies the mass function of Dempster-Shafer theory. Any relevant information selected from an expert’s knowledge database is “efficient” if the data is fully in compliance with the requirement of the ER rule. The logical process of the EIA model proceeds with a set of portfolio strategies from the information recommended by top financial analysts. Then, as a result, the model enables the ER rule to make an evaluation of all strategies for helping investors make decisions. Experiments were carried out to back-test the investment strategy using data from the China Stock Market & Accounting Research (CSMAR) Database for the four-year period between 2009 and 2012. The data contained more than 270,000 reports from more than 4,600 financial analysts. The risk-adjusted average annual return of the strategy outperformed that of the CSI300 index by as much as 10.69% for an investment horizon of six months, with the p value from Student’s t-test as low as 0.02%. The EIA model serves as the first successful application adapting the ER Rule for a new and effective decision-making process in financial investment, and this work is the only empirical study applying the ER Rule to the opinions of financial analysts, to the best of my knowledge.

The crime of human trafficking is a crime of complex nature and presents problems in its interpretation in the Peruvian law and jurisprudence. These problems require the author to study the legally protected goods behind this criminal prohibition, the consent of minors, analysis of the previous domain relationship above the victim or caused by the perpetrator, and the evidential approach that this reinterpretation of the crime of trafficking assumes. According to the previous legal framework, the author stands his proposal, which is characterized by criticism and being contrary to the Peruvian supreme jurisprudence. / El delito de trata de personas es un delito de naturaleza compleja y presenta problemas dentro de su interpretación en la jurisprudencia peruana. Estos problemas obligan al autor a estudiar la delimitación del bien jurídico protegido detrás de este delito, el consentimiento de menores de edad, el análisis de la relación de dominio que antecede o es provocada por el agresor sobre la víctima y el enfoque probatorio que supone esta reinterpretación del delito de trata de personas. De acuerdo con el marco jurídico anterior, el autor analiza casos observados por la Corte Suprema peruana y las diversas posturas doctrinales para finalmente emitir una postura propia, la cual se caracteriza por ser crítica y contraria a la interpretación presente en la jurisprudencia suprema peruana.

Este trabalho explora três possíveis significados para o \"romance histórico\" Amphitryon, de Ignacio Padilla, e, a partir disso, estuda modos de interação entre história e ficção. O primeiro sentido analisa-o como um livro na história, uma obra marcante que se escreveu sob o contexto de um manifesto célebre por tentar reorganizar a tradição literária latinoamericana. O segundo sentido toma-o como um livro de história, um romance que não só se aproxima de um gênero literário afeito aos fatos historiográficos, mas que, sobretudo, articula conteúdos históricos, literários e culturais sob formas narrativas que também derivam do século passado. O terceiro sentido, finalmente, o lê como um livro da história, uma narrativa que, pela disposição de relatos e narradores, simula o próprio mecanismo do fazer histórico. Conduzindo essas argumentações estão os princípios da dialética formaabertura e da interação autor-obra-leitor, derivados de Umberto Eco, a noção de leitura e escritura como forma de conhecer, cara a Jorge Luis Borges, e o paradigma indiciário, de Carlo Ginzburg. A tentativa final é fazer de um exercício de crítica literária uma reflexão sobre a história. Para tanto, insiste-se na analogia entre verificação e interpretação de dados e as mediações livro-leitor-leituras, e adota-se a \"espiral de leituras\" historicizadas como modo operativo que aproxima discursos ficcionais e discursos históricos, esboçando-se, então, paralelos e limites nos percursos da produção historiográfica e da produção ficcional ao longo do século XX. / This work explores three possible meanings for the \"historical novel\" Amphitryon, of Ignacio Padilla, and, from this, it studies ways of interaction between history and fiction. The first meaning analyzes it as a book in history, a remarkable work that was written under the context of a manifest notable for trying to reorganize the Latin American literary tradition. The second meaning takes it as a history book, a novel that not only comes close to the historiographies facts, but, above all, it articulates historical, literary and cultural contents under narrative forms that are also drawn from the last century. The third meaning, finally, reads it as a book of history, a narrative that, for the disposal of stories and narrators, simulates the mechanism of history. As a guide line for these arguments, there are the principles of the dialectic form-opening and the interaction author-work-reader, derived from Umberto Eco, the notion of reading and writing as forms of knowledge, from Jorge Luis Borges, and the evidential paradigm, of Carlo Ginzburg. The final attempt is to make a reflection on history from a literary critical exercise. In such way, one must insist on the analogy between verification and interpretation of data and the relations bookreader- readings, and adopts the historic \"spiral of readings\" as an operative way that approaches fiction and historical speeches, and so, outlining parallels and limits in the course of the historiography and fictional production throughout the 20th century.

The real estate selection process might be regarded as a typical Multi-Criteria Decision Making (MCDM) problem. With current literature concentrating predominantly on institutional investment decision making, additional effort should be directed towards studying inexperienced homebuyers who want to buy a property in which to live. In this context, authors have described the decision environment as a complex decision process with restricted access to property data, high financial burdens partially due to the illiquid nature of the investment, the unfamiliarity with the decision task and low transparency in information aggregation. Consequently, this situation could benefit from a more structured approach that assists homebuyers in their actions. In order to guide the decision making process and provide a suitable support mechanism, it is necessary to first structure the problem and extract the required information. A thorough literature review shows that little guidance is available for MCDM problem structuring. Consequently, this research first proposes an MCDM problem-structuring framework to decompose complex problems into smaller parts. Foremost, the application is intended for high-involvement consumer products and services. This framework is derived from MCDM and methodology literature, where the former provides the elements that need to be defined in any MCDM problem situation, and the latter suggests suitable data collection and analysis methods to obtain the information. As a result, the first contribution to existing literature is the introduction of an MCDM problem-structuring framework, which consists of a carefully designed sequential exploratory mixed method procedure. Next, following the proposed structure, the real estate selection problem in Majorca (Spain) is defined. Whilst providing the inherent problem elements and establishing a comprehensive list of evaluation criteria to assess luxury properties, the fieldwork also offers behavioural insights, contributing and supplementing existing real estate research. In particular, major misunderstandings and false assumptions during real estate agent and client interactions are observed, stressing the need to optimise communication and targeting strategies. On the basis of the relevant real estate evaluation criteria, a dataset of alternative houses is created and subsequently rated by prospective luxury-homebuyers. This provides the basis for the third research focus, the construction of a decision support model for real estate selection. In accordance to the problem features and model requirements, the Evidential Reasoning (ER) rule is identified to offer a powerful and transparent evidence aggregation process, with the potential to have a superior performance than other methods in addressing the selection decision. Due to the ER rule’s short history (2013), application studies in general are practically non-existing and unprecedented in the real estate domain. Therefore, the use of a modified ER model can provide the real estate literature with a prescriptive multi-criteria decision support mechanism, whilst simultaneously offering an application study for the MCDM community and other relevant decision analysis domains. In closing, modelling a real problem using the ER rule highlights the method’s advantages and might in turn increase awareness, leading to more applications.