Thèses sur le sujet « Decisional models »

Miranda understanding, appreciation, and reasoning abilities are essential to courts' determinations of knowing and intelligent Miranda rights waivers. Despite the remarkable development of Miranda research in recent decades, studies have generally focused on understanding and appreciation of Miranda rights, but have not examined Miranda reasoning and waiver decisions. Therefore, examining the nature of defendants' decisional capacities constitutes a critical step in further developing theoretical and clinical models for competent Miranda waiver decisions. The current study evaluated Miranda waiver decisions for 80 pretrial defendants from two Tulsa-area Oklahoma jails. Previously untested, the current study examined systematically how rational decision abilities affect defendants' personal waiver decisions. Components from general models of legal decision making, such as decisional competence and judgment models, were examined to determine their applicability to Miranda waiver decisions.

The paper analyzes the enterprises decision-making models to approach International markets with special attention to operational practices developed by companies in the Milan hinterland territory. The analysis is preceded by a brief historical overview designed to outline the stages of the passage from the first decision-making models, which derived the expansion choices in international markets from the economic theory of organization, through decision-making related to the doctrine concerning the allocation of resources Hecksher and Ohlin, until the realization of the fact that these "traditional strategies limits" led to the emerging new "theory of monopolistic advantages and market imperfections", later superseded by the "theory of the product life cycle." It continues with the examination of the so-called Born Global and its success factors by focusing on innovative strategy to internationalize without following a predetermined path and outlined sequence, but with the networking created with other competitors, which allows the implementation of a shared control strategies with its network partners. The second part of the paper illustrates the results of the research performed in the territory and referred to the analysis of how the theories described in the previous chapters have been operationally deployed in the Italian reality. In particular the research focuses the attention on the analysis of internationalization models adopted by some companies in northern Italy. The paper shows the results of interviews conducted with representatives of Italian companies related to the way they defined the entry strategies, the selection of foreign markets and general management of the internationalization process

Thesis (M.S.)--School of Computing and Engineering. University of Missouri--Kansas City, 2005.
"A thesis in computer science." Typescript. Advisor: Deep Medhi. Vita. Title from "catalog record" of the print edition Description based on contents viewed May 30, 2006. Includes bibliographical references (leaves 91-93). Online version of the print edition.

The M3 theory contributes to new knowledge through original research and advanced scholarship by introducing a descriptive framework for strategic decision-making in uncertain and changing environments. Aided by the introduction of a Social Realism epistemology into management literature it is differentiated its ability to present complex strategic positions as essentialist (via modes), relative (via models), and dynamic (via momentum) to plot the dynamic trajectory of innovation emergence, change, adaptation and transformation over time. At a fundamental level, the M3 theory identifies a consistent set of rules that decision-makers intentionally or unintentionally engage with or ignore to take strategic positions based on four integrated yet polarized pairs of modes: systematic (+S) vs. responsive (+R) strategies, and conforming (+C), vs. differentiating (+D) strategies. Systematic strategies (+S) is the mode dedicated to increasingly sophisticated rational cognitive processes; these processes plan, purposefully compartmentalize, and regulate emotions. Responsive strategies (+R) conversely, is the mode dedicated to increasingly sensitized intuitive processes; these processes are reflective, associative, action-orientated and emotionally expressive. The second pair of modes intersects with the two aforementioned modes with conforming strategies (+C) moving towards convergence by adapting or conveying socially perceived superior norms; these processes include the exploitation of existing power. In contrast, differentiating strategies (+D) represents the mode dedicated to diverging from traditional norms with empowerment for exploration. These processes include novelty-seeking, sabotage, risk-taking, experimentation, play, flexibility, discovery, and higher levels of innovation. Finally, the dynamic (momentum) component informs how strategic modes and models under uncertainty improve and adjust in sophistication under the pressure and demands of the four drives (+L). The M3 theory is informed by three distinct but interrelated and simultaneous empirical streams of data: (i) field data from five ethnographic case studies, with research participant feedback loops; (ii) the mapping of 200+ peer reviewed decision-making models; and (iii) prototyping the principles in the construction of the emergent M3 theory.

This study is concerned with understanding decision-making in relation to out-of-authority educational placements for pupils with an Autistic Spectrum Condition (ASC) in one Local Authority (LA). The aims of this research were twofold. The first was to explore what factors have the greatest impact on the decision to educate pupils with an ASC outside the local authority. The second was to explore the perceptions of key informants about the process for deciding those placements. The study involved examining 24 pupil cases where out-of-authority ASC placements had been agreed and interviews with case-informants contributing to those placement decisions in order to analyse their beliefs and understandings about the processes of decision-making. The literature review highlights the limited research with regard to decision-making about pupils with Special Educational Needs (SEN) and draws examples from medical decision-making frameworks. Data analysis showed that two factors, complexity and range of pupil need and lack of LA provision to match the needs identified had the greatest influence on the decision to educate pupils outside the local authority. The response of the LA’s own schools, professionals and parents to those presenting needs as well as the consequent impact on the child/young person and others were recognised secondary factors. Case-informants offered a strong impression that for the majority of these pupils successful inclusion in their own LA would require increased and more integrated services in order to meet their identified needs. Data from the qualitative interviews provides a sense of the range of informants’ experiences relating to decision-making processes and the factors determining those perceptions. These related to whether the processes had been experienced as planned, were evidenced-based, child-focused and involved effective working with parents and other agencies. The findings, in part, reflect government concerns about the current statutory SEN framework and the case for change as made in the recent Green Paper (DfE, 2011). At a local level informants identified the need for a more explicit model of decision-making, ethically grounded with an emphasis for decision-making to be based on the holistic needs of the child and viewed that this would be better facilitated by having improved joint-working between services and stronger partnership engagement between the LA and parents/carers. Clinical professional-patient shared decision-making is discussed as a potential model which might be usefully applied to better understand and develop current SEN decision making.

New conceptual work in the judgment and decision-making research arena has suggested a nonconsequentialist perspective to decision-making. From this perspective, an emphasis is placed on emotions during the decision-making process, specifically positing that concurrent emotions may lead to decisions that are nonconsequentialist in nature. In the current study I develop the Nonconsequentialist Decision-Making Model (NDMM) and include indecisiveness as a vital construct in the model. In tune with much new research on emotions during the decision-making process, I examine how being indecisive is a product of negative concurrent emotions, and how indecisiveness affects the decision-making process. Using a natural decision-making setting, the current study had participants discuss the "biggest" decision they are currently facing in their lives. Data was collected regarding indecisiveness, nonconsequentialist dysfunctional decisional coping behavior, and decision difficulty. The findings show strong support for the NDMM and the nonconsequentialist perspective. Furthermore, the indecisiveness construct was measured successfully and showed to be a critical part of the decision-making process when dealing with difficult decisions.

This report describes a case study where a multi criteria decision model is used to make decisions regarding asset interventions for four social housing complexes, similar in terms of issues and possible interventions, at Christchurch City Council. The value judgements from the decision makers and their advisors that were necessary for creating the decision model were elicited through three workshops; selecting aspects, weighting and rating and lastly reviewing the output. An analysis performed shows that the decision model is logically consistent and does not suffer from the rank reversal phenomenon. The validation of the model also included creating four individual decision models, one for each social housing complex, comparing the results of applying the joint model and the individual models, and revisiting and reconsidering the value judgments made in the different models when discrepancies were found. This included utility difference analysis and asking trade-off questions to the decision makers. Part of the validation was also to get acceptance of the output of the joint model from the social housing team. Applying the decision model on the four social housing complexes and receiving an output which is accepted by the social housing team suggests that the aggregated model can be used for future decision problems of the same kind, provided that they are within the set level ranges of the aspects. Since the decision model is transparent in terms of which values or priorities have been applied and which prerequisites must be met in order to apply the model to future decisions, it is possible to use the decision model as a ‘live model’ with adjustment being made to it when required.

Thesis (MEng)--Stellenbosch University, 2013.
ENGLISH ABSTRACT: The management of physical assets has become a popular eld of study over recent years and is being acknowledged in multiple disciplines world wide. In this project, research on Physical Asset Management (PAM), maintenance and decision making are presented. PAM is a complex subject and requires the participation of multiple disciplines in order to successfully manage physical assets. Moreover, the management of maintenance makes a big contribution in achieving successful PAM. Decision making is a core element to manage maintenance e ciently, both on strategic and operational level. Various methods and techniques can be used to aid the decision making process such as, using past experience, xed decision making techniques and techniques involving numerical calculations, to mention only a few. However, using numerical calculations to make decisions are not very popular. This is due to various reasons, for example the inherent complexity of the mathematics and the time required to execute such calculations are disliked. People tend to avoid complex numerical calculations and rather rely on past experience and discussion of circulating opinions to make decisions. This is not ideal and can lead to inconsistent and inaccurate decisions. In this project, the importance of numerical decision making is researched, especially in maintenance related decisions. The focus is placed on the simpli cation of numerical decision making techniques with the aim to make it easy and quick to use to support operational PAM decisions. Di erent decisions regarding PAM, especially decisions with regards to managing maintenance in order to achieve PAM, are discussed by means of a literature study. This is done to clarify the applicability of using numerical decision making techniques to support this type of decisions. A few di erent available numerical techniques are highlighted that can be used to support the decision making process. The decisions together with numerical decision making techniques are evaluated in order to combine the most appropriate techniques in a simpli ed manner. The purpose of this is that it can be used by anyone with the necessary knowledge of a speci c system or operation. As a result a simpli ed numerical decision making toolbox is developed that can support maintenance related decision. This toolbox is applied to a real life situation by means of a case study, made possible by Anglo American Platinum Limited (Amplats). An evaluation and validation of the toolbox is done through the case study to conclude wether it has value in practice or not.
AFRIKAANSE OPSOMMING: Die bestuur van siese bates het die afgelope paar jaar 'n gewilde studieveld geword en word erken in verskeie dissiplines reg oor die w^ereld. In hierdie projek word navorsing gedoen oor Fisiese Bate Bestuur (FBB), instandhouding en besluitneming. FBB is 'n komplekse onderwerp en vereis die deelname van verskeie dissiplines om sukses te behaal. Die bestuur van instandhouding maak 'n groot bydrae tot suksesvolle FBB. 'n Kern element van doeltre ende instandhouding is besluitneming, beide op strategiese en operasionele vlak. Verskillende metodes en tegnieke kan gebruik word om die besluitnemingsproses te ondersteun soos byvoorbeeld om gebruik te maak van ondervinding en vorige gebeurtenisse, vaste besluitnemingstegnieke, tegnieke wat numeriese berekeninge gebruik en nog meer. Die gebruik van numeriese metodes om besluite te neem is nie baie gewild nie. Dit is as gevolg van verskeie redes soos byvoorbeeld die inherente kompleksiteit en ingewikkeldheid van die wiskunde en ook die tyd wat benodig word om sulke berekeninge uit te voer. Mense is geneig om ingewikkelde numeriese berekeninge te vermy en eerder staat te maak op vorige ervaring en die bespreking van menings om besluite te neem. Dit is nie ideaal nie en kan lei tot onkonsekwente besluite, of selfs verkeerde besluite. In hierdie projek is die belangrikheid van numeriese besluitneming nagevors, veral in die onderhoudsverwante besluite. Die fokus word geplaas op die vereenvoudiging van die numeriese besluitnemings tegnieke. Die doel is om dit op so 'n manier te vereenvoudig dat dit maklik en vinnig is om te gebruik vir operasionele FBB besluite. Verskillende besluite oor FBB, veral besluite met betrekking tot instandhouding om suksesvolle FBB te bereik, word bespreek deur middel van 'n literatuurstudie. Die literatuurstudie ondersoek die toepaslikheid van die gebruik van numeriese besluitnemingstegnieke vir hierdie soort besluite. 'n Paar verskillende beskikbare numeriese tegnieke wat gebruik kan word om die besluitnemingsproses te ondersteun word uitgelig. Die besluite, saam met numeriese besluitnemingtegnieke, word ge evalueer om die mees gepaste tegnieke te kombineer in 'n vereenvoudigde manier. Uiteindelik moet dit deur enige iemand met die nodige kennis van 'n spesi eke stelsel of proses gebruik kan word. As resultaat is 'n vereenvoudigde numeriese besluitnemingstegniekkombinasie ontwikkel wat besluite verwant aan instandhouding kan ondersteun. Hierdie tegniek-kombinasie word toegepas in 'n werklike situasie deur middel van 'n gevallestudie, wat moontlik gemaak is deur Anglo American Platinum Limited. 'n Evaluering en validering van die tegniek-kombinasie word gedoen in die gevallestudie om te bepaal of dit wel waarde het in die praktyk of nie.

Mestrado em Econometria Aplicada e Previsão
A relação entre a intervenção regulatória, as decisões de investimento dos operadores alternativos e o grau de concorrência nos mercados de comunicações eletrónicas tem sido intensamente discutida. O debate centra-se na possibilidade de obter um compromisso entre concorrência baseada em serviços e concorrência baseada em infraestruturas. A teoria da escada do investimento defende a conciliação destes dois objetivos pela intervenção adequada do regulador. Usando uma base de dados bastante completa e atendendo às fragilidades apontadas a outros estudos conclui-se que a informação sobre o mercado português comprova alguns pressupostos teóricos associados à teoria da escada do investimento: (i) a criação de condições para que os operadores alternativos entrem no mercado é um passo importante para que invistam em infraestrutura, e (ii) o regulador possui instrumentos para neutralizar o custo de oportunidade criado ao investimento em infraestruturas pelos lucros da concorrência baseada em serviços. O investimento em redes de fibra ótica pelos operadores alternativos é também considerado, avaliando os determinantes deste investimento e o respetivo efeito no nível de cobertura de uma área geográfica. É dada particular atenção à obtenção de uma especificação adequada para o modelo. Conclui-se que é preferível utilizar um modelo a duas partes em detrimento de um modelo a uma parte, pois os conjuntos de determinantes da decisão de investir numa área geográfica e da decisão relativa ao nível de cobertura a atingir nessa área não são idênticos. As características demográficas, económicas e sociais intrínsecas às áreas geográficas influenciam significativamente as decisões de investimento dos operadores alternativos.
The relation between regulation, the alternative operators' investment decisions and the degree of competition in the markets, has been an important policy issue over time. The discussions on this matter are mostly related with the possibility to achieve service-based competition in the short run, without compromising infrastructure-based competition in the long run. The investment ladder theory argues that both goals are achievable by appropriate regulatory intervention. By using a rich dataset and taking into account flaws pointed out in other studies, the present study finds reasonable evidence that the Portuguese market's data supports theoretical assumptions of the investment ladder theory: (i) creating conditions for alternative operators entering the market is an step in creating conditions for investment in infrastructure; (ii) the regulator has the tools to neutralise the opportunity cost for infrastructure investment created by service-based competition profits. The investment in fibre networks by alternative operators is also taken into consideration, with an evaluation of the investment determinants and their effect on coverage level of alternative operator's fibre networks. Particular attention is given to achieve an appropriate model specification. It is concluded that it is preferable to use a two-part model over a one part-model, which provides evidence that the determinants of the decision to invest in a geographical area are not entirely similar to the determinants of the decision on the coverage level in that area. The present study found that the intrinsic demographic, economic and social characteristics of a given geographical area influence investment decisions of alternative operators.

Mathematical models of water distribution systems (WDS) serve as tools to represent the real systems for many different purposes. Calibration is the process of fine tuning the model parameters so that the real system is well-represented. In practice, calibration is performed considering all information is deterministic. Recent researches have incorporated uncertainties caused by field measurements into the calibration process. Parameter (D-optimality) and predictive (I-optimality) uncertainties have been used as indicators of how well a system is calibrated.This study focuses on a methodology that extends previous work by considering the impact of uncertainty on decisions that are made using the model. A new sampling strategy that would take into account the accuracy needed for different model objectives is proposed.The methodology uses an optimization routine that minimizes square differences between the observed and model calculated head values by adjusting the model parameters. Given uncertainty in measurements, the parameters from this nonlinear regression are imprecise and the model parameter uncertainties are computed using a first order second moment (FOSM) analysis. Parameter uncertainties are then propagated to model prediction uncertainties through a second FOSM analysis. Finally, the prediction uncertainty relationships are embedded in optimization problems to assess the effect of the uncertainties on model-based decisions. Additional data is collected provided that the monetary benefits of reducing uncertainties can be addressed.The proposed procedure is first applied on a small hypothetical network for a system expansion design problem using a steady state model. It is hypothesized that the model accuracy and data required calibrating WDS models with different objectives would require different amount of data. A real-scale network for design and operation problems is studied using the same methodology for comparison. The effect of a common practice, grouping pipes in the system, is also examined in both studies.Results suggest that the cost reductions are related to the convergence of the mean parameter estimates and the reduction of parameter variances. The impact of each factor changes during the calibration process as the parameters become more precise and the design is modified. Identification of the cause of cost changes, however, is not always obvious.

Within the practice of public health at a global level there is now a greater interest in conceptualizing, materialize, and achieve the development of "healthy communities". Such communities would require the approach of new approaches in the planning to tackle its challenges based on timely access to information and collaborative planning of their local processes. However, until the moment little research focuses on methods based on these principles collaborative support to this type of response required in health planning of these communities. In addition in developing countries has been implementing the Model of Integral Health Care Family, Community and Intercultural Education (MAIS-FCI), whose design requires articulate an artifact agile to integrate its public officials, health teams and community in the identification, analysis and decision-making of comprehensive health actions of its territory. Thereby allowing the planning and direction of strategic actions where the community becomes the active subject of the construction and health care, raising sustainable solutions through a proper analysis of its geographical area. In this way, in this thesis is presented an investigation that was intended to propose and evaluate a conceptual and methodological model of collective planning for the identification of health services in the territory. The model, through a collective intelligence geospatial system sought to provide an ideal tool to optimize the process of decision-making in health planning (for example: the distribution of health services) and facilitate so the development of healthy communities located in varied and complex geographic settings. By adopting the methodology of Participatory Action Research (IAP by its acronym in Spanish) along with a helper method as is the documentary research, generated an artifact called Geospatial Health Planning Model of Collective Intelligence (MPSIC), with which it was able to promote a process of collective intelligence in the health communities participating in the evaluation of sites suitable for the distribution of health services. This model was evaluated in two different situations of location of sanitary campaigns in Ecuador: The first for post of supply of screening tests of HIV and the second for days of collection of blood. Annual extramural The main results obtained exhibited an acceptable level of relevance and adequacy of the proposed model in health planning, as well as an empowerment of the knowledge of the participants, improve the participation and consensus in the decision-making process for the allocation of health services. It also managed to determine geographic sites suitable to highly specialized problems in the field of health through a group of multidisciplinary experts without previous knowledge in geo-technological tools.
Dentro de la práctica de salud pública a nivel global existe en la actualidad un interés mayor por conceptualizar, materializar y alcanzar el desarrollo de "comunidades saludables". Tales comunidades, requerirían el planteamiento de nuevos enfoques en la planificación que permitan afrontar sus desafíos basándose en el acceso oportuno a la información y la planificación colaborativa de sus procesos locales. Sin embargo, hasta el momento existe poca investigación que se enfoque en métodos basados en estos principios colaborativos que apoyen a este tipo de respuesta requerida en la planificación sanitaria de dichas comunidades. Además en los países en vías de desarrollo se viene implantando el Modelo de Atención Integral en Salud Familiar, Comunitario e Intercultural (MAIS-FCI), cuya concepción requiere articular un artefacto ágil que permita integrar a sus funcionarios públicos, equipos de salud y comunidad, en la identificación, análisis y toma de decisiones de acciones integrales de salud de su territorio, permitiendo de esta forma la planeación y dirección de acciones estratégicas donde la comunidad se convierta en sujeto activo de la construcción y cuidado de la salud, planteando soluciones sustentables mediante un adecuado análisis de su espacio geográfico. De esta manera, en la presente tesis se presenta una investigación que tuvo como propósito proponer y evaluar un modelo conceptual y metodológico de planificación colectiva para la identificación de servicios sanitarios en el territorio. El modelo, a través de un sistema geoespacial de inteligencia colectiva buscó aportar una herramienta idónea para optimizar el proceso de toma de decisiones en la planificación sanitaria (por ejemplo: la distribución de servicios sanitarios) y facilitar así el desarrollo de comunidades saludables localizadas en variados y complejos escenarios geográficos. Adoptando la metodología de Investigación-Acción Participativa (IAP por su acrónimo en español) junto con un método auxiliar como lo es la Investigación Documental, se generó un artefacto denominado Modelo de Planificación Sanitaria Geoespacial de Inteligencia Colectiva (MPGSIC), con el cual se logró fomentar un proceso de inteligencia colectiva en las comunidades de salud participantes en la evaluación de sitios idóneos para la distribución de servicios sanitarios. Este modelo se evaluó en dos situaciones diferentes de ubicación de campañas sanitarias en el Ecuador: La primera para puesto de suministro de pruebas de tamizaje de VIH y la segunda para jornadas de colecta extramural anual de sangre. Los principales resultados obtenidos exhiben un nivel aceptable de pertinencia y adecuación del modelo propuesto en la planificación sanitaria, así como un empoderamiento del conocimiento de los participantes, mejorando la participación y el consenso en el proceso de toma de decisiones para la asignación de servicios sanitarios. Además se consiguió determinar sitios geográficos idóneos a problemas altamente especializados en el campo de salud a través de un grupo de expertos multidisciplinarios sin previos conocimientos en herramientas geotecnológicas.

Esta tese reformula o modelo de teoria de decisão de Savage relaxando a hipótese implícita de que uma conseqüência é uma descrição perfeita de uma determinada situação. Axiomas comportamentais sobre preferências definidas no espaço de atos são introduzidos e uma representação na forma de Utilidade Esperada é derivada. Em particular, como em Savage, há uma única probabilidade subjetiva sobre os estados da natureza. O ganho de flexibilidade da reformulação apresenta uma solução para o paradoxo de Ellsberg que não faz uso de múltiplas probabilidades subjetivas, e uma reinterpretação da aversão ao risco no modelo de Utilidade Esperada convencional.
This thesis reformulates the Savage\'s Decision Theory model relaxing the implicit hypothesis that a consequence must be a perfect description of a situation. We introduce Behavioral axioms on preferences defined over the set of acts and derive a new Expected Utility functional representation. Like on Savage\'s, there is an unique prior over states of world. The flexibility gain of this representation presents a solution for the Ellsberg paradox that does not rely on multiple priors, and allows for a new interpretation of risk aversion on the Expect Utility model.

The revolution of ICT technologies in Digital Era combined with progress of ComplexitySciences have lead to the emergence of Computational Social Science, whose purposeis to analyze society in a quantitative way. Physics, can provide some tools in suchendeavor. Here, a new experimental framework called Pop-Up Experiments, based in Citizen Science philosophy and scientic rigor of Behavioral Experiments, is proposed.Experiments and mathematical models, in agreement with them, are used to validate and explain stylized facts of social phenomena concerning three diferent areas regarding human decision-making: Social Dilemmas, Pedestrian Movement and Financial Markets. In first place, two experiments are carried out on a variety of Social Dilemmas. Results demonstrate that children present a more volatile and random behavior than adults and the fact that individuals can be assigned to a small number of behavioral phenotypes. Secondly, pedestrian movement of visitors in a out-door fair have also been empiricallystudied. A model based in Langevin Dynamics that includes orientation persistence andpotential landscape is able to reproduce observed movement patterns. Moreover, suchtested model is theoretically adapted for describing other related issues like searching patterns. Finally, an experiment on human behavior with nancial data, where participants try to anticipate real price series, reveals that under high uncertain environment humans develop intuitive strategies that ease them to make decisions. One of such found strategies consist in imitating what the price did before. The existence of such strategyin real financial Big Data recordings is tested and afterwards confirmed by using Mutual Information and Transfer of Entropy techniques. Those mathematical methods also constitute a powerful tool to analyze the behavior of a population of investors in terms of synchronization and anticipation. In summary, all these experiences demonstrate thatstochastic models and mathematical tools belonging to Physics combined with Pop-UpExperiments constitute a very powerful tool at the service of Computational Social Science for the study of human decision-making. The impact of such fact based approachmodels is can be important in promoting cooperative environments in educational process, redesigning spaces to allow a better mobility in areas like fairs or understanding how people manage their investments to establish better financial market policies.
Las tecnologías de la revolucionarias de la Era Digital combinadas con el progreso de las Ciencias de la Complejidad han llevado a la creación de la Ciencia Social Computacional, cuyo proposito consiste en analizar la sociedad de forma cuantitativa. La Física puede aportar ciertas herramientas para dicho empeño. En esta tesis se pro-pone un nuevo marco experimental llamado \Pop-Up Experiments" basado tanto en la losofía de la Ciencia Ciudadana como en los experimentos de comportamiento humano. Estos experimentos son utilizados para valorar modelos que tratan de explicar regularidades y patrones en fenomenos sociales de tres areas distintas: Los dilemas sociales, el movimiento de las personas y el comportamiento en mercados financieros. En el primer caso, se llevan a cabo experimentos sobre una variedad de dilemas sociales demostrandoque un pequeño grupo de fenotipos es suficiente para clasificar el comportamiento de los humanos. Ademas, en el caso del Dilema del Prisionero con varias rondas, los resultados muestran que los niños presentan un comportamiento mas aleatorio y volatil. En la segunda area se analiza experimentalmente el movimiento de los visitantes de un recinto ferial. Un modelo basado en la Dinamica de Langevin que incluye persistencia en la orientacion y un campo de potencial permite reproducir los patrones observados. La versatilidad del modelo tambien permite su adaptacion para describir patrones de busqueda. Finalmente, un experimento sobre comportamiento humano en mercados financieros, donde los participantes tratan de predecir series reales de precios, revela que los humanos desarrollan estrategias intuitivas que les ayudan a tomar decisiones. Una de dichas estrategias, que consiste en imitar el mercado, es posteriormente con rmadautilizando analisis de Informacion Mutua y Transferencia de Entropia en registros de un fondo de inversion. En conclusion, estas experiencias demuestran que los modelosestoc asticos y herramientas de la Fisica, combinadas con los experimentos \Pop-Up" con-stituye una poderosa herramienta al servicio de las Ciencias Sociales Computacionalespara el estudio del proceso de toma de decisi on humano. El impacto de estos modelosbasados en hechos comprende desde la promoci on de entornos cooperativos en el procesoeducativo, hasta el redise~no de espacios para una mejor movilidad, pasando por un mejorentendimiento de c omo las personas manejan sus inversiones para una mejor regulaci onen mercados financieros.

Ingeniera Civil Industrial
El presente estudio tiene por objetivo desarrollar e implementar una metodología de priorización de infraestructura hidráulica de riego, de forma que los proyectos de embalses sean estudiados y evaluados en su contexto. Para la elaboración de la herramienta de priorización, se utilizó el método MCDM, el cual corresponde a un enfoque que permite comparar alternativas en base a múltiples atributos y características. Entre los métodos estudiados, se optó por la metodología AHP para determinar los ponderadores, ya que permite incluir, de forma simultánea, características cuantitativas y cualitativas dentro de un mismo modelo. Se desarrolló un modelo con 5 niveles de criterios y subcriterios. En el primer nivel se especifica el objetivo del estudio. En el segundo nivel jerárquico, se consideran los criterios macro, los cuales corresponden a factores de desarrollo y estratégicos. Los criterios de desarrollo están asociados directamente con las características de los proyectos en sí y los criterios estratégicos corresponden a aspectos políticos y administrativos. En el tercer nivel se identifican los factores económico, social y ambiental. El subcriterio económico busca medir las rentabilidades y riesgos del modelo, el subcriterio social incluye los impactos que este tipo de obras tienen sobre las comunidades y el subcriterio ambiental buscar medir el impacto que tienen los embalses en el territorio donde estos se localizan. Este modelo fue aplicado en cuatro proyectos de embalses, a través del cual se obtuvo un Índice Multi-Criterio (IMC), donde a mayor IMC, mayor prioridad tiene el proyecto. Los resultados obtenidos son: en primer lugar el Embalse A, seguido del Embalse C, luego se encuentra el Embalse B y, en último lugar se encuentra el Embalse D. Los resultados muestran que los métodos multicriterio de priorización permiten incluir características cualitativas y cuantitativas de forma simultánea en un modelo. Sin embargo, es necesario perfeccionar este tipo de metodologías en aspectos tales como la selección de criterios, subcriterios e indicadores, de forma que los proyectos de embalses no sean sub o sobre valorados. Adicional a esto, para la correcta aplicación de la metodología multicriterio, es necesario disponer de información actualizada tanto de los proyectos, como de las comunidades y territorio donde estos se localizan.

El presente trabajo se estructura en dos bloques claramente diferenciados: el primero presenta un panorama teórico de la situación actual de la disciplina de Marketing, especialmente centrado en la imagen de Marca. El segundo, de tipo estrictamente cuantitativo, plantea el trabajo de investigación realizado y presenta un Método de análisis y predicción del posicionamiento y reposicionamiento de marca, que permite, al profesional y a la empresa, decidir las actuaciones en función de unos resultados numéricos dados. De esta forma, el trabajo se propone establecer un modelo para la toma de decisiones que afectan a la imagen y el posicionamiento de las marcas, es decir, que ayuden a elegir las más adecuadas sobre los posibles posicionamientos o reposicionamientos para las marcas. Se intenta predecir los resultados de un movimiento en imagen y posicionamiento de las marcas en cualquier mercado y para cualquier marca. Además, el modelo que se propone pretende poner de manifiesto que antes de simular los movimientos en posicionamiento, es necesario conocer la imagen y el posicionamiento actual de la marca y las marcas que concurren en el mismo mercado y determinar qué características son clave para establecer la estrategia de posicionamiento. Y es a partir de ahí, que el modelo calcula la relevancia de los beneficios (características) que el consumidor busca en una marca y determina en qué medida cada marca ofrece con garantías estos beneficios buscados. Finalmente, muestra el camino para determinar si es posible ocupar una posición determinada en la mente del consumidor y, en el caso de no serlo, que otros caminos son posibles para posicionar la marca. El Modelo de Simulación Perceptual Integrado (MSPI) pretende ser el primer modelo de simulación de estrategias de posicionamiento de marcas, ya que hasta el momento sólo existen modelos parciales que no permiten simular posicionamientos y/o reposicionamientos de marcas, ni siquiera en el plano teórico. El modelo que se desarrolla puede aplicarse en cualquier mercado, cualquier producto o categoría de productos, marcas e incluso tipo de datos, siempre y cuando estos midan la imagen o el posicionamiento de las marcas. El análisis de toda esta información que se posee sobre el posicionamiento e imagen de las marcas que se pretende simular, forma parte de la fase inicial del modelo y se denomina partida. Las siguientes fases (de la 2ª a la 5ª) sirven para crear los beneficios únicos (2ª fase) y determinar: su composición (3ª fase), su importancia (4ª fase) y también la importancia de los niveles que componen cada beneficio (5ª fase). Las últimas tres etapas proyectan el posicionamiento pretendido con el fin de observar si la propuesta será exitosa o no para la marca. Son la sexta etapa de decisión del futuro posicionamiento en base a los beneficios hallados en fases anteriores, la séptima para estimar los valores de los beneficios y sus niveles y la octava, y última, para proyectar el posicionamiento con los valores estimados en la fase anterior. El trabajo no sólo explica el desarrollo teórico del modelo sino que además expone, paso a paso, su funcionamiento a partir de un estudio cuantitativo realizado en el mercado de la telefonía móvil en España a 1.222 usuarios actuales y potenciales de 18 a 55 años. A lo largo del trabajo se ha intentado que el Marco Teórico fundamente y sustente el Proyecto de Investigación, de modo y manera que los resultados de la exposición del método desarrollado, sean justificados por las necesidades planteadas en el posicionamiento o reposicionamiento de la imagen de marca, y se adecuen a las perspectivas deseadas.

Threat assessment and reliable motion-prediction of surrounding vehicles are some of the major challenges encountered in autonomous vehicles' safe decision-making. Predicting a threat in advance can give an autonomous vehicle enough time to avoid crashes or near crash situations. Most vehicles on roads are human-driven, making it challenging to predict their intentions and movements due to inherent uncertainty in their behaviors. Moreover, different driver behaviors pose different kinds of threats. Various driver behavior predictive models have been proposed in the literature for motion prediction. However, these models cannot be trusted entirely due to the human drivers' highly uncertain nature. This thesis proposes a novel trust-based driver behavior prediction and stochastic reachable set threat assessment methodology for various dangerous situations on the road. This trust-based methodology allows autonomous vehicles to quantify the degree of trust in their predictions to generate the probabilistically safest trajectory. This approach can be instrumental in the near-crash scenarios where no collision-free trajectory exists. Three different driving behaviors are considered: Normal, Aggressive, and Drowsy. Hidden Markov Models are used for driver behavior prediction. A "trust" in the detected driver is established by combining four driving features: Longitudinal acceleration, lateral acceleration, lane deviation, and velocity. A stochastic reachable set-based approach is used to model these three different driving behaviors. Two measures of threat are proposed: Current Threat and Short Term Prediction Threat which quantify present and the future probability of a crash. The proposed threat assessment methodology resulted in a lower rate of false positives and negatives. This probabilistic threat assessment methodology is used to address the second challenge in autonomous vehicle safety: crash avoidance decision-making. This thesis presents a fast, proactive decision-making methodology based on Stochastic Model Predictive Control (SMPC). A proactive decision-making approach exploits the surrounding human-driven vehicles' intent to assess the future threat, which helps generate a safe trajectory in advance, unlike reactive decision-making approaches that do not account for the surrounding vehicles' future intent. The crash avoidance problem is formulated as a chance-constrained optimization problem to account for uncertainty in the surrounding vehicle's motion. These chance-constraints always ensure a minimum probabilistic safety of the autonomous vehicle by keeping the probability of crash below a predefined risk parameter. This thesis proposes a tractable and deterministic reformulation of these chance-constraints using convex hull formulation for a fast real-time implementation. The controller's performance is studied for different risk parameters used in the chance-constraint formulation. Simulation results show that the proposed control methodology can avoid crashes in most hazardous situations on the road.
Master of Science
Unexpected road situations frequently arise on the roads which leads to crashes. In an NHTSA study, it was reported that around 94% of car crashes could be attributed to driver errors and misjudgments. This could be attributed to drinking and driving, fatigue, or reckless driving on the roads. Full self-driving cars can significantly reduce the frequency of such accidents. Testing of self-driving cars has recently begun on certain roads, and it is estimated that one in ten cars will be self-driving by the year 2030. This means that these self-driving cars will need to operate in human-driven environments and interact with human-driven vehicles. Therefore, it is crucial for autonomous vehicles to understand the way humans drive on the road to avoid collisions and interact safely with human-driven vehicles on the road. Detecting a threat in advance and generating a safe trajectory for crash avoidance are some of the major challenges faced by autonomous vehicles. We have proposed a reliable decision-making algorithm for crash avoidance in autonomous vehicles. Our framework addresses two core challenges encountered in crash avoidance decision-making in autonomous vehicles: 1. The outside challenge: Reliable motion prediction of surrounding vehicles to continuously assess the threat to the autonomous vehicle. 2. The inside challenge: Generating a safe trajectory for the autonomous vehicle in case of future predicted threat. The outside challenge is to predict the motion of surrounding vehicles. This requires building a reliable model through which future evolution of their position states can be predicted. Building these models is not trivial, as the surrounding vehicles' motion depends on human driver intentions and behaviors, which are highly uncertain. Various driver behavior predictive models have been proposed in the literature. However, most do not quantify trust in their predictions. We have proposed a trust-based driver behavior prediction method which combines all sensor measurements to output the probability (trust value) of a certain driver being "drowsy", "aggressive", or "normal". This method allows the autonomous vehicle to choose how much to trust a particular prediction. Once a picture is painted of surrounding vehicles, we can generate safe trajectories in advance – the inside challenge. Most existing approaches use stochastic optimal control methods, which are computationally expensive and impractical for fast real-time decision-making in crash scenarios. We have proposed a fast, proactive decision-making algorithm to generate crash avoidance trajectories based on Stochastic Model Predictive Control (SMPC). We reformulate the SMPC probabilistic constraints as deterministic constraints using convex hull formulation, allowing for faster real-time implementation. This deterministic SMPC implementation ensures in real-time that the vehicle maintains a minimum probabilistic safety.

Desenvolveu-se uma metodologia para integração de dois modelos matemáticos, um de quantidade de água, em rede de fluxo, denominado ACQUANET com outro de qualidade de água, de uma dimensão, aplicado a represas, denominado CE-QUAL-R1. Para tanto, foi elaborada uma INTERFACE em linguagem de programação possibilitando que as vazões resultantes, simuladas pelo ACQUANET, servissem como dados de entrada ao CE-QUAL-R1 para simular a distribuição vertical das variáveis de qualidade de água em uma represa. Essa metodologia foi aplicada à Represa Jaguari-Jacareí no Sistema Cantareira em São Paulo, Brasil, como alternativa de gerenciamento quali-quantitativo, além de possibilitar o uso de retirada de água em diferentes profundidades, através da operação de tomadas d’água seletivas existentes.
A methodology was developed for the integration of two mathematical models, one of water quantity in network named ACQUANET with other of water quality, in one dimension, applied in revervoirs, named CE-QUAL-R1. In order to achieve this goal, an INTERFACE was developed to link the CE-QUAL-R1 with ACQUANET outflow results. It should be highlighted that ACQUANET has been used for beginning values of CE-QUAL-R1 and to simulate the vertical distribution of water quality variables in a reservoir. This methodology was applied to Jaguari-Jacarei Reservoir, of Cantareira System in Sao Paulo, Brazil, as a management quality and quantity tool of the system and it showed the use possibility of withdrawal of outflowing waters from different depths, through existing selective withdrawals ports operation.

The research objective of this thesis is to formulate and demonstrate a computational framework for modeling the design decisions of engineers. This framework is intended to be descriptive in nature as opposed to prescriptive or normative; the output of the model represents a plausible result of a designer's decision making process. The framework decomposes the decision into three elements: the problem statement, the designer's beliefs about the alternatives, and the designer's preferences. Multi-attribute utility theory is used to capture designer preferences for multiple objectives under uncertainty. Machine-learning techniques are used to store the designer's knowledge and to make Bayesian inferences regarding the attributes of alternatives. These models are integrated into the framework of a Markov decision process to simulate multiple sequential decisions. The overall framework enables the designer's decision problem to be transformed into an optimization problem statement; the simulated designer selects the alternative with the maximum expected utility. Although utility theory is typically viewed as a normative decision framework, the perspective in this research is that the approach can be used in a descriptive context for modeling rational and non-time critical decisions by engineering designers. This approach is intended to enable the formalisms of utility theory to be used to design human subjects experiments involving engineers in design organizations based on pairwise lotteries and other methods for preference elicitation. The results of these experiments would substantiate the selection of parameters in the model to enable it to be used to diagnose potential problems in engineering design projects. The purpose of the decision-making framework is to enable the development of a design process simulation of an organization involved in the development of a large-scale complex engineered system such as an aircraft or spacecraft. The decision model will allow researchers to determine the broader effects of individual engineering decisions on the aggregate dynamics of the design process and the resulting performance of the designed artifact itself. To illustrate the model's applicability in this context, the framework is demonstrated on three example problems: a one-dimensional decision problem, a multidimensional turbojet design problem, and a variable fidelity analysis problem. Individual utility functions are developed for designers in a requirements-driven design problem and then combined into a multi-attribute utility function. Gaussian process models are used to represent the designer's beliefs about the alternatives, and a custom covariance function is formulated to more accurately represent a designer's uncertainty in beliefs about the design attributes.

The three chapters included in this dissertation aim to improve our understanding of the distance between people's desires regarding the timing and the number of children they want to have and what they finally achieve. The other recurring topic of this dissertation is the attempt to improve the ways in which we model fertility decisions. Chapter 1 discusses economic uncertainty, one of the key determinants of the decision to have a child in contemporary societies. The chapter also offers an innovative way of modelling employment instability which takes into account more information from labor market trajectories than do previously used methods. Chapter 2 starts exploring dynamic modelling. An agent-based computational approach is applied to the evolution of the mean age at first birth, in which the decisions of couples are a function of their education and their economic (employment) situation but also of the decisions of their friends and the prevailing norms regarding the timing of the transition to adulthood in their society. Instead of focusing on the decision to have a first child, Chapter 3 presents a model that covers the entire reproductive process in an attempt to generate the evolution of the Total Fertility Rate from the bottom-up. The reproductive process is seen here as a hurdle race in which agents move towards their family size target as they navigate a series of obstacles, with a special focus on gender dynamics at the household level.
Los tres capítulos incluídos en esta tesis tienen el objetivo de mejorar nuestro conocimiento de la distancia que existe entre los deseos de las personas con respecto al momento y la cantidad de hijos que quieren tener y lo que finalmente consiguen. El otro tema recurrente de este trabajo es el intento por mejorar las formas en las que modelamos las decisiones con respecto a la fecundidad. El primer capítulo se enfoca en la incertidumbre económica, uno de los determinantes principales de la decisión de tener hijos en las sociedades contemporáneas. El capítulo también ofrece una propuesta novedosa para modelar la inestabilidad laboral, que tiene en cuenta una mayor cantidad de información sobre las trayectorias laborales que otras medidas más comunmente utilizadas. El capítulo 2 comienza a explorar modélos dinámicos. La evolución de la edad media al primer hijo es modelada a través de un enfoque computacional y basado en agentes, en el que las decisiones de las parejas son una función de su educación y su situación laboral, pero también de las decisiones de sus pares y de las normas sociales que regulan el timing de la transición a la adultez. El capítulo 3 utiliza un enfoque similar pero esta vez el modelo representa todo el proceso reproductivo en el intento de generar la evolución de la Tasa Global de Fecundidad desde las interacciones a nivel micro. El proceso reproductivo es modelado aquí como una carrera de obstáculos en la que las personas intentan alcanzar su número de hijos ideal contra una serie de restricciones, con especial foco en las dinámicas de género a nivel de la pareja.

Aquesta tesi presenta un marc per als models oberts de suport a les decisions en les organitzacions. El treball es vehicula a través d’un compendi d’articles on s’analitzen els fluxos d’entrada i de sortida de coneixement en les organitzacions, així como les tecnologies existents de suport a les decisions. Es presenten els factors subjacents que impulsen nous models per a formes obertes de suport a la decisió. La tesis presenta un estudi de les distintes tipologies de models de suport a les decisions tenint en compte diferents tipus d’organitzacions. En el primer estudi, paper#, es presenta l’evolució de les tecnologies de suport a les decisions i l’avançament de les noves tecnologies per als models oberts. Aquest estudi proporciona una visió des d’una perspectiva evolutiva de la relació entre el coneixement expert i la seva utilització en les tecnologies de suport a les decisions. La investigació revela l’entorn canviant que la tecnologia ofereix a l’hora de adquirir coneixement per a la presa de decisions i obre horitzons sobre el nou paper que els experts tenen en aquests entorns. Es suggereix que un canvi significatiu en la presa de decisions es basa en el desafiament entre el paper tradicional dels experts i no experts. Per últim, aquest treball explora les oportunitats d’integració de la intel•ligència artificial en la tecnologia de suport a les decisions i quins beneficis addicionals poden aportar les eines d’ intel•ligència col•lectiva en la presa de decisions. El segon estudi, paper#2, investiga sobre la tipologia anomenada "agregada" dins del marc d’entorns oberts per al suport a la presa de decisions. S’utilitza un problema de predicció com a fil conductor per a posar en relleu la complexitat de la previsió de la demanda dins de la industria del cinema. S’analitza com es pot utilitzar la tecnologia per a millorar l’eficàcia en les decisions. La investigació compara dues tecnologies de suport a les decisions: sistemes experts i eines d’intel•ligència col•lectiva, i il•lustra com l’industria del cinema utilitza cada una d’aquestes tecnologies en la previsió dels ingressos de taquilla. Per últim, aquest article explora els beneficis de l’ integració d’aquestes tecnologies de suport per a l’obtenció de prediccions més precises. El tercer estudi, article#3, presenta un estudi longitudinal durant un període de 10 anys que utilitza IBM “Innovation Jams” como un context per a la col•laboració a gran escala dins de la tipologia anomenada "plataforma". Aquest article investiga el paper de les “Innovation Jams”, en el canvi organitzacional i com IBM es compromet amb un nou model d’innovació en les organitzacions. En ell es descriuen les “Innovation Jams”, que han impulsat la innovació i consolidat la pràctica de la innovació oberta en IBM. En aquest article s’utilitza el gènere musical d’una "jamband" com una metàfora per a descriure el desenvolupament emergent i l’ús de les “Innovation Jams”, com una manera d’entendre el canvi organitzatiu. Aquest estudi longitudinal ofereix una visió actualitzada de la recerca en “Innovation Jams”, mostrant com han evolucionat des d’un concepte, a una eina de gestió i finalment a un servei. L’article conclou amb una discussió sobre les implicacions dels resultats i com aquests permeten teoritzar sobre nous models d’ innovació i el canvi en les organitzacions. La recerca duta a terme en aquesta tesi ofereix un marc per als models oberts de suport a la decisió, i suggereix que, les fonts internes i externes de coneixement poden ser utilitzades, més enllà de la innovació del producte o serveis, per a la presa de decisions amb el suport de tecnologies emergents. Les contribucions teòriques d’aquesta tesi sostenen que les organitzacions ja no poden confiar en la tecnologia de suport a les decisions que únicament es centren en la reducció de la frontera entre els aspectes racionals i no racionals de la conducta social humana, sinó que pel contrari, han de considerar la xarxa dinàmica de la organització per al suport a la decisió. D’altra banda, les implicacions pràctiques d’aquesta tesi animen les organitzacions a pensar estratègicament sobre com les tecnologies emergents poden ajudar en la presa de decisions i també com els models de decisió resultants poden ser utilitzats per a navegar per l’entorn complex existent, i, a la vegada, forjar vincles més forts amb els clients, proveïdors i la xarxa de l’organització.
Esta tesis presenta un marco para modelos abiertos de soporte a las decisiones en las organizaciones. El trabajo se vehicula a través de un compendio de artículos dónde se analizan los flujos de entrada y salida de conocimiento en las organizaciones, así como las tecnologías existentes de soporte a las decisiones. Se presentan los factores subyacentes que impulsan nuevos modelos para formas abiertas de soporte a la decisión. La tesis presenta un estudio de las distintas tipologías de modelos de soporte a las decisiones teniendo en cuenta distintos tipos de organizaciones. En el primer estudio paper#1 se presenta la evolución de las tecnologías de apoyo a las decisiones y el avance de las nuevas tecnologías para los modelos abiertos. Este estudio proporciona una visión desde una perspectiva evolutiva de la relación entre conocimiento experto y su utilización en las tecnologías de soporte a las decisiones. La investigación revela el entorno cambiante que la tecnología ofrece a la hora de adquirir conocimiento para la toma de decisiones y abre horizontes sobre el nuevo papel que los expertos tienen en estos entornos. Se sugiere que un cambio significativo en la toma de decisiones se basa en el desafío entre el papel tradicional de los expertos y no expertos. Por último, este trabajo explora las oportunidades de integración de la inteligencia artificial en la tecnología de soporte de decisiones y que beneficios adicionales pueden aportar las herramientas de inteligencia colectiva en la toma de decisiones. El segundo estudio, paper#2, investiga sobre la tipología llamada "agregada" dentro del marco de entornos abiertos para el soporte a la toma de decisiones. Se utiliza un problema de predicción como hilo conductor para poner en relieve la complejidad de la previsión de la demanda dentro de la industria del cine. Se analiza cómo se puede utilizar la tecnología para mejorar la eficacia en las decisiones. La investigación compara dos tecnologías de soporte a las decisiones: sistemas expertos y herramientas de inteligencia colectiva, e ilustra cómo la industria del cine utiliza cada una de estas tecnologías en la previsión de los ingresos de taquilla. Por último, este artículo explora los beneficios de la integración de estas tecnologías de apoyo para la obtención de predicciones más precisas. El tercer estudio, artículo #3, presenta un estudio longitudinal durante un período de 10 años que utiliza IBM “Innovation Jams”, como un contexto para la colaboración a gran escala dentro de la tipología llamada "plataforma". Este artículo investiga el papel de las “Innovation Jams”, en el cambio organizacional y como IBM se compromete con un nuevo modelo de innovación de la organización. En él se describen las “Innovation Jams”, que han impulsado la innovación y consolidado la práctica de la innovación abierta en IBM. En este artículo se utiliza el género musical de una "jamband" como una metáfora para describir el desarrollo emergente y el uso de las “Innovation Jams”, como una manera de entender el cambio organizativo. Este estudio longitudinal ofrece una visión actualizada de la investigación en “Innovation Jams”, mostrando cómo han evolucionado desde un concepto, a una herramienta de gestión y finalmente a un servicio. El artículo concluye con una discusión sobre las implicaciones de los resultados y como ellos permiten teorizar sobre nuevos modelos de innovación y el cambio en las organizaciones. La investigación llevada a cabo en esta tesis ofrece un marco para los modelos abiertos de apoyo a la decisión, y sugiere que el uso de fuentes internas y externas de conocimiento pueden ser utilizadas más allá de la innovación del producto o servicio para la toma de decisiones con el soporte de tecnologías emergentes. Las contribuciones teóricas de esta tesis sostienen que las organizaciones ya no pueden confiar en la tecnología de apoyo a las decisiones que únicamente se centran en la reducción de la frontera entre los aspectos racionales y no racionales de la conducta social humana, sino por el contrario, deben considerar la red dinámica de la organización para el apoyo a la decisión. Por otra parte, las implicaciones prácticas de esta tesis alienta a las organizaciones a pensar estratégicamente acerca de cómo las tecnologías emergentes pueden ayudar a la toma de decisiones y también cómo los modelos de decisión resultantes pueden ser utilizados para navegar por el entorno complejo existente y, a su vez, forjar vínculos más fuertes con los clientes, proveedores y más amplios de la red de la organización.
This thesis presents a framework for open models of decision support through a compendium of papers that links research on the inward and outward flows of knowledge to the organization and decision support technologies. The framework presents underlying factors driving new and more open models of decision support. A typology of decision support models is offered considering types of problems organizations and managers charged with decision-making face. Thesis essay #1 suggests a perspective of the changing landscape for decision support technology and the advancement of new technology for open models of decision support. This study provides insight from an evolutionary perspective of expertise that has shaped the field of decision support technologies. The investigation sets out to reveal the changing landscape of expertise in supporting decision-making using technology and sheds light on the new role that experts will play in organizational decision-making. It suggests that a significant change in how decision-making is being supported which challenge the traditional role of experts and non-experts. Finally, this paper explores opportunities for decision support technology integration and the added benefits artificial intelligence can bring to collective intelligence tools. Thesis essay #2 investigates the ‘aggregate’ typology within the open model decision support framework. A forecasting problem is used to highlight the complexity of demand forecasting in supply-chain management within the film industry and how technology is leveraged for effective supply-chain management decisions. The investigation compares two decision support technologies: expert systems and collective intelligence tools and illustrates how the film industry uses each in forecasting box-office revenue. Finally, this essay explores the combined benefits in integrating each support technology for more accurate forecasting. Thesis essay #3 is a longitudinal study over a 10 year period that uses IBM Innovation Jams as a context for large-scale collaboration within the ‘platform’ typology. This essay investigates the role of innovation jams on organizational change as IBM learned to engage with a new model of organizing innovation. It describes the role innovation jams have played in shaping the practice of open innovation at IBM. This essay uses the musical genre of a “jamband” as a metaphor to describe the emergent development and use of innovation jams as a way to understand organizational change. This longitudinal study brings innovation jam research up-to-date and presents innovation jams as they evolved from a concept, a management tool, and service. The essay concludes with a discussion on the implications of the findings for theorizing about new models of organizing innovation for organizational change. Research conducted in this thesis offers a framework of open models of decision support that suggests that the use of internal and external sources of knowledge can be leveraged beyond product or service innovation, to include decision-making supported by emerging technology. Theoretical contributions of this thesis argues that organizations can no longer rely on decision support technology that solely focus on bridging the boundary between rational and non-rational aspects of human social behavior but instead, must consider the larger dynamic organizational network for decision support. Moreover, practical implications of this thesis encourages organizations to think strategically about how emerging technology can support decision making and the resulting decision support models to navigate the complex environment they work in and in turn, to forge stronger links with customers, suppliers, and the wider organizational network.

In this dissertation, we present a collection of manuscripts describing the development of prognostic models designed to assist clinical decision making. This work is motivated by limitations of commonly used techniques to produce accessible prognostic models with easily interpretable and clinically credible results. Such limitations hinder prognostic model widespread utilization in medical practice. Our methodology is based on Rough Set Theory (RST) as a mathematical tool for clinical data anal- ysis. We focus on developing rule-based prognostic models for end-of life care decision making in an effort to improve the hospice referral process. The development of the prognostic models is demonstrated using a retrospective data set of 9,103 terminally ill patients containing physiological characteristics, diagnostic information and neurological function values. We develop four RST-based prognostic models and compare them with commonly used classification techniques including logistic regression, support vector machines, random forest and decision trees in terms of characteristics related to clinical credibility such as accessibility and accuracy. RST based models show comparable accuracy with other methodologies while providing accessible models with a structure that facilitates clinical interpretation. They offer both more insight into the model process and more opportunity for the model to incorporate personal information of those making and being affected by the decision.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 1986.
MICROFICHE COPY AVAILABLE IN ARCHIVES AND ENGINEERING.
Bibliography: leaves 227-232.
by Paul Thayer Chapman.
Ph.D.

In this thesis we formalise and study computational aspects of group decision making for rational, self-interested agents. Specifically, we are interested in systems where agents reach consensus according to endogenous thresholds. Natural groups have been shown to make collective decisions according to threshold-mediated behaviours. An individual will commit to some collective endeavour only if the number of others having already committed exceeds their threshold. Consensus is reached only where all individuals express commitment. We present a family of models that describe fundamental aspects of cooperative behaviour in multi-agent systems. These include: coalition formation, participation in joint actions and the achievement of individuals’ goals over time. We associate novel solution concepts with our models and present results concerning the computational complexity of several natural decision problems arising from these. We demonstrate potential applications of our work by modelling a group decision problem common to many cohesive groups: establishing the location of the group. Using model checking tools we compute the effects of agents’ thresholds upon outcomes. We consider our results within an appropriate research context.

Humanitarian relief and aid organizations all over the world implement efforts aimed at recovering from disasters, reducing poverty and promoting human rights. The purpose of this dissertation is to develop a series of quantitative decision models to help address some of the challenges faced by humanitarian logistics. The first study discusses the development of a spreadsheet-based multicriteria scheduling model for a small development aid organization in a South American developing country. Development aid organizations plan and execute efforts that are primarily directed towards promoting human welfare. Because these organizations rely heavily on the use of volunteers to carry out their social mission, it is important that they manage their volunteer workforce efficiently. In this study, we demonstrate not only how the proposed model helps to reduce the number of unfilled shifts and to decrease total scheduling costs, but also how it helps to better satisfy the volunteers’ scheduling preferences, thus supporting long-term retention and effectiveness of the workforce. The purpose of the second study is to develop a decision model to assist in the management of humanitarian relief volunteers. One of the challenges faced by humanitarian organizations is that there exist limited decision technologies that fit their needs while it has also been pointed out that those organizations experience coordination difficulties with volunteers willing to help. Even though employee workforce management models have been the topic of extensive research over the past decades, no work has focused on the problem of managing humanitarian relief volunteers. In this study, we discuss a series of principles from the field of volunteer management and develop a multicriteria optimization model to assist in the assignment of both individual volunteers and volunteer groups to tasks. We present illustrative examples and analyze two complementary solution methodologies that incorporate the decision maker's preferences and knowledge and allow him/her to trade-off conflicting objectives. The third study discusses the development of a decision model for the procurement of goods in humanitarian efforts. Despite the prevalence of procurement expenditures in humanitarian efforts, procurement in humanitarian contexts is a topic that has only been discussed in a qualitative manner in the literature. In our paper, we introduce a two stage decision model with recourse to improve the procurement of goods in humanitarian relief supply chains and present an illustrative example. Conclusions, limitations, and directions for future research are also discussed.
Ph. D.

Passenger flow studies in airport terminals have shown consistent statistical relationships between airport spatial layout and pedestrian movement, facilitating prediction of movement from terminal designs. However, these studies are done at an aggregate level and do not incorporate how individual passengers make decisions at a microscopic level. Therefore, they do not explain the formation of complex movement flows. In addition, existing models mostly focus on standard airport processing procedures such as immigration and security, but seldom consider discretionary activities of passengers, and thus are not able to truly describe the full range of passenger flows within airport terminals. As the route-choice decision-making of passengers involves many uncertain factors within the airport terminals, the mechanisms to fulfill the capacity of managing the route-choice have proven difficult to acquire and quantify. Could the study of cognitive factors of passengers (i.e. human mental preferences of deciding which on-airport facility to use) be useful to tackle these issues? Assuming the movement in virtual simulated environments can be analogous to movement in real environments, passenger behaviour dynamics can be similar to those generated in virtual experiments. Three levels of dynamics have been devised for motion control: the localised field, tactical level, and strategic level. A localised field refers to basic motion capabilities, such as walking speed, direction and avoidance of obstacles. The other two fields represent cognitive route-choice decision-making. This research views passenger flow problems via a "bottom-up approach", regarding individual passengers as independent intelligent agents who can behave autonomously and are able to interact with others and the ambient environment. In this regard, passenger flow formation becomes an emergent phenomenon of large numbers of passengers interacting with others. In the thesis, first, the passenger flow in airport terminals was investigated. Discretionary activities of passengers were integrated with standard processing procedures in the research. The localised field for passenger motion dynamics was constructed by a devised force-based model. Next, advanced traits of passengers (such as their desire to shop, their comfort with technology and their willingness to ask for assistance) were formulated to facilitate tactical route-choice decision-making. The traits consist of quantified measures of mental preferences of passengers when they travel through airport terminals. Each category of the traits indicates a decision which passengers may take. They were inferred through a Bayesian network model by analysing the probabilities based on currently available data. Route-choice decision-making was finalised by calculating corresponding utility results based on those probabilities observed. Three sorts of simulation outcomes were generated: namely, queuing length before checkpoints, average dwell time of passengers at service facilities, and instantaneous space utilisation. Queuing length reflects the number of passengers who are in a queue. Long queues no doubt cause significant delay in processing procedures. The dwell time of each passenger agent at the service facilities were recorded. The overall dwell time of passenger agents at typical facility areas were analysed so as to demonstrate portions of utilisation in the temporal aspect. For the spatial aspect, the number of passenger agents who were dwelling within specific terminal areas can be used to estimate service rates. All outcomes demonstrated specific results by typical simulated passenger flows. They directly reflect terminal capacity. The simulation results strongly suggest that integrating discretionary activities of passengers makes the passenger flows more intuitive, observing probabilities of mental preferences by inferring advanced traits make up an approach capable of carrying out tactical route-choice decision-making. On the whole, the research studied passenger flows in airport terminals by an agent-based model, which investigated individual characteristics of passengers and their impact on psychological route-choice decisions of passengers. Finally, intuitive passenger flows in airport terminals were able to be realised in simulation.

Ingeniero Civil
Los modelos de elección discreta basados en utilidades o disposiciones a pagar han sido muy utilizados para representar el comportamiento de los consumidores en sistemas de Uso de Suelo y de Transporte. En este trabajo se desarrolla un modelo de elección que usa disposiciones a pagar para representar la elección en mercados de remate, donde los bienes se asignan a los mejores postores. Los modelos tradicionales no consideran el hecho de que estas funciones están siempre acotadas superiormente porque los consumidores tienen restricciones presupuestarias. Si bien se han generado enfoques para representar esta situación, estas son heurísticas solamente. En este trabajo se deduce de un modo riguroso un nuevo modelo de elección discreta binomial donde las funciones de valor son disposiciones a pagar acotadas superiormente. En particular, se obtiene una expresión analítica en el caso en que estas funciones son variables aleatorias idénticamente distribuidas Gumbel. Este modelo junto con incorporar en su formulación la influencia de la restricción presupuestaria, tiene la interesante propiedad de permitir estimar un número mayor de parámetros que en el caso del modelo Logit Binomial clásico. Junto con caracterizar este modelo, se probó empíricamente usando bases sintéticas que es posible estimar sus parámetros y que en muchos casos, entrega estimaciones más precisas que el modelo Logit Binomial tradicional.

The recent extending empirical evidence regarding the power of factor models versus the traditional CAPM has motivated the research in the current thesis. Substantial controversy has been raised over two issues: 1) Are the new factors, market value and book-to-market equity, the most important sources of risk? and 2) Is it time to consider CAPM as a useless model? Effectively, these are the main questions we attempt to address in the current research within a unified framework of firm attributes and more aspects of the econometrical applied approaches. The main findings of the empirical research in this thesis show that, firstly the beta portfolio returns exhibit the highest volatility, confirming thus the beta as the most significant risk source. Secondly, the market portfolio absorbs the excess returns of the majority of value-weighted factor portfolios which is partly attributed to the mitigation of the January effect. In the seasonality area, we identify a strong October effect with high volatility but not high returns, a phenomenon that cannot be explained with a rational story. The re-examination of the Fama and French 1992 model with corrections of econometrical problems and the application of panel data methodology reveals that the sole significant factor over all the candidate variables is the price variable. Yet, even the power of the price factor is eliminating with the application of non-linear systems where the CAPM constraints are directly validated but with a negative sign. However, the presence of negative risk premium is consistent with the valid application of CAPM in a financial world where the occurrence of bad states of world is more frequent than the presence of up markets. Overall, the results of this thesis contribute to a thorough understanding of the factor models' performance which plays a key role in the financial investment decisions. The implication is that the CAPM should be still regarded as the basic financial model in the risk-return management process.

This doctoral thesis consists of five research articles on the general topic of optimal decision making under uncertainty in a Bayesian framework. The papers are preceded by three introductory chapters. Papers I and II are dedicated to the problem of finding an optimal stopping strategy to liquidate an asset with unknown drift. In Paper I, the price is modelled by the classical Black-Scholes model with unknown drift. The first passage time of the posterior mean below a monotone boundary is shown to be optimal. The boundary is characterised as the unique solution to a nonlinear integral equation. Paper II solves the same optimal liquidation problem, but in a more general model with stochastic regime-switching volatility. An optimal liquidation strategy and various structural properties of the problem are determined. In Paper III, the problem of sequentially testing the sign of the drift of an arithmetic Brownian motion with the 0-1 loss function and a constant cost of observation per unit of time is studied from a Bayesian perspective. Optimal decision strategies for arbitrary prior distributions are determined and investigated. The strategies consist of two monotone stopping boundaries, which we characterise in terms of integral equations. In Paper IV, the problem of stopping a Brownian bridge with an unknown pinning point to maximise the expected value at the stopping time is studied. Besides a few general properties established, structural properties of an optimal strategy are shown to be sensitive to the prior. A general condition for a one-sided optimal stopping region is provided. Paper V deals with the problem of detecting a drift change of a Brownian motion under various extensions of the classical Wiener disorder problem. Monotonicity properties of the solution with respect to various model parameters are studied. Also, effects of a possible misspecification of the underlying model are explored.

The quantification of choice has been a major area of research for behavioural scientists for several decades. This is, in part, due to the discovery of the matching law (Herrnstein, 1961) which stipulates that relative response rates on concurrently available alternatives “match” the available relative reinforcement rates. This theoretical construct has been developed to describe response allocation in more complex situations, such as concurrent chains, and successfully describes both human and non-human behaviour. Typically, this phenomenon becomes evident when behaviour settles at an asymptote after several sessions of training where contingencies are held constant, and is often called “steady-state” behaviour. However, a fundamental question still remains: what causes matching – that is, what are the underlying momentary process(es) that produce matching? Researchers have suggested that what is necessary to answer this question is to take a molecular approach to the analysis of choice behaviour, thereby assessing choice in transition (Grace, 2002a). Recently, a new model of choice acquisition has been developed that appears to offer promise. It combines two separate mechanisms; a “winner-takes-all” categorical discrimination, and a linear-operator acquisition process (Grace & McLean, 2006). The initial results suggest this model could provide an alternative explanation for what underlies matching – that two separate processes are cooccurring in the acquisition of choice behaviour – allowing response allocation to be either linear or non-linear. This thesis extends the Grace and McLean model to include the situation of response strength ‘carrying-over’ from session to session to describe the process of acquisition gradually accumulating with experience. Moreover, additional assumptions have been added to describe temporal phenomena 2 and presumed discounting of previous experience on current choice behaviour. A steady-state version of the extended model was derived and, when fitted to published data sets, describes choice behaviour equally well when compared to existing models of steady-state choice. As a consequence of these additions, the Extended Decision Model (EDM) predicts a unique response allocation pattern – choice behaviour follows a bitonic function when initial-link durations were increased and the terminallink delays were held constant. The results from experiments presented in this thesis support this prediction, whilst steady-state analyses found the EDM was parameter invariant – differences between parameters from two schedule types across several archival data sets were non-significant, while existing steady-state models had significant differences. These findings provide further support for the claim that the EDM and the Decision Model (DM) mechanisms provide unique and accurate descriptions of the molecular processes governing choice behaviour. Moreover, the implication from these results is that the underlying assumption of the EDM and DM – that choice is determined by the propensity to respond rather than conditioned reinforcement – appears to have further foundation. This challenges the assumptions of existing models of choice behaviour and presents the possibility that probabilistic approaches are perhaps more appropriate for describing response allocations than discrete estimates of relative value when contingencies change.

Standard finance theory portrays investors as rational utility maximisers. Persisting market anomalies and observed investor practice, however, have led to widespread recognition that the fundamental axioms of rationality are often violated. In response to the limitations inherent in standard theory, the Behavioural Finance approach relaxes the rationality assumption and takes account of psychological influences on individuals’ decision-making processes. Adopting the behavioural approach, this thesis, which includes two empirical studies, examines why, and to what extent, investors depart from rational or optimal investment practices. The thesis examines the effect of Myopic Loss Aversion (MLA) suggested by Benartzi and Thaler (1995) as a response to the Equity Premium Puzzle highlighted by Mehra and Prescott (1985). While previous studies are almost exclusively based on experiments in a laboratory setting, this approach provides more compelling empirical evidence by investigating the effects of MLA on real individual investors’ portfolio allocations through the use of the Dutch National Bank Household Survey. For the first time, the concept of MLA is identified through the interaction of two separate effects, firstly, individuals’ myopia, reflected in portfolio evaluation and rebalancing frequencies, and secondly, loss aversion. The thesis finds that individuals who are less affected by MLA invest more in risky financial assets. Further, individuals who are less myopic increase their share of risky assets invested in their financial portfolios over time, although this is unrelated to their loss aversion. These findings support the prediction of MLA theory that short investment horizons and high loss aversion lead to a significantly lower share of risky investments. In summary, the high equity premium can be explained by the notion of MLA. If individuals evaluate their investment performance over the long-term, they perceive much smaller risks relative to stockholding returns; consequently, they will be prepared to accept smaller equity premiums. The findings suggest possible interventions by policy makers and investment advisors to encourage individuals to remain in the stock market, such as providing long-term investment instruments, or restricting evaluation frequency to the annual reporting of investment performance. In response to the stockholding puzzle (Haliassos and Bertaut, 1995), this thesis also investigates individuals’ stock market returns expectations and their varying levels of risk aversion. Previous studies find that individuals’ heterogeneous stock market expectations determine variations in their stockholdings. The thesis accounts for the effect of risk aversion on stock market expectations, as well as on stockholding decisions. Additionally, the causality issue as between individuals’ expectations and stockholding status is controlled. The thesis finds that more risk averse individuals hold lower stock market expectations, and that the stock market return expectations of more risk averse individuals affect their stock market participation decisions negatively. The portfolio allocation decisions of individuals who already hold stocks are only affected by their expectations, with risk aversion being no longer significant. The thesis argues that persistent risk aversion effects cause individuals to hold pessimistic views of stock market returns, thus contributing to the enduring stockholding puzzle. The thesis reinforces existing perceptions that individuals in the real world may not make fully rational decisions due to their judgments which are based on heuristics and affected by cognitive biases. Individual investors often fail to maximise their utility given their preferences and constraints. Consequently, this thesis draws attention to the possible role of institutions, policy makers, and financial advisory bodies in providing effective interventions and guidelines to improve individuals’ financial decisions.

This work presents the first known attempt to model the dairy business from a multiscale modelling perspective. The multiscale nature of the dairy industry is examined with emphasis on those key decision making and process scales involved in production. Decision making scales identified range from the investor level to the plant operator level, and encompass business, production, plant, and operational levels. The model considers scales from the production manager to the unit operation scale. The cheese making process is used to demonstrate scale identification in the context of the important phenomena and other natural levels of scrutiny of interest to decision makers. This work was a first step in the establishment of a multiscale system model capable of delivering information for process troubleshooting, scheduling, process and business optimization, and process control decision-making for the dairy industry. Here, only material transfer throughout a process, use of raw materials, and production of manufactured product is modelled. However, an implementation pathway for adding other models (such as the precipitation of milk protein which forms curd) to the system model is proposed. The software implementation of the dairy industry multiscale model presented here tests the validity of the proposed: • object model (object and collection classes) used to model unit operations and integrate them into a process, • mechanisms for modelling material and energy streams, • method to create simulations over variable time horizons. The model was implemented using object oriented programming (OOP) methods in conjunction with technologies such as Visual Basic .NET and CAPE-OPEN. An OOP object model is presented which successfully enabled the construction of a multiscale model of the cheese making process. Material content, unit operation, and raw milk supply models were integrated into the multiscale model. The model is capable of performing simulations over variable time horizons, from 1 second, to multiple years. Mechanisms for modelling material streams, connecting unit operations, and controlling unit operation behaviour were implemented. Simple unit operations such as pumps and storage silos along with more complex unit operations, such as a cheese vat batch, were modelled. Despite some simplifications to the model of the cheese making process, the simulations successfully reproduced the major features expected from the process and its constituent unit operations. Decision making information for process operators, plant managers, production managers, and the dairy business manager can be produced from the data generated. The multiscale model can be made more sophisticated by extending the functionality of existing objects, and incorporating other scale partial models. However, increasing the number of reported variables by even a small number can quickly increase the data processing and storage demands of the model. A unit operation’s operational state of existence at any point of time was proposed as a mechanism for integrating and recalculating lower scale partial models. This mechanism was successfully tested using a unit operation’s material content model and is presented here as a new concept in multiscale modelling. The proposed modelling structure can be extended to include any number of partial models and any number of scales.

This thesis examines the dynamic decision making behavior of fishermen. Two models are developed: (l) an intraseasonal model of vessel movement on the fishing ground during each season; and (2) an interseasonal model for investment decision making from year to year. Both decision models are driven by single economic objectives and the fisherman-decision maker is assumed to make rational choices to optimize the stated objective. In this competitive market intraseasonal decisions are assumed to be made in the short-run to maximize the net operating income of each fishing enterprise. These decisions about where to fish to obtain the maximum return to fishing effort over the course of the season are modelled by a partially observable Markov decision process which incorporates the key elements of the problem facing each fisherman. The state space for this process is derived from total seasonal biomass. This aggregate description of the state space renders the problem practicable and solvable. The normative model is developed formally and applied to freezer trollers of the British Columbia commercial fishing fleet. Model results for average income and catch per troller, and seasonal fishing distribution over the fishing grounds reflect major tendencies in statistics arising from actual intraseasonal decisions made by this group of fishermen. Interseasonal decisions concerning longer-term investment strategies are made in an environment which is highly variable from season-to-season. Extensive variability implies that economic survival is a primary consideration in the investment decision process. The investment decision making process is modelled as a probabilistic dynamic programming problem in discrete time. Investors are assumed to make rational decisions based on income expectations and subject to survivability conditions to maximize the net worth of the fishing enterprise at the end of a finite planning horizon. The formal analysis of the investment model is presented. The model is applied to all trollers of the British Columbia commercial fishing fleet. The pattern of actual investment by troller is simulated by tuning behavioral components of the investment model. These results provide insight into the behavioral basis of investment decision making by this group of fishermen. This modelling framework has implications for planning and regulation in fisheries. Insight gained into the key factors behind fishermen's decisions can provide a basis for the development of strategic policies which anticipate fishermen's behavior and are aimed at stabilizing the economic viability of the fishing sector. The approach represents a movement away from reactive, short-term policies which have characterized fisheries regulation to date.
Business, Sauder School of
Graduate

This thesis is concerned with an investigation of solution methods for continuous multiple objective decision models (MODM's). A number of different solution methods which have appeared in the literature are reviewed with an emphasis on the underlying concepts of the methods. The following chapter examines the solution of MODM’S from the other side, namely the behavioural aspects of decision making. Having gained an appreciation of exactly how people do make decisions I the intent of the thesis is twofold. Firstly, to develop new solution methods which can accommodate the decision maker (DM) in whatever his or her particular decision strategy is. And secondly, it is to empirically examine four solution methods with respect to users' preferences among them. Of these four solution methods, three are among the most well known in the literature and all can cite practical application. Two new solution methods have been developed. Both of these methods are based on a specific formulation of the MODM which is known as the maxmin formulation. The theory of the maxmin formulation is developed in Chapter 4. By using the Lagrange multipliers at the optimal solution, suitable pairwise tradeoff information can be presented to the DM. This forms the basis for the first solution method, which interacts with the DM as he or she progressively provides preference information. The other solution method makes use of a branch of Psychology called Social Judgement Theory and incorporates this into the solution method. This second method is especially applicable to the multiple DM situation. In the empirical examination of solution methods it was found that one solution method was clearly preferred over the other three. The thesis concludes with a discussion of approaches for reducing the number of objectives in a MODM.

There are a wide range of techniques on offer to decision makers choosing between options where each option exhibits a range of attributes. Many of these techniques involve eliciting weights to represent the relative importance of each attribute. This thesis offers a mathematical explanation for the consistent differences in the distribution of weights experienced when a fixed sum method, Point Allocation (PA), and a fixed scale method, Direct Rating (DR), are used. Fixed scale and fixed sum simulations, sampling from the Uniform distribution, produce different weight profiles matching those found in practical applications. Formulae are found representing the distribution of weights produced by the simulations. These enable ranked weights to be calculated, which can be used as surrogates for 'true' weights. In particular, a second major aspect of the study concerns the discovery of a family of piecewise probability density functions to represent the distributions of ranked weights generated using the DR method. The means of the distributions are the Rank Order Distribution (ROD) surrogate weights. These are compared with the Rank Order Centroid (ROC) weights. As the number of attributes in a decision problem increases, the ROD weights approximate to the more easily calculated Rank Sum weights. An interactive survey is conducted, using students and the internet, to compare the PA and DR methods in terms of their ease of use, time taken, accuracy achieved, and user's confidence, in producing weights that represented a series of known "true" weights presented in graphical form. Statistical analysis found significant effects of method, time, accuracy achieved, and confidence of participants, favouring the DR approach. The methods are also compared as a means of obtaining the importance weights given by students to seven listed attributes of universities. A significant difference is found between the pattern of decision weights produced using the two methods confirming previously published studies.

Die Dynamik des Erdsystems im Anthropozän wird durch eine zunehmende Verschränkung von Prozessen auf physikalischer und ökologischer sowie auf sozioökonomischer Ebene bestimmt. Wenn Modelle als Entscheidungshilfen in diesem Umfeld nützlich sein sollen, müssen sie diese komplexen Rückkopplungen ebenso berücksichtigen wie die inhärent emergenten und heterogenen Qualitäten gesellschaftlicher Dynamik. Diese Arbeit schlägt vor, den Menschen als begrenzten rationalen Entscheidungsträger zu modellieren, die (soziales) Lernen nutzen, um Entscheidungsheuristiken zu erwerben, die in einer gegebenen Umgebung gut funktionieren. Dies wird in einem Wirtschaftsmodell mit zwei Sektoren veranschaulicht, in dem ein Sektor eine fossile Ressource für die wirtschaftliche Produktion verwendet und die Haushalte ihre Investitionsentscheidungen in der zuvor beschriebenen Weise treffen. In der Modellökonomie können individuelle Entscheidungsfindung und soziale Dynamik die CO 2 Emissionen nicht auf ein Niveau begrenzen, das eine globale Erwärmung über 1,5◦C verhindert. Eine Kombination aus kollektivem Handeln und koordinierter öffentlicher Politik allerdings kann. Eine Folgestudie analysiert das soziale Lernen der individuellen Sparquoten in einer Ein-Sektor-Wirtschaft. Hier nähert sich die aggregierte Sparquote der eines intertemporär optimierenden allwissenden Sozialen Planers an, wenn die soziale Interaktionsrate ausreichend niedrig ist. Gleichzeitig führt eine abnehmende Interaktionsrate einem plötzlichen Übergangs von einer unimodalen zu einer stark bimodalen Verteilung des Vermögens unter den Haushalten. Schließlich schlägt diese Arbeit eine Kombination verschiedener Methoden vor, die zur Ableitung analytischer Näherungen für solche vernetzten heterogenen Agentenmodelle verwendet werden können, bei denen Interaktionen zwischen Agenten sowohl auf individueller als auch auf aggregierter Ebene auftreten.
The trajectory of the Earth system in the Anthropocene is governed by an increasing entanglement of processes on a physical and ecological as well as on a socio-economic level. If models are to be useful as decision support tools in this environment, they ought acknowledge these complex feedback loops as well as the inherently emergent and heterogeneous qualities of societal dynamics. This thesis improves the capability of social-ecological and socio-economic models to picture emergent social phenomena and uses and extends techniques from dynamical systems theory and statistical physics for their analysis. It proposes to model humans as bounded rational decision makers that use (social) learning to acquire decision heuristics that function well in a given environment. This is illustrated in a two sector economic model in which one sector uses a fossil resource for economic production and households make their investment decisions in the previously described way. In the model economy individual decision making and social dynamics can not limit CO 2 emissions to a level that prevents global warming above 1.5 ◦ C. However, a combination of collective action and coordinated public policy actually can. A follow up study analyzes social learning of individual savings rates in a one sector investment economy. Here, the aggregate savings rate in the economy approaches that of an intertemporarily optimizing omniscient social planner if the social interaction rate is sufficiently low. Sumultaneously, a decreasing interaction rate leads to emergent inequality in the model in the form of a sudden transition from a unimodal to a strongly bimodal distribution of wealth among households. Finally, this thesis proposes a combination of different moment closure techniques that can be used to derive analytic approximations for such networked heterogeneous agent models where interactions between agents occur on an individual as well as on an aggregated level.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Economics, September, 2020
Cataloged from student-submitted PDF of thesis.
Includes bibliographical references.
This thesis consists of three separate papers concerning the use of communication channels and intermediaries in organizations. A noisy-signalling model of strategic communication is introduced in the first chapter, and expanded upon in the remainder of the thesis. In the second part of the first chapter, I use the core noisy-signalling model to study organizational design of a single channel of communication. The results of the analysis provide a rational for the variation in communication processes observed across organizations, as well as costly political lobbying and advertising campaigns. In the second chapter, I extend the core model to allow the informed party to choose among multiple communication channels when conversing with the decision maker. The model suggests that polarization across communication channels may be an efficient response to "bandwidth" concerns facing decision-makers of large corporations or unqualified management. Conversely, coexistence of partisan and non-partisan channels within an organization or community (e.g. tabloids and professional news sources in the journalism industry) may also be socially efficient for other environments. In the third chapter, I consider a different extension of the core model by allow- ing the two parties to communicate via a strategic intermediary. I use the model to provide a possible explanation for the variety of roles communication intermediaries play in different organizations, the correlation between control-rights and communication hierarchies in organizations, as well usage of third-party, conflict-resolution arrangements.
by Ali Fakhruddin Palida.
Ph. D.
Ph.D. Massachusetts Institute of Technology, Department of Economics

Coreference is an important and frequent concept in any form of discourse, and Coreference Resolution (CR) a widely used task in Natural Language Understanding (NLU). In this thesis, we implement and explore two recent models that include the concept of coreference in Recurrent Neural Network (RNN)-based Language Models (LM). Entity and reference decisions are modeled explicitly in these models using attention mechanisms. Both models learn to save the previously observed entities in a set and to decide if the next token created by the LM is a mention of one of the entities in the set, an entity that has not been observed yet, or not an entity. After a theoretical analysis where we compare the two LMs to each other and to a state of the art Coreference Resolution system, we perform an extensive quantitative and qualitative analysis. For this purpose, we train the two models and a classical RNN-LM as the baseline model on the OntoNotes 5.0 corpus with coreference annotation. While we do not reach the baseline in the perplexity metric, we show that the models’ relative performance on entity tokens has the potential to improve when including the explicit entity modeling. We show that the most challenging point in the systems is the decision if the next token is an entity token, while the decision which entity the next token refers to performs comparatively well. Our analysis in the context of a text generation task shows that a wide-spread error source for the mention creation process is the confusion of tokens that refer to related but different entities in the real world, presumably a result of the context-based word representations in the models. Our re-implementation of the DeepMind model by Yang et al. 2016 performs notably better than the re-implementation of the EntityNLM model by Ji et al. 2017 with a perplexity of 107 compared to a perplexity of 131.

The focus of this thesis is on the application of decision models to the economic evaluation of health care technologies. The primary objective addresses the correct choice of modelling technique, as the attributes of the chosen technique could have a significant impact on the process, as well as the results, of an evaluation. Separate decision models, a Markov process and a discrete event simulation (DES) model are applied to a case study evaluation comparing alternative adjuvant therapies for early breast cancer. The case study models are built and analysed as stochastic models: whereby probability distributions are specified to represent the uncertainty about the true values of the model input parameters. Three secondary objectives are also specified. Firstly, the empirical application of the alternative decision models requires the specification of a 'modelling process' that is not well defined in the health economics literature. Secondly, a comparison of alternative methods for specifying probability distributions to describe the uncertainty in the model's input parameters is undertaken. The final secondary objective covers the application of methods for valuing the collection of additional information to inform the resource allocation decision. The empirical application of the two relevant modelling techniques clarifies the potential advantages derived from the increased flexibility provided by DES over Markov models. The thesis concludes that the use of DES should be strongly considered if either of the following issues appear relevant: model parameters are a function of the time spent in particular states, or the data describing the timing of events are not in the form of transition probabilities. The full description of the modelling process provides a resource for health economists wanting to use decision models. No definitive process is established, however, as there exist competing methods for various stages of the modelling process. The main conclusion from the comparison of methods for specifying probability distributions around the input parameters is that the theoretically specified distributions are most likely to provide a common baseline for comparisons between evaluations. The central question that remains to be addressed is which method is the most theoretically correct? The application of a Vol analysis provides useful insights into the methods employed and leads to the identification of particular methodological issues requiring future research in this area.

We present a method for finding the optimal decision on Random Variables in a graphical model. Upper and lower bounds on the exact value for each decision are used to reduce the complexity of the algorithm, while we still ensure that the decision chosen actually represents the exact optimal choice. Since the highest lower bound value is also a lower bound on the value of the optimal decision, we rule out any candidate with an upper bound of lower value than the highest lower bound. By this strategy, we try to reduce the number of candidates to a number we can afford to do exact calculations on.We generate five Bayesian Networks with corresponding value functions, and apply our strategy to these. The bounds on the values are obtained by use of an available computer program, where the complexity is controlled by an input constant. We study the number of decisions accepted for different values of this input constant. From the first Network, we learn that the bounds does not work well unless we split the calculations into parts for different groups of the nodes. We observe that this splitting works well on the next three Networks, while the last Network illustrates how the method fails when we add more edges to the graph. We realize that our improved strategy is successful on sparse graphs, while the method is unsuccessful when we increase the density of edges among the nodes.