Manufacturers address the distinct operational objectives of product innovation and manufacturing efficiency by having separate fabrication facilities (“fabs”) for development and manufacturing. Additionally, the industrial manufacture of a semiconductor product proceeds through several stages of production. These are typically a research and development (R&D) stage, a ramping stage, and a manufacturing stage. These production stages are distributed over the different fabs. These differences in fabrication environment and stage of production result in differences in the characteristics of production of a semiconductor product over its manufacturing lifetime. Some examples of these differences are device yield, breadth of processing conditions, throughput, number of reaction chambers operating in parallel, metrology, and data collection. These differences are reflected in the data available in the fab databases. This research explores the use of a neural network modeling and genetic algorithm optimization method with these different datasets. The focus is on a high-aspect-ratio etch process across the different fabs and production stages. Models are built from process input variables to post-process metrology, and from process input variables to yield metrics. In the latter case, there can be tens of processes occurring between the model input and output variables. I demonstrate the usefulness and industrial application of neural network process modeling and genetic algorithm recipe optimization by performing a reaction chamber matching exercise on a manufacturing line. The performance of a reaction chamber can deviate from target, either in terms of its post-process metrology or its associated yield metrics. The method developed herein generated an optimized recipe that brought the outlying behavior of a chamber closer to target and closer to that of the other chambers (“chamber matching”). This is one of many possible applications. It was chosen because it demonstrates both the fidelity of the process models and the effectiveness of the optimization algorithm.

While hollow glass microspheres are commonly used in syntactic foam, their abrasive and brittle properties usually result in poor processability and have adverse effects on the foam performance. Therefore, a number of attempts have been made in the industry to replace hollow glass microspheres with polymeric foamed microspheres. Among many choices, expandable thermoplastic (ETP) microspheres filled syntactic foam has shown its high potential to become a novel class of engineering materials, especially for lightweight structural applications. However, conventional processing techniques for syntactic foam usually experience difficulties such as high processing viscosity, low loading of foam fillers, and ineffective microsphere expansion. To address these emerging issues, a microwave expansion process to produce thermoset-matrix syntactic foam containing thermoplastic foam beads was developed in this thesis work. In this process, unexpanded ETP microspheres were directly foamed in uncured thermoset matrix via microwave heating. Expandable polystyrene (EPS) microspheres and epoxy resin were chosen as a model material system. The resin viscosity and specific microwave energy are found to be the two primary control parameters determining the process window. Mechanical characterization showed that the syntactic foam can outweigh neat polymer in lightweight structural applications and was effectively toughened by foamed EPS. Furthermore, the microwave expansion process was found to be capable of molding syntactic foam parts of relatively sophisticated geometry with smooth surfaces. In order to broaden its impact, the microwave expansion process was extended to produce composite EPS foam. This process converts an expandable suspension into a composite foam with a honeycomb-like barrier structure. The suspension viscosity was found to highly influence the foam morphology. Results from mechanical tests showed that the existence of the barrier structure can considerably improve the mechanical performance of the composite foam. Fire-retardation tests demonstrated that the barrier structure can effectively stop the fire path into the foam, suppress toxic smoke generation, and maintain foam structure integrity. A general formulation was developed to model the EPS expansion to optimize the microwave expansion process. A semi-analytical solution was first obtained based on the case of a single bubble expansion in an infinite matrix. The dimensionless bubble radius and pressure are defined and found to be as exponential functions of dimensionless expansion time. The semi-analytical solution can qualitatively predict the radial expansion of EPS microsphere observed in a real-time experiment. To have an accurate prediction, a numerical solution was obtained to the model that couples the nucleation and expansion of multiple bubbles in a finite matrix. The results show that the numerical solution can quantitatively predict the radial expansion of EPS. A parameter sensitivity study was performed to examine the effect of each parameter over the expansion process.

This thesis considers the requirement engineering process from the small scale industries point of view. It begins with a brief introduction to software requirements and then proceeds to a detailed study of the requirement engineering process. The later part of the thesis considers Business Process Modeling and how it helps in understanding any business in a better way and how conceptual model helps in extracting business requirements from any particular business scenario. The Object Oriented Technique is used for building the conceptual model. The thesis concludes with a case study. / text

Molecular contaminants on the surface of nanoparticles (NPs) are critical in determining the environmental safety and health (ESH) impacts of NPs. In order to characterize the surface properties that relate to adsorption and desorption interactions, a method has been developed for studying the dynamic interactions of adsorbing species on NP samples. The results are analyzed using a process simulator to determine fundamental properties such as capacity, affinity, rate expressions, and activation energies of NP interactions with contaminants. The method is illustrated using moisture as a representative model compound and particles of SiO₂, HfO₂, and CeO₂, which are three oxides used in semiconductor manufacturing. The effect of particle size and temperature on the surface properties of porous oxide NPs was investigated. Infrared spectra peaks corresponding to the stretching vibration of water molecules were monitored by in-site Fourier transform infrared (FTIR) spectroscopy. These are related to the moisture concentration on the surface of NPs. A transient multilayer model was developed to represent the fundamental steps in the process. The thermal stability of adsorbed species and the strength of bonding to the surface were evaluated by determining the activation energies of the various steps. The results indicate that the surface interaction parameters are dependent on species, temperature, and particle size. SiO₂ has the highest adsorption capacity and therefore is most prone to the adsorption of moisture and similar contaminants. However, the affinity of the NPs for H₂O retention is highest for CeO₂ and lowest for SiO₂. As temperature decreases, NPs exhibit a higher saturated moisture concentration and are more prone to the adsorption of moisture and similar contaminants. Furthermore, smaller NPs have a higher saturated surface concentration and a slower response to purging and desorption. Factors contributing to the environmental and health impact of NPs (extent of surface coverage, capacity, and activation energy of retention) have been investigated during this study. The second objective of this study is to develop a method to measure and control the contamination in ultra-high-purity (UHP) gas delivery systems. Modern semiconductor manufacturing plants have very stringent specifications for the moisture content at the point-of-use, usually below several parts per billion (ppb). When the gas delivery system gets contaminated, a significant amount of purge time is required for recovery of the background system. Therefore, it is critical for high-volume semiconductor manufacturers to reduce purge gas usage as well as purge time during the dry-down process. A method consisting of experimental research and process simulations is used to compare steady-state purge (SSP) process of constant pressure and flow rate with the pressure-cycle purge (PCP) process of cyclic pressure and flow rate at a controlled interval. The results show that the PCP process has significant advantages over the SSP process under certain conditions. It can reduce the purge time and gas usage when the gas purity at point-of-use is the major concern. The process model is validated by data congruent with the experimental results under various operating conditions and is useful in conducting parametric studies and optimizing the purge process for industrial applications. The effect of key operational parameters, such as start time of PCP process as well as choice of PCP patterns has been studied.

Today, organizations often need to modify their business processes to cope with changes in the environment, such as mergers/acquisitions, new government regulations, and new customer demand. Most organizations also have a set of business policies defining the way they conduct their business. Although there has been extensive research on process analysis and design, how to systematically extract workflow models from business policies has not been studied, resulting in a missing link between the specification of business policies and the modeling of business processes.Given that process changes are often determined by executives and managers at the policy level, the aforementioned missing link often leads to inefficient and inaccurate implementation of process changes by business analysts and process designers. We refer to this problem as the policy mismatch problem in business process management. For organizations with large-scale business processes and a large number of business policies, solving the policy mismatch problem is very difficult and challenging.In this dissertation, we attempt to provide a formal link between business policies and workflow models by proposing a logic-based methodology for process analysis and design. In particular, we first propose a Policy-driven Process Design (PPD) methodology to formalize the procedure of extracting workflow models from business policies. In PPD, narrative process policies are parsed into precise information on various workflow components, and a set of process design rules and algorithms are applied to generate workflow models from that information.We also develop a logic-based process modeling language named Unified Predicate Language (UPL). UPL is able to represent all workflow components in a single logic format and provides analytical capability via logic inference and query. We demonstrate UPL's expressive power and analytical ability by applying it to process design and process change analysis. In particular, we use UPL to define and classify process change anomalies and develop algorithms to verify and enforce process consistency.The Policy-driven Process Design, Unified Predicate Language, and process change analysis approach found in this dissertation contribute to business process management research by providing a formal methodology for resolving the policy mismatch problem.

The purpose of this research was to determine if patient journey process modeling could act as a change management tool to support electronic health record (EHR) adoption, at a tertiary-care mental health centre. This research study was based on a pre/post design, which evaluated the attitudes of clinicians??? pre and post implementation of the EHR. A survey was used to assess the attitudes of various healthcare professionals, such as physicians, nurses and a spectrum of allied health disciplines, at various phases of the planning and implementation process. In addition to the surveys, current and future state PaJMa (patient journey modeling architecture) models representing technology use and process flows of all units were created by observational studies, and served as change management tools. These PaJMa models were then presented as part of an intervention that was held in the form of an educational session to highlight the benefits of technology, and to address the common concerns identified from the initial survey results. The centre for mental health sciences facility was used as the case study to apply the PaJMa model and assess its change management functionality. Since, the organization was moving from paper to electronic based patient charts it was an ideal choice for this research. It was predicted that the attitudes and opinions of clinicians towards the EHR implementation, and EHRs in general, would change and become more positive with increased knowledge and education. This in-turn would increase EHR adoption and hence lead to a successful implementation.

Two of the hottest areas in porous materials research in the last decade have been in energy storage, mainly hydrogen and methane, and in carbon capture and sequestration (CCS). Although these topics are intricately linked in terms of our future energy landscape, the specific materials needed to solve these problems must have significantly different properties. High pressure gas storage is most often linked with high surface areas and pore volumes, while carbon capture sorbents require high sorption enthalpies to achieve the needed selectivity. The latter typically involves separating CO2 from mixed gas streams of mostly nitrogen via a temperature swing adsorption (TSA) process. Much of the excitement has arisen because of the potential of metal-organic frameworks (MOFs) and porous polymer networks (PPNs). Both classes of materials have extremely high surface areas (upwards of 4000 m2/g) and can be modified to have specific physical properties, thus enabling high performance materials for targeted applications. This dissertation focuses on the synthesis and characterization of these novel materials for both applications by tuning framework topologies, composition, and surface properties. Specifically, two routes to synthesize a single molecule trap (SMT) highlight the flexibility of MOF design and ability to tune a framework to interact with specifically one guest molecule; computational and experimental evidence of the binding mechanism are shown as well. Furthermore, eight PPNs are synthesized and characterized for post-combustion carbon capture and direct air capture applications. In addition a high-throughput model, grounded in thermodynamics, to calculate the energy penalty associated with the carbon capture step is presented in order to evaluate all materials for TSA applications provide a comparison to the state of the art capture technologies. This includes results of working capacity and energy calculations to determine parasitic loads (per ton of CO2 captured) from readily available experimental data of any material (adsorption isotherms and heat capacities) using a few simple equations. Through various systematic investigations, trends are analyzed to form structure property relationships that will aid future material development.

Das Gebiet der Serviceentwicklung im Rahmen serviceorientierter Architekturen, als auch der Bereich der Geschäfts­prozessmodellierung gewannen in den letzten Jahren zunehmend an Bedeutung. Aus fachlicher Sicht stellt die Analyse, Simulation und Optimierung aktueller und zukünftiger Geschäftsprozesse ein wertvolles Instrument zur Erkennung von Schwachstellen und der Verbesserung der Leistungsfähigkeit von Unternehmen dar. Aus IT-Sicht bietet die Service­orientierung die Möglichkeit, Unternehmensressourcen über öffentliche und private Schnittstellen verfügbar, und damit die Position des Unternehmens im Netzwerk von Lieferanten, Kunden und Geschäfts­partnern zu stärken. Den Kern dieser Arbeit bildet die Untersuchung des Spannungsfeldes zwischen fachlicher Geschäfts­­­prozess- und Servicemodellierung, sowie der Ausführung von Geschäftsprozessen auf techni­scher Ebene, die durch Services unterstützt werden. Im Fokus der Betrachtung liegt der unter­nehmens- bzw. organisations­übergreifende Bereich, insbesondere der Finanzdienst­leistungs­sektor. Zu Beginn der Arbeit wird ein allgemeines Begriffsverständnis zu Bereichen Service- und Prozess­modellierung geschaffen, die im Hauptteil der Arbeit verfeinert und ergänzt werden, um die Beson­der­heiten und Problemfelder für den überbetrieblichen Bereich verständlich zu machen. Dabei wird das RosettaNet-Frame­work als Rahmenwerk zur Gestaltung von elektronischen Geschäfts­prozessen beschrieben und im letzten Teil der Arbeit zur Implementierung eines konkreten Geschäfts­­prozesses im Vertriebsprozess heran­gezogen. Zu diesem Zweck wird der ORACLE WebLogic Application Server verwendet.

Orientador: Renato de Campos / Banca: Wilson Massashiro Yonezawa / Banca: José Alcides Gobbo Junior / Resumo: O entendimento dos processos de Modelo de Referência de Operações da Cadeia de Suprimentos (ou Supply Chain Reference - SCOR model) pode ser facilitado pelo uso de uma metodologia de modelagem de processos. O emprego de uma notação com semântica adequada para representar esses modelos de processos, associadas a ferramentas de modelagem e gestão, podem contribuir para a melhor compreensão e o gerenciamento de uma cadeia de suprimentos. Existem várias notações propostas na literatura, como por exemplo, a Notação para Modelagem de Processos de Negócios (ou Bussiness Process Modeling Notation - BPMN), a Cadeia de Processos Dirigida por Eventos (ou Event-Driven Process Chain - EPC) e a Linguagem de Modelagem Unificada (ou Unified Modeling Language - UML). Também podem ser identificadas algumas ferramentas de suporte computacional para a modelagem dos processos, como por exemplo, o Bonita Studio, o ARIS Express e o Intalio Designer. Este trabalho analisou e comparou algumas notações e ferramentas de suporte computacional para melhor representar os modelos de processos SCOR. Definidos as notações, ferramentas e critérios de comparação, as análises indicaram a notação BPMN e a ferramenta Intalio como sendo as mais adequadas. Como resultado final, a notação e a ferramenta foram utilizadas para a geração de uma biblioteca de modelos a partir de processos padrão de terceiro nível do modelo SCOR. Exemplos de desenvolvimento de modelos são apresentados, visando demonstrar a adequação da notação e ferramenta, e exemplificar como a biblioteca de modelos pode ser usada visando o desenvolvimento de modelos de quarto nível, a partir da biblioteca de modelos, por possível usuários. Como trabalho futuro, propõe-se a geração de modelos de referência de software com base nos modelos SCOR / Abstract: The understanding of the process of Supply Chain Operation Reference - SCOR model can be facilitated by use of a methodology for process modeling. The use of a notation with semantics appropriate for representing these types of processes associated with the modeling and management tools, can contribute to better understanding and management of a supply chain. There are several notations proposed in the literature, for exemple, the notation for Business Modeling Notation - BPMN, the Event-Driven Process Chain - EPC and Unified Modeling Language - UML). Can also be identified some tools for computational modeling of processes, such as the Bonita Studio, ARIS Express and Intalio Designer. This study analyzed and compared some notations and support tools to best represent the computational process models SCOR. Defined notations, tools and benchmarks, the analysis indicated the Intalio BPMN notation and the most appropriate. As a final result, the notation and the tool was used to generate a library of models from standard processes of the third level of the SCOR model. Examples of model development are presented to demonstrate the adequacy of the notation and tool, ans illustrate how the model library can be used for the develpment of models of the fourth level, from the library of models for potential users. As future work, we propose the generation of role models for developing software systems based on the SCOR model / Mestre

A Notação e Modelo de Processo de Negócio (BPMN - Business Process Model and Notation) possibilita apresentar informações sobre o fluxo de dados de um processo por meio de dados e artefatos, tais como objetos, associações e repositórios de dados. No entanto, esses elementos apresentam poder de expressão limitado, principalmente quando se referem ao mapeamento das fontes de dados (ex.: serviço web) em um modelo de processo. Neste contexto, o presente trabalho propõe uma abordagem para associar fontes de dados a elementos de BPMN (ex.: tarefa de serviço). Tal abordagem deve servir como um ponto de partida para profissionais de gerenciamento de processos de negócio quanto ao mapeamento, na etapa de modelagem, da fonte de dados que está sendo utilizada pelo processo, através da utilização de elementos de BPMN. Visando demonstrar os resultados, foram evidenciadas cinco correlações, que são chamadas de “propostas de definições”, seguidas por sua descrição textual e um exemplo de uso. Essas propostas de definições, juntamente com as fontes de dados utilizadas, foram avaliadas por meio de uma pesquisa de opinião. Como resultado, destaca-se que as fontes de dados selecionadas foram validadas pelos participantes da pesquisa de opinião, que demonstraram conhecimento em ao menos uma das fontes de dados, fornecendo indícios que as fontes de dados selecionadas são as mais apropriadas para o trabalho em questão. Ainda, como resultados, as propostas de definições foram validadas com os participantes, obtendo resultados positivos quanto aos fragmentos de processo elaborados para representar as fontes de dados internos, transferência de arquivos e base de dados compartilhada. A principal contribuição deste trabalho é auxiliar na identificação de fontes de dados de um processo de negócio, a partir do conjunto de elementos disponibilizado pela BPMN 2.0.2, onde, anteriormente, tal identificação era possível apenas através da documentação do processo. / Business Process Model and Notation (BPMN) makes it possible to display information about the data flow of a process through data and artifacts such as objects, associations and data stores. However, these elements provide a limited capacity, especially when referring to the mapping of the data sources (e.g. web service) in a process model. In this context, this paper proposes an approach to associate data sources to BPMN elements (e.g. service task). Such approach aims to serve as a starting point to business process management professionals in order to map, in design time, the data source that is being used by the process, through the use of BPMN elements. In order to demonstrate the results, five correlations were evidenced, which are called “definitions proposals”, followed by their textual description and an example of use. These definitions proposals, together with the data sources used, were evaluated through a survey. As a result, the selected data sources were validated by the participants of the survey that demonstrated knowledge in at least one of the data sources, providing evidence that the selected data sources are the most appropriated for the work concerned. Also, as results, the definitions proposals were validated with the participants, obtaining positive results regarding to the process fragments designed to represent the data sources internal data, file transfer and shared database. The main contribution of this thesis is to assist the identification of data sources of a business process, from the set of elements provided by BPMN 2.0.2, where, previously, such identification was only possible through the process documentation.