The value creating mechanism of business networks has been acknowledged in research (Casson & Della Giusta, 2007, p. 221; Granovetter, 1973; Jones, Hesterly & Borgatti, 1997, p. 912). Within small business research, focus has been on network structure (Jack, Moult, Anderson & Dodd. 2010, p. 317; Shaw, 2006, p. 5; Curran et al., 1993, p. 13), disregarding dynamic factors such as network content, strength and processes. Network studies within start-up businesses have received attention (Witt, 2004, p 392); hence limited research exists concerning more well-established small businesses’ networks. These research gaps will here be addressed from the perspective of a smaller business, a micro business, at a stage of firm maturity. Small business research suffers from methodological fragmentation and underdevelopments, favoring positivistic and quantitative methods (Blackburn & Kovalainen, 2009, p. 128-129). To address this, this research seeks to explore qualitative alternatives, which can be increasingly fruitful for the small business field when studying networks (Jack et al., 2010, p. 317). Recognizing the social and contextual embeddedness of the micro business and its sensitivity to external changes (Curran & Blackburn, 2001, p. 7), this research opts for a qualitative in-depth case study to explore these contextual mechanisms. This research adopts a critical realist stance using ethnographic and process research methods of grounded theory, taking the sequence of events into account in the analysis. This present research combines strands of research of core disciplines (sociology) to newer disciplines (entrepreneurship) to produce a conceptual mapping. This deductive conceptual mapping is used as basis for the data collection and as a tool for analysis in this by nature inductive study. This study finds that network tie content and strength deepen the understanding of networks and their maintenance, beyond the static mapping of structure. Network tie content and strength does influence structure and is in turn influenced by context. This provides answer to why the network appears as it does. The appearance of the mature micro business’ network indicates that growth ambitions can be inhibited by contextual factors such as limited access to new opportunities. This present study finds and agrees with Van Maanen (2011a, p. 226-227) that the single case study and ethnographic observant-participant methods, often criticized for not producing substantial findings, are informative as they allow exploring the influence of contextual factors. The intended audience of research academics and local/regional/national entrepreneurship policy makers should recognize that research which enhances tools for apprehending contextual factors can provide interesting results. The research design disregards generalization capabilities in favor of accounting for the research process of the researcher and its subject. This is argued to give a fruitful version of truth of network appearance in terms of content and strength and why it appears as such.

Made available in DSpace on 2014-12-17T14:57:55Z (GMT). No. of bitstreams: 1 AdrianoSB_TESE.pdf: 4047692 bytes, checksum: 5b0ec3af1fa350ee85a33e2b05f88fc7 (MD5) Previous issue date: 2012-11-30 / The static and cyclic assays are common to test materials in structures.. For cycling assays to assess the fatigue behavior of the material and thereby obtain the S-N curves and these are used to construct the diagrams of living constant. However, these diagrams, when constructed with small amounts of S-N curves underestimate or overestimate the actual behavior of the composite, there is increasing need for more testing to obtain more accurate results. Therewith, , a way of reducing costs is the statistical analysis of the fatigue behavior. The aim of this research was evaluate the probabilistic fatigue behavior of composite materials. The research was conducted in three parts. The first part consists of associating the equation of probability Weilbull equations commonly used in modeling of composite materials S-N curve, namely the exponential equation and power law and their generalizations. The second part was used the results obtained by the equation which best represents the S-N curves of probability and trained a network to the modular 5% failure. In the third part, we carried out a comparative study of the results obtained using the nonlinear model by parts (PNL) with the results of a modular network architecture (MN) in the analysis of fatigue behavior. For this we used a database of ten materials obtained from the literature to assess the ability of generalization of the modular network as well as its robustness. From the results it was found that the power law of probability generalized probabilistic behavior better represents the fatigue and composites that although the generalization ability of the MN that was not robust training with 5% failure rate, but for values mean the MN showed more accurate results than the PNL model / Quando da utiliza??o de materiais em estruturas ? comum ? realiza??o de ensaios est?ticos e c?clicos. Para os ensaios c?clicos avalia-se o comportamento ? fadiga do material e com isso obt?m-se as curvas S-N e estas s?o utilizadas para construir os diagramas de vida constante. Por?m, estes diagramas, quando constru?dos com pequenas quantidades de curvas S-N, subestimam ou sobrestimam o comportamento real do comp?sito, havendo necessidade crescente de se fazer mais ensaios para a obten??o de maior precis?o nos resultados. Pensando assim, uma forma de diminuir custos ? a an?lise estat?stica do comportamento ? fadiga. Assim, este trabalho possui o intuito de avaliar o comportamento probabil?stico ? fadiga de materiais comp?sitos e foi dividido em tr?s partes. A primeira parte consiste em como associar a equa??o de probabilidade de Weilbull ?s equa??es comumente utilizadas no modelamento da curva S-N de mat?rias comp?sitos, quais sejam, a equa??o exponencial e a lei de pot?ncia e suas respectivas generaliza??es. Na segunda parte utilizou-se os resultados obtidos pela equa??o que melhor representa as curvas S-N de probabilidade e treinou-se uma rede modular ? 5% de falha. Na terceira parte, realizou-se um estudo comparativo dos resultados obtidos usando o modelo n?o linear por partes (PNL) com os resultados de uma arquitetura de rede modular (MN) na an?lise do comportamento ? fadiga. Para tanto utilizouse uma base de dados com dez materiais obtidos da literatura para se avaliar a capacidade de generaliza??o da rede modular, bem como sua robustez. A partir dos resultados verificou-se que a lei de pot?ncia generalizada de probabilidade representa melhor o comportamento probabil?stico ? fadiga de comp?sitos e que apesar da capacidade de generaliza??o da MN esta n?o se mostrou robusta para um treinamento com 5% de falha, por?m para valores m?dios a MN apresentou resultados mais precisos que o modelo PNL