Dissertations / Theses on the topic 'Decision logic'

Různé typy logik se často používají jako prostředky pro formální specifikaci systémů. Slabá monadická logika druhého řádu s k následníky (WSkS) je jednou z nich a byť má poměrně velkou vyjadřovací sílu, stále je rozhodnutelná. Ačkoliv složitost testování splnitelnosti WSkS formule není ani ve třídě ELEMENTARY, tak existují přístupy založené na deterministických automatech, implementované např. v nástroji MONA, které efektně řeší omezenou třídu praktických příkladů, nicméně nefungují pro jiné. Tato práce rozšiřuje třídu prakticky řešitelných příkladů, a to tak, že využívá nedávno vyvinutých technik pro efektní manipulaci s nedeterministickými automaty (jako je například testování universality jazyka pomocí přístupu založeného na antichainech) a navrhuje novou rozhodovací proceduru pro WSkS využívající právě nedeterministické automaty. Procedura je implementována a ve srovnání s nástrojem MONA dosahuje v některých případech řádově lepších výsledků.

The value of actual outcomes or states of affairs often depends on what could have been. Such dependencies create well-known “paradoxes” for decision theory, the best-known perhaps being the so-called Allais Paradox. The primary aim of this PhD thesis is to enrich decision theory such that it includes counterfactual prospects in the domains of desirability (or utility) functions, and show that, as a result, the paradoxes in question disappear. Before discussing the way in which counterfactual propositions influence the desirability of actual outcomes, I discuss the way in which the truth of one factual proposition influences the desirability of another. This examination leads me to reject the Invariance assumption, which states that the desirability of a proposition is independent of whether it is true. The assumption plays an important role in David Lewis’ famous arguments against the so-called Desire-as-Belief thesis (DAB). The unsoundness of Lewis’ argument does of course not make DAB true. In fact, I provide novel arguments against different versions of DAB, without assuming Invariance. To justify the assumptions I make when extending decision theory to counterfactual prospects, I discuss several issues concerning the logic, metaphysics and epistemology of counterfactuals. For instance, I defend a version of the so-called Ramsey test, and show that Richard Bradley’s recent Multidimensional Possible World Semantics for Conditionals is both more plausible and permissive than Bradley’s original formulation of it suggested. I use the multidimensional semantics to extend Richard Jeffrey’s decision theory to counterfactuals, and show that his desirability measure, extended to counterfactuals, can represent the various different ways in which counterfactuals influence the desirability of factual propositions. And I explain why the most common alternatives to Jeffrey’s theory cannot be similarly extended. I conclude the thesis by using Jeffrey’s extended decision theory to construct an ethical theory I call Modal Consequentialism, and argue that it better satisfies certain entrenched moral intuitions than Non-Modal Consequentialism (such as classical utilitarianism and welfare economics).

International Telemetering Conference Proceedings / October 26-29, 1987 / Town and Country Hotel, San Diego, California
A Binary Decision Machine (BDM) is described as a means of achieving logical control of data acquisition equipment and telemetry systems. The basic architecture of a BDM is initially presented followed by a description of its implementation as a Very Large Scale Integration (VLSI) device. Performance characteristics, programming, and ease of use as a controller are then presented via actual applications. The results of these endeavors led to a means of digitizing and extracting doppler data in a missile telemetry system.

Thesis (Ph. D.)--University of California, Riverside, 2008.
Includes abstract. Title from first page of PDF file (viewed March 10, 2010). Available via ProQuest Digital Dissertations. Includes bibliographical references (p. 166-170). Also issued in print.

During last three decades, companies have offshored their manufacturing activities across international borders in order to pursue lower manufacturing costs. Despite having accomplished their purpose, companies have also suffered from issues, especially poor quality of products and a poor response to customer demand. Therefore, companies consider relocating some of the manufacturing activities back to the home country, a process that is known as manufacturing reshoring. There is paucity of scholarly attention on how manufacturing reshoring decisions are evaluated and supported. Therefore, the purpose of this thesis is to develop decision-support tools to evaluate manufacturing reshoring decisions. In order to fulfil this, it is important to know how industry experts reason while making manufacturing reshoring decisions (RQ1), and how their reasoning can be modeled into decision-support tools (RQ2). Therefore, three studies were conducted including a multiple case study and two modeling studies. The multiple case study addressed the criteria that are considered by the industry experts in these decisions, while the two modeling studies, based on fuzzy logic and analytical hierarchy process (AHP), used a part of these criteria to develop decision-support tools. The findings indicate that a holistic set of criteria were considered by industry experts in arriving at a manufacturing reshoring decision. A large portion of these criteria occur within competitive priority category and among them, high importance is given to quality, while low importance to sustainability. Fuzzy logic modeling was used to model the criteria from the perspective of competitive priority at an overall level. Three fuzzy logic concepts were developed to capture industry experts’ reasoning and facilitate modeling of manufacturing reshoring decisions. Furthermore, two configurations and sixteen settings were developed, of which, the best ones were identified. AHP-based tools were used to capture experts’ reasoning of the competitive priority criteria by comparing the criteria. It was observed that fuzzy logic-based tools are able to better emulate industry experts’ reasoning of manufacturing reshoring. This research contributes to theory with a holistic framework of reshoring decision criteria, and to practice with decision-support tools for evaluation of manufacturing reshoring decisions.
Under de tre senaste decennierna har många företag flyttat sin produktion till lågkostnadsländer för att kunna utnyttja lägre lönekostnader. Många gånger har företagen genom denna åtgärd lyckats sänka sin tillverkningskostnad men samtidigt drabbats av oförutsedda problem kopplat till exempelvis produkt-kvalitet och möjligheten att kundanpassa produkter. Hanteringen av problemen har lett till ytterligare kostnader som många gånger överstigit besparingen i tillverkningskostnad. Detta har lett till att allt fler företag börjat flytta tillbaka sin produktion till hemlandet, så kallad reshoring. Reshoring är ett ungt område där det saknas forskning gällande bland annat hur den här typen av beslut på bästa sätt kan utvärderas och vilken typ av beslutstöd som kan underlätta den här typen av beslut. Därför är syftet med den här avhandlingen är att utveckla beslutsstödverktyg för utvärdering av reshoring beslut. För att uppfylla syftet har två forskningsfrågor formulerats. Den första frågan handlar om hur industriexperter resonerar kring reshoring beslut (RQ1) medan den andra frågan handlar om hur deras resonemang kan modelleras i beslutsstödverktyg (RQ2). Tre studier har genomförts för att besvara forskningsfrågorna, en fallstudie och två modelleringsstudier. Fallstudien fokuserar på att identifiera vilka kriterier som industriexperter beaktar medan modelleringsstudierna fokuserar på att utveckla beslutstödsverktyg där en del av dessa kriterier beaktas, med hjälp av fuzzy logic och analytical hierarchy process (AHP). Resultaten från forskningen visar att industriexperter bedömer reshoring beslut utifrån ett holistiskt perspektiv. En stor del av dessa beslutskriterier finns inom konkurrenskraft kategorin och inom dessa, har industriexperterna lagt högst vikt på kvalitet och lägst vikt på hållbarhet. Genom fuzzy logic modellering modellerades kriterierna på en övergripande nivå. Tre nya fuzzy logic koncept utvecklades för att fånga experternas resonemang. Dessutom utvecklades två konfigurationer med sexton olika inställningar, och de bästa identifierades. AHP-baserade verktyg utvecklades för att fånga experternas resonemang om kriterierna för konkurrenskraft prioriteringar. Fuzzy logic-baserade verktyg kan bättre fånga experternas resonemang kring reshoring beslut. Denna forskning bidrar till teori med en holistisk lista över beslutskriterier för reshoring beslut, och till praktik med beslutsstöd verktyg för utvärdering av reshoring beslut.

Thesis (M.S. in Operations Research)--Naval Postgraduate School, June 2010.
Thesis Advisor(s): Szechtman, Roberto ; Atkinson, Michael P. ; Second Reader: Kress, Moshe. "June 2010." Description based on title screen as viewed on July 14, 2010. Author(s) subject terms: Decision analysis, systems analysis, influence diagram, multi agent, MAID, Hezbollah, Hizbullah, Lebanese, Lebanon, Israel, Iran, Syria, Genie, Imad Mughniyah, nuclear, Iran, Middle East, temporal, dynamic programming, backward induction, political stakeholders, political actors, decision support, decision logic, decision forecast. Includes bibliographical references (p. 149-151). Also available in print.

This dissertation focuses on the problem of uncertainty handling during learning by agents dealing in stochastic environments by means of reinforcement learning. Most previous investigations in reinforcement learning have proposed algorithms to deal with the learning performance issues but neglecting the uncertainty present in stochastic environments. Reinforcement learning is a valuable learning method when a system requires a selection of actions whose consequences emerge over long periods for which input-output data are not available. In most combinations of fuzzy systems with reinforcement learning, the environment is considered deterministic. However, for many cases, the consequence of an action may be uncertain or stochastic in nature. This work proposes a novel reinforcement learning approach combined with the universal function approximation capability of fuzzy systems within a probabilistic fuzzy logic theory framework, where the information from the environment is not interpreted in a deterministic way as in classic approaches but rather, in a statistical way that considers a probability distribution of long term consequences. The generalized probabilistic fuzzy reinforcement learning (GPFRL) method, presented in this dissertation, is a modified version of the actor-critic learning architecture where the learning is enhanced by the introduction of a probability measure into the learning structure where an incremental gradient descent weight- updating algorithm provides convergence. XXIABSTRACT Experiments were performed on simulated and real environments based on a travel planning spoken dialogue system. Experimental results provided evidence to support the following claims: first, the GPFRL have shown a robust performance when used in control optimization tasks. Second, its learning speed outperforms most of other similar methods. Third, GPFRL agents are feasible and promising for the design of adaptive behaviour robotics systems.

When non-recurrent road traffic congestion happens, the operator of the traffic control centre has to select the most appropriate traffic control measure or combination of measures in a short time to manage the traffic network. This is a complex task, which requires expert knowledge, much experience and fast reaction. There are a large number of factors related to a traffic state as well as a large number of possible control measures that need to be considered during the decision making process. The identification of suitable control measures for a given non-recurrent traffic congestion can be tough even for experienced operators. Therefore, simulation models are used in many cases. However, simulating different traffic scenarios for a number of control measures in a complicated situation is very time-consuming. In this paper we propose an intelligent traffic control decision support system (ITC-DSS) to assist the human operator of the traffic control centre to manage online the current traffic state. The proposed system combines three soft-computing approaches, namely fuzzy logic, neural network, and genetic algorithm. These approaches form a fuzzy-neural network tool with self-organization algorithm for initializing the membership functions, a GA algorithm for identifying fuzzy rules, and the back-propagation neural network algorithm for fine tuning the system parameters. The proposed system has been tested for a case-study of a small section of the ring-road around Riyadh city. The results obtained for the case study are promising and show that the proposed approach can provide an effective support for online traffic control.

The design and verification of high performance circuits is becoming increasingly challenging due to the complex sets of constraints that must hold to ensure correct operation. The complexity of digital circuits increases rapidly, which in turn results in an increased complexity of the verification process and increased user effort. Efficient and easy to use tools are therefore required for timed verification.
In this thesis we propose two methods to aid the verification process. Firstly, we introduce a new verification methodology that combines the advantages of several existing successful approaches. In particular, our verification technique uses a combination of untimed, relative timing and timed verification. Secondly, we propose and evaluate a novel method of solving CSPs (constraint satisfaction problems) using BDDs (binary decision diagrams). We investigate two different implementations of a BDD-based CSP solver and their capacity to bridge the gap between untimed and timed verification. Finally, we present two case studies to demonstrate the proposed techniques.

A tomada de decisÃo à um processo cognitivo que conduz à seleÃÃo de um plano de escolha dentre vÃrios. Este pode ser concebido atravÃs do juÃzo de um ou vÃrios indivÃduos, os quais serÃo definidos como agentes. O trabalho em questÃo terà como cerne ambientes onde grupo de indivÃduos atuam simultaneamente influenciando uns aos outros, ou seja, iremos trabalhar com sistemas multiagentes. Dentre a classe de problemas envolvidos pela tomada de decisÃo, destacam-se os Problemas de DecisÃo MulticritÃrio. Estes sÃo uma variaÃÃo dos Problemas de DecisÃo usuais onde a correta tomada de decisÃo se processa atravÃs da apreciaÃÃo de vÃrios critÃrios, os quais sÃo utilizados para descrever o objeto/fato a ser decidido. Para que seja possÃvel a tomada de decisÃo, se faz necessÃrio uma estratÃgia que analise o problema em questÃo, de modo a determinar as alternativas sobre as quais o tomador de decisÃo deverà escolher, avalie cada critÃrio que compÃe a alternativa, diante dos possÃveis valores que estes podem assumir, para, assim, realizar a tomada de decisÃo. Dentre as diversas estratÃgias utilizadas para resolver este tipo de problemas, destacam-se aquelas que usam a LÃgica MatemÃtica como tÃcnica de modelagem e soluÃÃo. Amplamente estudada por pesquisadores vinculados à InteligÃncia Artificial, a LÃgica MatemÃtica utiliza-se de conceitos especÃficos de sua sintaxe e semÃntica para modelar ambientes complexos e estabelecer por mÃtodos especÃficos a tomada de decisÃo. Dentre as abordagens relevantes vinculadas à LÃgica MatemÃtica, destacam-s aquelas que empregam a LÃgica de Primeira Ordem. Inspirado nas LÃgica de PreferÃncias clÃssicas, este trabalho propÃe a LÃgica de Primeira Ordem para problemas de DecisÃo com AgregaÃÃo de PreferÃncias FODPA. Esta à capaz de modelar e resolver Problemas de DecisÃo MulticritÃrio em ambientes multiagentes atravÃs de tÃcnicas relacionadas Ãs LÃgicas com AgregaÃÃo de PreferÃncias e à LÃgica de Primeira Ordem.
Decision making is a cognitive procedure that lead to selection of a plan of choice among several. This can be designed through the judgment of one or more individuals, who are defined as agents. The work in question will have as core environments where several individuals act simultaneously, that is, we will work with multi-agent systems. Among the problems involved by the decision making processes, stand out the Multicriteria Decision Problems. These are a variation of the usual Decision Problems where the correct decision-making processes through the assessment of various criteria, which are used to describe the object / fact to be decided. To be able the correct decision making, a strategy is necessery to analyze the problem, determine the alternatives on which the decision maker must choose, evaluate each criterion that compose an alternative towards the possible values that each criterion can assume, thus, take the decision. Among the several strategies used to solve such problems, stand out those that use the Mathematical Logic as modeling and solution techniques. Extensively studied by researchers in Artificial Intelligence, the Mathematical Logic uses specific concepts of its syntax and semantics to model complex environments and establish decision-making. Inspired by the Classical Preferences Logics, this work aims to propose the First Order Logic for Decison problems with Preference Aggregation {FODPA}. This is able to model and solve Multicriteria Decisiom Problems in multi-agent environments by techniques related to Preferences Logics and the First Order Logic.

Multi-Criteria Group Decision Making (MCGDM) is a decision tool which is able to find a unique agreement from a group of decision makers (DMs) by evaluating various conflicting criteria. However, the current multi-criteria decision making with a group of DMs (MCGDM) techniques do not effectively deal with the large number of possibilities that cause disagreement between different judgements and the variety of ideas and opinions among the decision makers which lead to high_uncertainty levels. There is a growing interest to investigate techniques to handle the faced uncertainties in many decision making applications. Studies in fuzzy decision making have grown rapidly in the utilisation of extended fuzzy set theories (i.e., Type-2 Fuzzy Sets, Intuitionistic Fuzzy Sets, Hesitant Fuzzy Sets, Vague Sets, Interval-valued Fuzzy Sets; etc.) to evaluate the faced uncertainties.

This thesis, which is divided into five papers, explores biased decision making in naturalistic environments and its implications for the efficiency of financial markets and forecasts of competitive event outcomes. Betting markets offer a valuable real world decision making context, allowing analysis that is not possible using regular financial market data. The first paper surveys studies that have employed betting markets to investigate biased decision making and discusses why the extent of these biases is significantly less than in the laboratory. The second paper addresses unresolved issues relating to noise trading and herding in financial markets, by showing that noise trading is associated with increased market efficiency, that the extent of herding differs depending on the direction and timing of changes in market prices, and that this results in an economically significant inefficiency. The findings of this paper have important policy implications for wider financial markets: regulatory measures to protect investors from the destabilizing effects of noise appear to be self-defeating and herding is particularly prevalent when uninformed traders perceive that informed traders are participating in the market. The third and fourth papers address the favourite-longshot bias (FLB), where market prices under-/over-estimate high/low probability outcomes. These papers demonstrate that previous explanations of the bias are inconsistent with evidence of trading in UK betting markets by developing and testing the predictions of models that explain the bias in terms of competition between market makers and the demand preferences of bettors. Moreover, it is definitively shown that, when no market maker is involved, the bias is due to cognitive errors of traders rather than their preference for risk, because only prospect theory, and not risk-love, can explain a reduced FLB in events with strong favourites. The final chapter explores methodological concerns relating to estimates of forecast accuracy in models of discrete choice, and arrives at a much more rigorous understanding of the value of these estimates.

We describe contributions to algorithmic proof techniques for deciding the satisfiability of boolean combinations of many-variable nonlinear polynomial equations and inequalities over the real and complex numbers. In the first half, we present an abstract theory of Grobner basis construction algorithms for algebraically closed fields of characteristic zero and use it to introduce and prove the correctness of Grobner basis methods tailored to the needs of modern satisfiability modulo theories (SMT) solvers. In the process, we use the technique of proof orders to derive a generalisation of S-polynomial superfluousness in terms of transfinite induction along an ordinal parameterised by a monomial order. We use this generalisation to prove the abstract (“strategy-independent”) admissibility of a number of superfluous S-polynomial criteria important for efficient basis construction. Finally, we consider local notions of proof minimality for weak Nullstellensatz proofs and give ideal-theoretic methods for computing complex “unsatisfiable cores” which contribute to efficient SMT solving in the context of nonlinear complex arithmetic. In the second half, we consider the problem of effectively combining a heterogeneous collection of decision techniques for fragments of the existential theory of real closed fields. We propose and investigate a number of novel combined decision methods and implement them in our proof tool RAHD (Real Algebra in High Dimensions). We build a hierarchy of increasingly powerful combined decision methods, culminating in a generalisation of partial cylindrical algebraic decomposition (CAD) which we call Abstract Partial CAD. This generalisation incorporates the use of arbitrary sound but possibly incomplete proof procedures for the existential theory of real closed fields as first-class functional parameters for “short-circuiting” expensive computations during the lifting phase of CAD. Identifying these proof procedure parameters formally with RAHD proof strategies, we implement the method in RAHD for the case of full-dimensional cell decompositions and investigate its efficacy with respect to the Brown-McCallum projection operator. We end with some wishes for the future.

Intelligent systems that assist users in fulfilling complex tasks need a concise and processable representation of incomplete and uncertain information. In order to be able to choose among different options, these systems also need a compact and processable representation of the concept of preference. Preferences can provide an effective way to choose the best solutions to a given problem. These solutions can represent the most plausible states of the world when we model incomplete information, the most satisfactory states of the world when we express user preferences, or optimal decisions when we make decisions under uncertainty. Several domains, such as, reasoning under incomplete and uncertain information, user preference modeling, and qualitative decision making under uncertainty, have benefited from advances on preference representation. In the literature, several symbolic approaches of nonclassical reasoning have been proposed. Among them, logic programming under answer set semantics offers a good compromise between symbolic representation and computation of knowledge and several extensions for handling preferences. Nevertheless, there are still some open issues to be considered in logic programming. In nonmonotonic reasoning, first, most approaches assume that exceptions to logic program rules are already specified. However, sometimes, it is possible to consider implicit preferences based on the specificity of the rules to handle incomplete information. Secondly, the joint handling of exceptions and uncertainty has received little attention: when information is uncertain, the selection of default rules can be a matter of explicit preferences and uncertainty. In user preference modeling, although existing logic programming specifications allow to express user preferences which depend both on incomplete and contextual information, in some applications, some preferences in some context may be more important than others. Furthermore, more complex preference expressions need to be supported. In qualitative decision making under uncertainty, existing logic programming-based methodologies for making decisions seem to lack a satisfactory handling of preferences and uncertainty. The aim of this dissertation is twofold: 1) to tackle the role played by preferences in logic programming from different perspectives, and 2) to contribute to this novel field by proposing several frameworks and methods able to address the above issues. To this end, we will first show how preferences can be used to select default rules in logic programs in an implicit and explicit way. In particular, we propose (i) a method for selecting logic program rules based on specificity, and (ii) a framework for selecting uncertain default rules based on explicit preferences and the certainty of the rules. Then, we will see how user preferences can be modeled and processed in terms of a logic program (iii) in order to manage user profiles in a context-aware system and (iv) in order to propose a framework for the specification of nested (non-flat) preference expressions. Finally, in the attempt to bridge the gap between logic programming and qualitative decision under uncertainty, (v) we propose a classical- and a possibilistic-based logic programming methodology to compute an optimal decision when uncertainty and preferences are matters of degrees.
Els sistemes intel.ligents que assisteixen a usuaris en la realització de tasques complexes necessiten una representació concisa i formal de la informació que permeti un raonament nomonòton en condicions d’incertesa. Per a poder escollir entre les diferents opcions, aquests sistemes solen necessitar una representació del concepte de preferència. Les preferències poden proporcionar una manera efectiva de triar entre les millors solucions a un problema. Aquestes solucions poden representar els estats del món més plausibles quan es tracta de modelar informació incompleta, els estats del món més satisfactori quan expressem preferències de l’usuari, o decisions òptimes quan estem parlant de presa de decisió incorporant incertesa. L’ús de les preferències ha beneficiat diferents dominis, com, el raonament en presència d’informació incompleta i incerta, el modelat de preferències d’usuari, i la presa de decisió sota incertesa. En la literatura, s’hi troben diferents aproximacions al raonament no clàssic basades en una representació simbòlica de la informació. Entre elles, l’enfocament de programació lògica, utilitzant la semàntica de answer set, ofereix una bona aproximació entre representació i processament simbòlic del coneixement, i diferents extensions per gestionar les preferències. No obstant això, en programació lògica es poden identificar diferents problemes pel que fa a la gestió de les preferències. Per exemple, en la majoria d’enfocaments de raonament no-monòton s’assumeix que les excepcions a default rules d’un programa lògic ja estan expressades. Però de vegades es poden considerar preferències implícites basades en l’especificitat de les regles per gestionar la informació incompleta. A més, quan la informació és també incerta, la selecció de default rules pot dependre de preferències explícites i de la incertesa. En el modelatge de preferències del usuari, encara que els formalismes existents basats en programació lògica permetin expressar preferències que depenen d’informació contextual i incompleta, en algunes aplicacions, donat un context, algunes preferències poden ser més importants que unes altres. Per tant, resulta d’interès un llenguatge que permeti capturar preferències més complexes. En la presa de decisions sota incertesa, les metodologies basades en programació lògica creades fins ara no ofereixen una solució del tot satisfactòria pel que fa a la gestió de les preferències i la incertesa. L’objectiu d’aquesta tesi és doble: 1) estudiar el paper de les preferències en la programació lògica des de diferents perspectives, i 2) contribuir a aquesta jove àrea d’investigació proposant diferents marcs teòrics i mètodes per abordar els problemes anteriorment citats. Per a aquest propòsit veurem com les preferències es poden utilitzar de manera implícita i explícita per a la selecció de default rules proposant: (i) un mètode basat en l’especificitat de les regles, que permeti seleccionar regles en un programa lògic; (ii) un marc teòric per a la selecció de default rules incertes basat en preferències explícites i la incertesa de les regles. També veurem com les preferències de l’usuari poden ser modelades i processades usant un enfocament de programació lògica (iii) que suporti la creació d’un mecanisme de gestió dels perfils dels usuaris en un sistema amb reconeixement del context; (iv) que permeti proposar un marc teòric capaç d’expressar preferències amb fòrmules imbricades. Per últim, amb l’objectiu de disminuir la distància entre programació lògica i la presa de decisió amb incertesa proposem (v) una metodologia basada en programació lògica clàssica i en una extensió de la programació lògica que incorpora lògica possibilística per modelar un problema de presa de decisions i per inferir una decisió òptima.
Los sistemas inteligentes que asisten a usuarios en tareas complejas necesitan una representación concisa y procesable de la información que permita un razonamiento nomonótono e incierto. Para poder escoger entre las diferentes opciones, estos sistemas suelen necesitar una representación del concepto de preferencia. Las preferencias pueden proporcionar una manera efectiva para elegir entre las mejores soluciones a un problema. Dichas soluciones pueden representar los estados del mundo más plausibles cuando hablamos de representación de información incompleta, los estados del mundo más satisfactorios cuando hablamos de preferencias del usuario, o decisiones óptimas cuando estamos hablando de toma de decisión con incertidumbre. El uso de las preferencias ha beneficiado diferentes dominios, como, razonamiento en presencia de información incompleta e incierta, modelado de preferencias de usuario, y toma de decisión con incertidumbre. En la literatura, distintos enfoques simbólicos de razonamiento no clásico han sido creados. Entre ellos, la programación lógica con la semántica de answer set ofrece un buen acercamiento entre representación y procesamiento simbólico del conocimiento, y diferentes extensiones para manejar las preferencias. Sin embargo, en programación lógica se pueden identificar diferentes problemas con respecto al manejo de las preferencias. Por ejemplo, en la mayoría de enfoques de razonamiento no-monótono se asume que las excepciones a default rules de un programa lógico ya están expresadas. Pero, a veces se pueden considerar preferencias implícitas basadas en la especificidad de las reglas para manejar la información incompleta. Además, cuando la información es también incierta, la selección de default rules pueden depender de preferencias explícitas y de la incertidumbre. En el modelado de preferencias, aunque los formalismos existentes basados en programación lógica permitan expresar preferencias que dependen de información contextual e incompleta, in algunas aplicaciones, algunas preferencias en un contexto puede ser más importantes que otras. Por lo tanto, un lenguaje que permita capturar preferencias más complejas es deseable. En la toma de decisiones con incertidumbre, las metodologías basadas en programación lógica creadas hasta ahora no ofrecen una solución del todo satisfactoria al manejo de las preferencias y la incertidumbre. El objectivo de esta tesis es doble: 1) estudiar el rol de las preferencias en programación lógica desde diferentes perspectivas, y 2) contribuir a esta joven área de investigación proponiendo diferentes marcos teóricos y métodos para abordar los problemas anteriormente citados. Para este propósito veremos como las preferencias pueden ser usadas de manera implícita y explícita para la selección de default rules proponiendo: (i) un método para seleccionar reglas en un programa basado en la especificad de las reglas; (ii) un marco teórico para la selección de default rules basado en preferencias explícitas y incertidumbre. También veremos como las preferencias del usuario pueden ser modeladas y procesadas usando un enfoque de programación lógica (iii) para crear un mecanismo de manejo de los perfiles de los usuarios en un sistema con reconocimiento del contexto; (iv) para crear un marco teórico capaz de expresar preferencias con formulas anidadas. Por último, con el objetivo de disminuir la distancia entre programación lógica y la toma de decisión con incertidumbre proponemos (v) una metodología para modelar un problema de toma de decisiones y para inferir una decisión óptima usando un enfoque de programación lógica clásica y uno de programación lógica extendida con lógica posibilística.
Sistemi intelligenti, destinati a fornire supporto agli utenti in processi decisionali complessi, richiedono una rappresentazione dell’informazione concisa, formale e che permetta di ragionare in maniera non monotona e incerta. Per poter scegliere tra le diverse opzioni, tali sistemi hanno bisogno di disporre di una rappresentazione del concetto di preferenza altrettanto concisa e formale. Le preferenze offrono una maniera efficace per scegliere le miglior soluzioni di un problema. Tali soluzioni possono rappresentare gli stati del mondo più credibili quando si tratta di ragionamento non monotono, gli stati del mondo più soddisfacenti quando si tratta delle preferenze degli utenti, o le decisioni migliori quando prendiamo una decisione in condizioni di incertezza. Diversi domini come ad esempio il ragionamento non monotono e incerto, la strutturazione del profilo utente, e i modelli di decisione in condizioni d’incertezza hanno tratto beneficio dalla rappresentazione delle preferenze. Nella bibliografia disponibile si possono incontrare diversi approcci simbolici al ragionamento non classico. Tra questi, la programmazione logica con answer set semantics offre un buon compromesso tra rappresentazione simbolica e processamento dell’informazione, e diversi estensioni per la gestione delle preferenze sono state proposti in tal senso. Nonostante ció, nella programmazione logica esistono ancora delle problematiche aperte. Prima di tutto, nella maggior parte degli approcci al ragionamento non monotono, si suppone che nel programma le eccezioni alle regole siano già specificate. Tuttavia, a volte per trattare l’informazione incompleta è possibile prendere in considerazione preferenze implicite basate sulla specificità delle regole. In secondo luogo, la gestione congiunta di eccezioni e incertezza ha avuto scarsa attenzione: quando l’informazione è incerta, la scelta di default rule può essere una questione di preferenze esplicite e d’incertezza allo stesso tempo. Nella creazione di preferenze dell’utente, anche se le specifiche di programmazione logica esistenti permettono di esprimere preferenze che dipendono sia da un’informazione incompleta che da una contestuale, in alcune applicazioni talune preferenze possono essere più importanti di altre, o espressioni più complesse devono essere supportate. In un processo decisionale con incertezza, le metodologie basate sulla programmazione logica viste sinora, non offrono una gestione soddisfacente delle preferenze e dell’incertezza. Lo scopo di questa dissertazione è doppio: 1) chiarire il ruolo che le preferenze giocano nella programmazione logica da diverse prospettive e 2) contribuire proponendo in questo nuovo settore di ricerca, diversi framework e metodi in grado di affrontare le citate problematiche. Per prima cosa, dimostreremo come le preferenze possono essere usate per selezionare default rule in un programma in maniera implicita ed esplicita. In particolare proporremo: (i) un metodo per la selezione delle regole di un programma logico basato sulla specificità dell’informazione; (ii) un framework per la selezione di default rule basato sulle preferenze esplicite e sull’incertezza associata alle regole del programma. Poi, vedremo come le preferenze degli utenti possono essere modellate attraverso un programma logico, (iii) per creare il profilo dell’utente in un sistema context-aware, e (iv) per proporre un framework che supporti la definizione di preferenze complesse. Infine, per colmare le lacune in programmazione logica applicata a un processo di decisione con incertezza (v) proporremo una metodologia basata sulla programmazione logica classica e una metodologia basata su un’estensione della programmazione logica con logica possibilistica.

Data mining approaches have been increasingly used in recent years in order to find patterns and regularities in large databases. In this study, the C4.5 decision tree approach was used for mining of Gene Logic database, containing biological data. The decision tree approach was used in order to identify the most relevant genes and risk factors involved in breast cancer, in order to separate healthy patients from breast cancer patients in the data sets used. Four different tests were performed for this purpose. Cross validation was performed, for each of the four tests, in order to evaluate the capacity of the decision tree approaches in correctly classifying ‘new’ samples. In the first test, the expression of 108 breast related genes, shown in appendix A, for 75 patients were used as input to the C4.5 algorithm. This test resulted in a decision tree containing only four genes considered to be the most relevant in order to correctly classify patients. Cross validation indicates an average accuracy of 89% in classifying ‘new’ samples. In the second test, risk factor data was used as input. The cross validation result shows an average accuracy of 87% in classifying ‘new’ samples. In the third test, both gene expression data and risk factor data were put together as one input. The cross validation procedure for this approach again indicates an average accuracy of 87% in classifying ‘new’ samples. In the final test, the C4.5 algorithm was used in order to indicate possible signalling pathways involving the four genes identified by the decision tree based on only gene expression data. In some of cases, the C4.5 algorithm found trees suggesting pathways which are supported by the breast cancer literature. Since not all pathways involving the four putative breast cancer genes are known yet, the other suggested pathways should be further analyzed in order to increase their credibility.

In summary, this study demonstrates the application of decision tree approaches for the identification of genes and risk factors relevant for the classification of breast cancer patients

In the construction industry, the adoption of concurrent engineering principles requires the development of effective enabling IT tools. Such tools need to address specific areas of need in the implementation of concurrent engineering in construction. Collaborative decision-making is an important area in this regard. A review of existing works has shown that none of the existing approaches to collaborative decision-making adequately addresses the needs of distributed construction project teams. The review also reveals that fuzzy logic offers great potential for application to collaborative decision-making. This thesis describes a Web-based collaborative decision-making system for construction project teams using fuzzy logic. Fuzzy logic is applied to tackle uncertainties and imprecision during the decision-making process. The prototype system is designed as Web-based to cope with the difficulty in the case where project team members are geographically distributed and physical meetings are inconvenient/or expensive. The prototype was developed into a Web-based software using Java and allows a virtual meeting to be held within a construction project team via a client-server system. The prototype system also supports objectivity in group decision-making and the approach encapsulated in the prototype system can be used for generic decision-making scenarios. The system implementation revealed that collaborative decision-making within a virtual construction project team can be significantly enhanced by the use of a fuzzybased approach. A generic scenario and a construction scenario were used to evaluate the system and the evaluation confirmed that the system does proffer many benefits in facilitating collaborative decision-making in construction. It is concluded that the prototype decision-making system represents a unique and innovative approach to collaborative decision-making in construction project teams. It not only contributes to the implementation of concurrent engineering in construction, but also it represents a substantial advance over existing approaches.

The mass-migration of refugees in the fall 2015 posed an immense humanitarian and logistical challenge: exhausted from their week-long journeys, refugees arrived in Vienna in need of care, shelter, food, medical aid, and onward transport. The refugee crisis was managed by an emerging polycentric and inter-sectoral collective of organizations. In this paper, we investigate how, during such a situation, leaders of these organizations made decisions in concert with each other and hence sustained the collective's capacity to act collectively. We ask: what was the logic of decision-making that orchestrated collective action during the crisis? In answering this question, we make the following contribution: departing from March's logics of consequences and appropriateness as well as Weick's work on sensemaking during crisis, we introduce an alternative logic that informed decision-making: the logic of tact. With this concept we (a) offer a better understanding of how managers make decisions under the condition of bounded rationality and the simultaneous transgression of their institutional identity in situations of crisis; and we (b) show that in decision-making under duress cognition is neither ahead of action, nor is action ahead of cognition; rather, tact explicates the rapid switching between cognition and action, orchestrating decision-making through this interplay.

Automated proof system for logic of correlated knowledge is presented in the dissertation. The system consists of the sequent calculus GS-LCK and the proof search procedure GS-LCK-PROC. Sequent calculus is sound, complete and satisfy the properties of invertibility of rules, admissibility of weakening, contraction and cut. The procedure GS-LCK-PROC is terminating and allows to check if the sequent is provable. Also decidability of logic of correlated knowledge has been proved. Using the terminating procedure GS-LCK-PROC the validity of all formulas of logic of correlated knowledge can be checked.
Automatinė įrodymų sistema koreliatyvių žinių logikai yra pristatoma disertacijoje. Sistemą sudaro sekvencinis skaičiavimas GS-LCK ir įrodymo paieškos procedūra GS-LCK-PROC. Sekvencinis skaičiavimas yra pagrįstas, pilnas ir tenkina taisyklių apverčiamumo, silpninimo, prastinimo ir pjūvio leistinumo savybes. Procedūra GS-LCK-PROC yra baigtinė ir leidžia patikrinti, ar sekvencija yra išvedama. Taip pat buvo įrodytas koreliatyvių žinių logikos išsprendžiamumas. Naudojant baigtinę procedūra GS-LCK-PROC, visų koreliatyvių žinių logikos formulių tapatus teisingumas gali būti patikrintas.

To date, several disciplines have broached the systems view of service and the engineering of service systems. Operations research applied to services began with a rather simplistic, macro view of resource integration in the form of data envelopment analysis (DEA), introduced by Charnes, Cooper and Rhodes in 1978 (Banker et al., 1984; Charnes et al., 1994). Micro models of service systems have tended to study the systems’ IT components (Hsu, 2009; Qiu 2009). Engineering, which has always been associated with ‘assembling pieces that work in specific ways’ (Ottino, 2004) and ‘a process of precise composition to achieve a predictable purpose and function’ (Fromm, 2010: 2), has contributed to greater scalability and purposeful control in service systems. However, the agents of the system are usually people whose activities may not easily be controlled by predictable processes and yet are critical aspects of the value-creating system (Ng et al., 2011b). There is need for a new combinative paradigm, such as third-generation activity theory, in which two or more activity systems come into contact, to explore dialogue, exchanging perspectives of multiple actors, resulting in networks or groups of activity systems that are constantly interacting (Marken, 2006; Nardi, 1996, Oliveros et al., 2010). While various systems approaches, such as general systems theory (von Bertalanffy, 1962); open systems theory (Boulding, 1956; Katz and Kahn, 1978); and viable systems approach (Barile, 2008; Beer, 1972; Golinelli, 2010), will not be reviewed here (see Ng et al., 2011a for a systems approach to service science), they share common tenets: boundaries, interfaces, hierarchy, feedback and adaptation to which most systems writers would add emergence, input, output and transformation (Kast and Rosenzweig, 1972). These terms may be used as a basis for a research agenda for the consideration of a service system.

L'accroissement et la diversification des services offerts par les systèmes informatiques modernes rendent la tâche de sécuriser ces systèmes encore plus complexe. D'une part, l'évolution du nombre de services système accroît le nombre des vulnérabilités qui peuvent être exploitées par des attaquants afin d'atteindre certains objectifs d'intrusion. D'autre part, un système de sécurité moderne doit assurer un certain niveau de performance et de qualité de service tout en maintenant l'état de sécurité. Ainsi, les systèmes de sécurité modernes doivent tenir compte des exigences de l'utilisateur au cours du processus de sécurité. En outre, la réaction dans des contextes critiques contre une attaque après son exécution ne peut pas toujours remédier à ses effets néfastes. Dans certains cas, il est essentiel que le système de sécurité soit en avance de phase par rapport à l'attaquant et de prendre les mesures nécessaires pour l'empêcher d'atteindre son objectif d'intrusion. Nous soutenons dans cette thèse que le processus de sécurité doit suivre un raisonnement intelligent qui permet au système de prévoir les attaques qui peuvent se produire par corrélation à une alerte détectée et d'appliquer les meilleures contre-mesures possibles. Nous proposons une approche qui génère des scénarios potentiels d'attaque qui correspondent à une alerte détectée. Ensuite, nous nous concentrons sur le processus de génération d'un ensemble approprié de contre-mesures contre les scénarios d'attaque générés. Un ensemble généré des contre-mesures est considéré comme approprié dans l'approche proposée s'il présente un ensemble cohérent et il satisfait les exigences de l'administrateur de sécurité (par exemple, la disponibilité). Nous soutenons dans cette thèse que le processus de réaction peut être considéré comme un débat entre deux agents. D'un côté, l'attaquant choisit ses arguments comme étant un ensemble d'actions pour essayer d'atteindre un objectif d'intrusion, et de l'autre côté l'agent défendant la cible choisit ses arguments comme étant un ensemble de contre-mesures pour bloquer la progression de l'attaquant ou atténuer les effets de l'attaque. D'autre part, nous proposons une approche basée sur une méthode d'aide à la décision multicritère. Cette approche assiste l'administrateur de sécurité lors de la sélection des contre-mesures parmi l'ensemble approprié des contre-mesures générées à partir de la première approche. Le processus d'assistance est basé sur l'historique des décisions de l'administrateur de sécurité. Cette approche permet également de sélectionner automatiquement des contre-mesures appropriées lorsque l'administrateur de sécurité est dans l'incapacité de les sélectionner (par exemple, en dehors des heures de travail, par manque de connaissances sur l'attaque). Enfin, notre approche est implémentée et testée dans le cadre des systèmes automobiles
The growth and diversity of services offered by modern systems make the task of securing these systems a complex exercise. On the one hand, the evolution of the number of system services increases the risk of causing vulnerabilities. These vulnerabilities can be exploited by malicious users to reach some intrusion objectives. On the other hand, the most recent competitive systems are those that ensure a certain level of performance and quality of service while maintaining the safety state. Thus, modern security systems must consider the user requirements during the security process.In addition, reacting in critical contexts against an attack after its execution can not always mitigate the adverse effects of the attack. In these cases, security systems should be in a phase ahead of the attacker in order to take necessary measures to prevent him/her from reaching his/her intrusion objective. To address those problems, we argue in this thesis that the reaction process must follow a smart reasoning. This reasoning allows the system, according to a detected attack, to preview the related attacks that may occur and to apply the best possible countermeasures. On the one hand, we propose an approach that generates potential attack scenarios given a detected alert. Then, we focus on the generation process of an appropriate set of countermeasures against attack scenarios generated among all system responses defined for the system. A generated set of countermeasures is considered as appropriate in the proposed approach if it presents a coherent set (i.e., it does not contain conflictual countermeasures) and it satisfies security administrator requirements (e.g., performance, availability). We argue in this thesis that the reaction process can be seen as two agents arguing against each other. On one side the attacker chooses his arguments as a set of actions to try to reach an intrusion objective, and on the other side the agent defending the target chooses his arguments as a set of countermeasures to block the attacker's progress or mitigate the attack effects. On the other hand, we propose an approach based on a recommender system using Multi-Criteria Decision Making (MCDM) method. This approach assists security administrators while selecting countermeasures among the appropriate set of countermeasures generated from the first approach. The assistance process is based on the security administrator decisions historic. This approach permits also, to automatically select appropriate system responses in critical cases where the security administrator is unable to select them (e.g., outside working hours, lack of knowledge about the ongoing attack). Finally, our approaches are implemented and tested in the automotive system use case to ensure that our approaches implementation successfully responded to real-time constraints

The harmonization of climate-adaptive behaviour with pre-existing decision-making processes is central to the way climate change adaptation is described in the literature. Yet such behaviour is largely understudied, making it difficult to predict whether and how individuals can integrate (i.e., ‘mainstream’) climate change with risk management processes for weather and other ‘normal’ stressors. In this dissertation, I examine the decision-making processes of South Africa’s commercial grain farmers, as a uniquely informative case, through five complementary studies of two original datasets. I seek to better understand the relationship between risk perceptions and climate-adaptive behaviour in this group, who are known to be sensitive to weather risks and who are adopting climate-resilient farming practices (i.e., Conservation Agriculture (CA)), but who are nonetheless perceived by local experts to be insensitive to climate change risks. In doing so, I distinguish between weather-sensitive decision-making, in which farmers perceive and react to weather risks in conjunction with other ‘normal’ risks, and climate-sensitive decision-making, in which they also perceive and react to the anticipated effects of climate change. Using mental models interviews in the Western Cape province (N = 90), I first reconceptualise farmers’ risk-based decision-making processes, drawing on theories of risk perception and framing from cognitive psychology and behavioural economics to interpret the empirical evidence. Second, I explain the variation in farmers’ adoption of climate-resilient CA practices based on the different cognitive frames (expressed as linguistic frames) that they use to perceive, interpret and respond to weather risks. Third, I use these results to guide the quantitative analysis of a national survey (N = 441), with which I assess the utility of the CA concept in promoting, monitoring and evaluating sustainable and climate-resilient farming practices, as envisioned by the Climate-Smart Agriculture and Sustainable Intensification frameworks. Fourth, I use the interview data to evaluate whether and how farmers integrate climate change and weather risks in farm-level decision-making. Fifth, I build on these findings by using the survey data to quantitatively test whether farmers perceive weather and climate change as equivalent risks.
Science, Faculty of
Resources, Environment and Sustainability (IRES), Institute for
Graduate

This work deals with decision-making problems with uncertain information in dynamic environment. It develops a new logic system, multi-valued temporal propositional logic, combining multi-valued logic and linear temporal logic. This new logic allows uncertain information to be represented with either a numerical truth value in the [0,1] interval or a linguistic value, and uses these values with both states and same-time and next-time rules. Multi-valued temporal propositional logic, a generic logic system, provides a simple calculus for analysing uncertain information with Lukasiewicz implication algebra. It introduces uncertainty and temporality into rules. Soundness and completeness theorems provide a theoretical foundation for the reasoning system. Based on the new logic, forward and backward reasoning algorithms are proposed, which offers simulation/prediction and query answering functions. The reasoning system, based on the forward and backward reasoning algorithms was programmed in Pro log. Three scenarios were then used to evaluate this reasoning system. The results of the evaluations showed that the proposed reasoning system could return reliable and reasonable feedback to users on their input. Furthermore, users are able to perfect their knowledge bases according to the feedback from the reasoning system on their input. A general comparison between the reasoning system and dynamic Bayesian networks with a simple scenario with uncertain information in a certain temporal environment was undertaken. Within this comparison, an analysis of advantages and disadvantages between the proposed reasoning system and dynamic ~ Bayesian networks has been provided.

The aim of this thesis is partly to create a dynamic decision model for Sandvik Machining Solutions distribution flows and partly to evaluate how the stock transfer system logic handle four specific exceptional situations. The purpose is to reduce the total costs while keeping or improving the service level. The thesis presents a total cost model and guidelines for the planning function when deciding the main supplier in the distribution. The thesis also presents a system logic evaluation of the stock transfer logic used by Sandvik Machining Solutions.

This work is motivated by the need of providing patients with a decision support system that facilitates the selection of the most appropriate treatment strategy in cancer treatment. Treatment options are currently subject to predetermined clinical pathways and medical expertise, but generally, do not consider the individual patient characteristics or preferences. Although genomic patient data are available, this information is rarely used in the clinical setting for real-life patient care. In the area of personalized medicine, the advancement in the fundamental understanding of cancer biology and clinical oncology can promote the prevention, detection, and treatment of cancer diseases. The objectives of this research are twofold. 1) To develop a patient-centered decision support model that can determine the most appropriate cancer treatment strategy based on subjective medical decision criteria, and patient's characteristics concerning the treatment options available and desired clinical outcomes; and 2) to develop a methodology to organize and analyze gene expression data and validate its accuracy as a predictive model for patient's response to radiation therapy (tumor radiosensitivity). The complexity and dimensionality of the data generated from gene expression microarrays requires advanced computational approaches. The microarray gene expression data processing and prediction model is built in four steps: response variable transformation to emphasize the lower and upper extremes (related to Radiosensitive and Radioresistant cell lines); dimensionality reduction to select candidate gene expression probesets; model development using a Random Forest algorithm; and validation of the model in two clinical cohorts for colorectal and esophagus cancer patients. Subjective human decision-making plays a significant role in defining the treatment strategy. Thus, the decision model developed in this research uses language and mechanisms suitable for human interpretation and understanding through fuzzy sets and degree of membership. This treatment selection strategy is modeled using a fuzzy logic framework to account for the subjectivity associated to the medical strategy and the patient's characteristics and preferences. The decision model considers criteria associated to survival rate, adverse events and efficacy (measured by radiosensitivity) for treatment recommendation. Finally, a sensitive analysis evaluates the impact of introducing radiosensitivity in the decision-making process. The intellectual merit of this research stems from the fact that it advances the science of decision-making by integrating concepts from the fields of artificial intelligence, medicine, biology and biostatistics to develop a decision aid approach that considers conflictive objectives and has a high practical value. The model focuses on criteria relevant to cancer treatment selection but it can be modified and extended to other scenarios beyond the healthcare environment.

This is a collaboration project between the Nepal Heart Foundation (NHF) and the University of Greenwich (UoG), United Kingdom (UK). Our mutual aim, agreed at the outset, has been to analyse, design and develop a cost effective Clinical Decision Support System (CDSS) for diagnosis and recognition of Acute Rheumatic Fever (ARF) and Rheumatic Heart Disease (RHD) at an early stage by developing/adopting UK’s and NHF’s treatment practices and procedures that would be appropriate for the Nepalese environment and lifestyle. The Application we developed was designed for use by community health workers and doctors in the rural areas of Nepal where laboratory facilities, expert services and technology are often deficient. The research undertaken investigated three problems that previously had not been addressed in the Artificial Intelligence (AI) community. These are: 1) ARF in Nepal has created a lot of confusion in diagnosis and treatment, due to the lack of standard procedures; 2) the adoption of foreign guidelines is often not effective and does not suit the Nepali environment and lifestyle and 3) the value of combining (our proposed) Hybrid Approach (Knowledge-based System (KBS), Temporal Theory (TT) and Fuzzy Logic (FL)) to design and develop an application to diagnose ARF cases at an early stage in English and Nepali. This research presents, validates and evaluates a proposed Hybrid Approach to diagnose ARF at three different stages: 1) Detected; 2) Suspected and 3) Not-detected and also to identify the severity level of detected ARF in the forms of Severe Case, Moderate Case or Mild Case. The Hybrid Approach is a combination of the KBS/Boolean Rule Model, Temporal Model and Fuzzy Model. The KBS/Boolean Rule Model has four components for design and implementation of KBS. These are: identifying the ARF stage in a case; Rule Pattern Matching; New Rule Formation and Rule Selection Mechanism. The Temporal Model has four components namely: Descriptive Explanation of ARF symptoms; Temporal Lookup-Table/Rules and Temporal Reasoning which produce a Temporal Template for demonstrating the relationship between the signs and ARF. The Fuzzification, Fuzzy Inferences and Defuzzification components are applied to design and implement a Fuzzy Model. The research undertaken divided the overall ARF diagnosis problem, in effect its requirements, into several sub-problems and each model of the Hybrid Approach addressed particular sub-problems for example, Identify the stage of the ARF component of the KBS/Boolean Rule Model used to solve the question of identifying the stage of ARF based on the symptoms presented. Each problem was therefore handled using a particular model’s components. This significantly helped to improve maintainability, reliability and the overall quality of our final ARF Diagnosis Application. The developed ARF Diagnosis Application was experimentally tested and evaluated by NHF’s experts and users through applying NHF’s data sets consisting of 676 real patients’ records. The ARF Diagnosis Application was found to match 99% of the cases derived from NHF’s datasets. The overall ARF diagnostics performance and accuracy was 99.36%. Therefore, the experiments and evaluations of our ARF Diagnosis Application proved that it was logically and technically feasible to employ the Hybrid Approach for developing a new and practical ARF Diagnosis Application. The Application was ultimately developed and succeeded in embracing NHF’s requirements and guidelines thereby matching the Nepalese setting and making it suitable for use in Nepal having fully by met the exigencies of the Nepalese environment and lifestyle. Application of a new ARF diagnosis system (ours) proved that the Hybrid Approach, applied methods of diagnosis of ARF, medication and treatment plan, including help and supporting information which were identified and defined, were shown to be appropriate to support effectively community health workers and doctors who actively care for ARF and RHD cases in rural Nepal.

This research pursues the use of powerful BDD-based functional circuit analysis to evaluate some approaches to test set generation. Functional representations of the circuit allow the measurement of information about faults that is not directly available through circuit simulation methods, such as probability of random detection and test-space overlap between faults. I have created a software tool that performs experiments to make such measurements and augments existing test generation strategies with this new information. Using this tool, I explored the relationship of fault model difficulty to test set length through fortuitous detection, and I experimented with the application of function-based methods to help reconcile the traditionally opposed goals of making test sets that are both smaller and more effective.

Thesis (Ph. D.)--University of Missouri-Columbia, 2007.
The entire dissertation/thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file (which also appears in the research.pdf); a non-technical general description, or public abstract, appears in the public.pdf file. Title from title screen of research.pdf file (viewed on February 14, 2008) Vita. Includes bibliographical references.

Strategic decision-making is ubiquitous in everyday life. The analysis of game strategies has been a research theme in game theory for several decades since von Neumann and Morgenstern. Sophisticated models and analysis tools have been developed with wide applications in Economics, Management Science, Social Science and Computer Science, especially in the field of Artificial Intelligence. However, \much of game theory is about the question whether strategic equilibria exist", as Johan van Benthem, a world-leading logician and game-theorist, points out, \but there are hardly any explicit languages for defining, comparing, or combining strategies". Without such a facility it is challenging for computer scientists to build intelligent agents that are capable of strategic decision-making. In the last twenty years, logical approaches have been proposed to tackle this problem. Pioneering work includes Game Logics, Coalition Logic and Alternating-time Temporal Logic (ATL). These logics either provide facilities for expressing and combining games or offer mechanisms for reasoning about strategic abilities of players. But none of them can solve the problem. The intrinsic difficulty in establishing such a logic is that reasoning about strategies requires combinations of temporal reasoning, counterfactual reasoning, reasoning about actions, preferences and knowledge, as well as reasoning about multi-agent interactions and coalitional abilities. More recently, a few new logical formalisms have been proposed by extending ATL with strategy variables in order to express strategies explicitly. However, most of these logics tend to have high computational complexity, because ATL introduces quantifications over strategies (functions), which leaves little hope of building any tractable inference system based on such a logic. This thesis takes up the challenge by using a bottom-up approach in order to create a balance between expressive power and computational efficiency. Instead of starting with a highly complicated logic, we propose a set of logical frameworks based on a simple and practical logical language, called Game Description Language (GDL), which has been used as an official language for General Game Playing (GGP) since 2005. To represent game strategies, we extend GDL with two binary prioritized connectives for combining actions in terms of their priorities specified by these connectives, and provide it with a semantics based on the standard state transition model. To reason about the strategic abilities of players, we further extend the framework with coalition operators from ATL for specifying the strategic abilities of players. More importantly, a unified semantics is provided for both GDL- and ATL- formulas, which allows us to verify and reason about game strategies. Interestingly, the framework can be used to formalize the fundamental game-playing principles and formally derive two well-known results on two-player games: Weak Determinacy and Zermelo's Theorem. We also show that the model-checking problem of the logic is not worse than that of ATL*, an extension of ATL. To deal with imperfect information games, we extend GDL with the standard epistemic operators and provide it with a semantics based on the epistemic state transition model. The language allows us to specify an imperfect information game and formalize its epistemic properties. Meanwhile, the framework allows us to reason about players' own as well as other players' knowledge during game playing. Most importantly, the logic has a moderate computational complexity, which makes it significantly different from similar existing frameworks. To investigate the interplay between knowledge shared by a group of players and its coalitional abilities, we provide a variant of semantics for ATL with imperfect information. The relation between knowledge sharing and coalitional abilities is investigated through the interplay of epistemic and coalition modalities. Moreover, this semantics is able to preserve the desirable properties of coalitional abilities. To deal with collective decision-making, we apply the approach of combining actions via their priorities for collective choice. We extend propositional logic with the prioritized connective for modelling reason-based individual and collective choices. Not only individual preferences but also aggregation rules can be expressed within this logic. A model-checking algorithm for this logic is thus developed to automatically generate individual and collective choices. In many real-world situations, a group making collective judgments may assign individual members or subgroups different priorities to determine the collective judgment. We design an aggregation rule based on the priorities of individuals so as to investigate how the judgment from each individual affects group judgment in a hierarchical environment. We also show that this rule satisfies a set of plausible conditions and has a tractable computational complexity.

Over the last century, there has been growing interest in changing the approach to maintenance management. The current practice for selecting critical equipment and making a decision on the most appropriate maintenance strategy is perceived to have serious limitations, principally because it lacks decision analysis. Due to the complex nature of decision-making in maintenance management, different models have been developed for selecting critical equipment. However, many of these models considered maintenance management as operational concern and ignored the strategic concerns of maintenance management. This thesis builds upon earlier works on decision-making for selecting critical equipment and maintenance strategy. It sets out to construct three hypotheses by introducing evidence from a comprehensive literature review, case study analysis and in-depth interviews. The thesis focuses on artificial intelligence and multi-criteria decision-making techniques (i.e. Fuzzy Logic and Analytical Hierarchy Process) to bridge this gap. It proposes a strategic decision-making model in maintenance and asset management for selecting critical equipment and deciding on a maintenance strategy. The novelty of model is to propose an approach in which maintenance strategy can be applied based on the equipment criticality while not making a trade-off between safety and cost but rather to combine the concern of safety with financial, operational and technical perspectives. The model provides an opportunity to consider safety as the first priority. The research output suggests that existing criticality assessment methods for optimising maintenance delivery have limited value and are suffering from a lack of strategic decision analysis. Multi-criteria decision-making tools could be used to improve decision-making of criticality assessment methods and hence maintenance strategy implementation. The validity of the proposed strategic decision-making model was tested through case study analysis and in-depth interviews. The results suggest that a strategic decision-making model could have a significant impact on improving safety, reliability and operational availability. The strategic decision-making model would enable asset managers to track the consequences of their decisions whilst dealing with maintenance. It is also an effective tool in the hands of a maintenance department to convince their asset managers to make a maintenance investment.

The increasing risk of river flooding or coastal submersion is already visible through recent events like the storm Xynthia and the floods in the Var department, which caused several dozens of deaths in France. These catastrophic events, even if their extent remains relatively limited, would have justified a preventive evacuation of high risk prone areas. However, the consequences for the population would be much more serious when large cities of hundreds of thousands of people will be partially or totally threatened by floods. This possibility is already an actual danger for large megacities like Bangkok and Alexandria, and also threatens French cities like Tours, Paris or Nice. Being more and more aware of this possibility, big coastal, estuarine and river cities in France, in Europe and in all continents are incited to prepare emergency and mass evacuation plans in order to prevent and cope with exceptional events. The elaboration of these plans is extremely complex and difficult due to technical, organizational, sociological and even political aspects. The great majority of cities in the world prone to large scale disasters do not already have this kind of plan at their disposal. Moreover, the existing state of the art shows that there are few operational tools to help territorial managers implement these plans in the phases of preparation and crisis management. Our work aims to contribute to the development of a support method for the evacuation decision taken in a crisis management context. This method is partly based on the information included in the provisional evacuation plans produced in the preparation phase. To reach this objective, we propose to adapt the tools of the fuzzy logic approach and apply them to a set of synthesized indicators. These indicators or decision criteria have been first selected from a method of evacuation planning previously developed by the research team Avenues-GSU. These criteria integrate classic data on the hazard level (overall forecast level and local flood water levels), the vulnerability of the territory and population and, which is more innovative, some information about the ability of the organization to evacuate and the security or the risk of the evacuation itself. The final result of this method, applied to the spatial dimension with the Matlab and ArcGIS software, is a map of the necessity to evacuate. This map shows the areas with the highest priority to be evacuated according to a fuzzy multicriteria analysis. It has been tested 5 at the pilot site of the city of Bordeaux located upstream in the Gironde estuary, and the theoretical results were compared with historical floods of 1981 and 1999. A hypothetic flood scenario was also studied taking into account the potential climate change impact and the consequences of a 1 meter sea level rise during the 21st century. This method and prototype tool should help policymakers to better understand a complex situation in pre-alert phase and assess the real need for urban zones evacuation on the basis of a limited but representative set of criteria. The maps of the necessity to evacuate represents an innovative proposal which extend and complement the existing official maps of flood forecasting (vigicrue) and its implications in terms of local impacts and crisis management anticipation.

Automated deduction procedures for modal logics, and related decidable fragments of first-order logic, are used in many real-world applications. A popular way of obtaining decision procedures for these logics is to base them on semantic tableau calculi. We focus on calculi that use unification, instead of the more widely employed approach of generating ground instantiations over the course of a derivation. The most common type of tableaux with unification are so-called free-variable tableaux, where variables are treated as global to the entire tableau. A long-standing open problem for procedures based on free-variable tableaux is how to ensure fairness, in the sense that "equivalent" applications of the closure rule are prevented from being done over and over again. Some solutions such as using depth-first iterative deepening are known, but those are unnecessary in theory, and not very efficient in practice. This is a main reason why there are hardly any decision procedures for modal logics based on free-variable tableaux. In this thesis, we review existing work on incorporating unification into first-order and modal tableau procedures, show how the closure fairness problem arises, and discuss existing solutions to it. For the first-order case, we outline a calculus which addresses the closure fairness problem. As opposed to free-variable tableaux, closure fairness is much easier to achieve in disconnection tableaux and similar clausal calculi. We therefore focus on using clausal first-order tableau calculi for decidable classes, in particular the two-variable fragment. Using the so-called unrestricted blocking mechanism for enforcing termination, we present the first ground tableau decision procedure for this fragment. Even for such a ground calculus, guaranteeing that depth-first terminations terminate is highly non-trivial. We parametrise our procedure by a so-called lookahead amount, and prove that this parameter is crucial for determining whether depth-first derivations terminate or not. Extending these ideas to tableaux with unification, we specify a preliminary disconnection tableau procedure which uses a non-grounding version of the unrestricted blocking rule.