The main objective of this dissertation is to assist in reducing the chaotic state of the software engineering discipline by providing insights into both the effectiveness of software defect reduction methods and ways these methods can be improved. The dissertation is divided into two main parts. The first is a quasi-experiment comparing the software defect rates and initial development costs of two methods of software defect reduction: software inspection and test-driven development (TDD). Participants, consisting of computer science students at the University of Arizona, were divided into four treatment groups and were asked to complete the same programming assignment using either TDD, software inspection, both, or neither. Resulting defect counts and initial development costs were compared across groups. The study found that software inspection is more effective than TDD at reducing defects, but that it also has a higher initial cost of development. The study establishes the existence of a defect-reduction gap between software inspection and TDD and highlights the need to improve TDD because of its other benefits.The second part of the dissertation explores a method of applying mutation analysis to TDD to reduce the defect reduction gap between the two methods and to make TDD more reliable and predictable. A new change impact analysis algorithm (CHA-AS) based on CHA is presented and evaluated for applications of software change impact analysis where a predetermined set of program entry points is not available or is not known. An estimated average case complexity analysis indicates that the algorithm's time and space complexity is linear in the size of the program under analysis, and a simulation experiment indicates that the algorithm can capitalize on the iterative nature of TDD to produce a cost savings in mutation analysis applied to TDD projects. The algorithm should also be useful for other change impact analysis situations with undefined program entry points such as code library and framework development.An enhanced TDD method is proposed that incorporates mutation analysis, and a set of future research directions are proposed for developing tools to support mutation analysis enhanced TDD and to continue to improve the TDD method.

Com o crescimento da demanda por aplicações orientadas a serviços, a complexidade da gestão de mudanças em serviços aumenta cada vez mais. Nesse contexto, provedores de portfólios de serviços têm como objetivo a minimização do impacto de tais mudanças para os seus clientes. Os trabalhos existentes abordam essencialmente a mudança sob o viés técnico, lidando com técnicas de versionamento e determinação de compatibilidade. No entanto, os provedores possuem dilemas de decisão que vão além dos aspectos técnicos, abrangendo o impacto na perspectiva de negócios, considerando clientes afetados, receitas, custos, violações de SLA, entre outros. Este trabalho faz parte de um framework voltado para a análise da evolução de serviços considerando a perspectiva de uso dos mesmos. O presente trabalho propõe o uso de técnicas de Business Intelligence e Data Warehousing para apoiar a tomada de decisões orientadas a negócios na gestão do ciclo de vida de serviços. Em tal contexto, considera-se o cenário de portfólio de serviços e mudanças profundas, abrangendo cadeias de serviços e clientes diretos/indiretos. Este trabalho identifica e analisa requisitos de decisão, juntamente com indicadores financeiros e de uso relacionados à provisão de serviços. Um data warehouse é modelado para prover uma visão unificada e integrada de tais indicadores, de acordo com diferentes perspectivas de análise. Uma arquitetura para a extração, transformação e carga dos dados também é apresentada com a proposição de alternativas para lidar com a integração e heterogeneidade das fontes de dados. Através de um estudo de caso inspirado em um cenário real, demonstra-se como a análise do impacto é analisada através da abordagem proposta. / With the growing demand for service-oriented applications, the complexity of service change management is increasing. In this context, service providers aim at the minimal impact of changes to its clients. Existing work essentially addresses change decisions from a technical perspective, with versioning techniques and compatibility assessments. However, providers have decisional dilemmas that go beyond the technical aspects, addressing the business impact in terms of clients affected, revenues, costs, SLA penalties, among others. This work integrates a framework addressing the service evolution analysis based on service usage perspective. We propose the use of Business Intelligence and Data Warehousing techniques to support business-oriented decisions on the service life-cycle management. Our approach covers service portfolios and the deep change context, i.e. services consumed in large scale and direct/indirect clients. This work is centered on the identification and analysis of decisional requirements and financial and usage indicators related to the service provision business. A data warehouse is modeled to provide a unified and integrated view of these indicators, according to different analysis perspectives. An architecture to extract, transform and loading data also is presented, proposing alternatives to lead with the integration and the heterogeneity of the data sources. Through a case study inspired by a real world scenario, we demonstrate how change impact is analyzed according our approach.

The world is urbanizing at an unprecedented rate, and cities are dominantly and increasingly becoming hubs for agglomerations of human population and economic activities, as well as major sources of environmental problems. Accordingly, humanity′s pursuit of global sustainability is becoming increasingly reliant on urban sustainability. Unfortunately, the traditional approaches of urbanization and urban stormwater management are inappropriate from the sustainability standpoint. By removing vegetation and topsoil and creating impervious structures, urbanization destroys natural biodiversity and hydrological processes. As a result, urban societies are disconnected from nature and deprived of ecosystem services including flood control, fresh air, clean water, and natural beauty. Due to disrupted hydrology, an urban landscape transforms most rainwater into stormwater runoff which is conveyed off the site through a system of curb-gutter-pipe, called gray infrastructure. While gray infrastructure efficiently mitigates the problem of flash floods in urban areas, it results in multiple other adverse environmental consequences such as loss of freshwater from urban landscapes, transfer of pollutants to receiving waters, and an increased potential of downstream flooding. Green infrastructure (GI) is regarded as a sound alternative that manages stormwater by revitalizing the natural processes of soil, water, and vegetation, and restoring ecosystem structures and functions. Thus, the approach re–establishes the lost socio–ecological connectivity and regenerates ecosystem services. However, despite being inevitably important for urban sustainability, and despite being the object of unrelenting expert advocacy for more than two decades, the approach is yet to become a mainstream practice. To widely implement GI, cities need to address two critical challenges. First, urban stormwater managers and decision makers should be ensured that the approach can adequately and reliably manage stormwater. In the time when flooding problems are rising due to climate change, this concern has become more prominent. Second, if there exist any other barriers, they should be replaced with strategies that help expedite the use of GI. This multidisciplinary research dealt with these two challenges. The study consisted of two major parts. In the first part, a computer model was developed for a combined sewer system of St. Louis, a city in the US state of Missouri, using U.S. EPA SWMM. Simulations for historical (1971-2000) and future (2041-2070) 50-yr 3-hr rainfall scenarios were then run on the model with and without GI. The simulation results showed a significant impact of increased precipitation on the system, which was considerably reduced after adding select GI measures to the modeled system. The following 4 types of GI were used: bio–retention cell, permeable pavement, green roof, and rain barrel. In the second part, a survey of relevant policies and governance mechanisms of eleven U.S. cities was conducted to identify potential barriers to GI and determine strategies to address them. The study also included the assessment of relevant city, state, and federal policies and governance structures. A total of 29 barriers were identified, which were grouped into 5 categories. Most of the identified barriers stem from cognitive barriers and socio–institutional arrangements. A total of 33 policies, also grouped into 5 groups, were determined to address the barriers. The investigation on governance revealed that current governance is highly technocratic and centralized, and hence has less opportunity for public involvement. Therefore, it is inherently inappropriate for GI, which requires extensive public involvement. This dissertation proposes a two–tier governance model suitable for implementing GI.

Com o crescimento da demanda por aplicações orientadas a serviços, a complexidade da gestão de mudanças em serviços aumenta cada vez mais. Nesse contexto, provedores de portfólios de serviços têm como objetivo a minimização do impacto de tais mudanças para os seus clientes. Os trabalhos existentes abordam essencialmente a mudança sob o viés técnico, lidando com técnicas de versionamento e determinação de compatibilidade. No entanto, os provedores possuem dilemas de decisão que vão além dos aspectos técnicos, abrangendo o impacto na perspectiva de negócios, considerando clientes afetados, receitas, custos, violações de SLA, entre outros. Este trabalho faz parte de um framework voltado para a análise da evolução de serviços considerando a perspectiva de uso dos mesmos. O presente trabalho propõe o uso de técnicas de Business Intelligence e Data Warehousing para apoiar a tomada de decisões orientadas a negócios na gestão do ciclo de vida de serviços. Em tal contexto, considera-se o cenário de portfólio de serviços e mudanças profundas, abrangendo cadeias de serviços e clientes diretos/indiretos. Este trabalho identifica e analisa requisitos de decisão, juntamente com indicadores financeiros e de uso relacionados à provisão de serviços. Um data warehouse é modelado para prover uma visão unificada e integrada de tais indicadores, de acordo com diferentes perspectivas de análise. Uma arquitetura para a extração, transformação e carga dos dados também é apresentada com a proposição de alternativas para lidar com a integração e heterogeneidade das fontes de dados. Através de um estudo de caso inspirado em um cenário real, demonstra-se como a análise do impacto é analisada através da abordagem proposta. / With the growing demand for service-oriented applications, the complexity of service change management is increasing. In this context, service providers aim at the minimal impact of changes to its clients. Existing work essentially addresses change decisions from a technical perspective, with versioning techniques and compatibility assessments. However, providers have decisional dilemmas that go beyond the technical aspects, addressing the business impact in terms of clients affected, revenues, costs, SLA penalties, among others. This work integrates a framework addressing the service evolution analysis based on service usage perspective. We propose the use of Business Intelligence and Data Warehousing techniques to support business-oriented decisions on the service life-cycle management. Our approach covers service portfolios and the deep change context, i.e. services consumed in large scale and direct/indirect clients. This work is centered on the identification and analysis of decisional requirements and financial and usage indicators related to the service provision business. A data warehouse is modeled to provide a unified and integrated view of these indicators, according to different analysis perspectives. An architecture to extract, transform and loading data also is presented, proposing alternatives to lead with the integration and the heterogeneity of the data sources. Through a case study inspired by a real world scenario, we demonstrate how change impact is analyzed according our approach.

Com o crescimento da demanda por aplicações orientadas a serviços, a complexidade da gestão de mudanças em serviços aumenta cada vez mais. Nesse contexto, provedores de portfólios de serviços têm como objetivo a minimização do impacto de tais mudanças para os seus clientes. Os trabalhos existentes abordam essencialmente a mudança sob o viés técnico, lidando com técnicas de versionamento e determinação de compatibilidade. No entanto, os provedores possuem dilemas de decisão que vão além dos aspectos técnicos, abrangendo o impacto na perspectiva de negócios, considerando clientes afetados, receitas, custos, violações de SLA, entre outros. Este trabalho faz parte de um framework voltado para a análise da evolução de serviços considerando a perspectiva de uso dos mesmos. O presente trabalho propõe o uso de técnicas de Business Intelligence e Data Warehousing para apoiar a tomada de decisões orientadas a negócios na gestão do ciclo de vida de serviços. Em tal contexto, considera-se o cenário de portfólio de serviços e mudanças profundas, abrangendo cadeias de serviços e clientes diretos/indiretos. Este trabalho identifica e analisa requisitos de decisão, juntamente com indicadores financeiros e de uso relacionados à provisão de serviços. Um data warehouse é modelado para prover uma visão unificada e integrada de tais indicadores, de acordo com diferentes perspectivas de análise. Uma arquitetura para a extração, transformação e carga dos dados também é apresentada com a proposição de alternativas para lidar com a integração e heterogeneidade das fontes de dados. Através de um estudo de caso inspirado em um cenário real, demonstra-se como a análise do impacto é analisada através da abordagem proposta. / With the growing demand for service-oriented applications, the complexity of service change management is increasing. In this context, service providers aim at the minimal impact of changes to its clients. Existing work essentially addresses change decisions from a technical perspective, with versioning techniques and compatibility assessments. However, providers have decisional dilemmas that go beyond the technical aspects, addressing the business impact in terms of clients affected, revenues, costs, SLA penalties, among others. This work integrates a framework addressing the service evolution analysis based on service usage perspective. We propose the use of Business Intelligence and Data Warehousing techniques to support business-oriented decisions on the service life-cycle management. Our approach covers service portfolios and the deep change context, i.e. services consumed in large scale and direct/indirect clients. This work is centered on the identification and analysis of decisional requirements and financial and usage indicators related to the service provision business. A data warehouse is modeled to provide a unified and integrated view of these indicators, according to different analysis perspectives. An architecture to extract, transform and loading data also is presented, proposing alternatives to lead with the integration and the heterogeneity of the data sources. Through a case study inspired by a real world scenario, we demonstrate how change impact is analyzed according our approach.

Submitted by Fabio Sobreira Campos da Costa (fabio.sobreira@ufpe.br) on 2016-07-01T12:38:10Z No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Dissertação-JoelsonIsidroDaSilvaAraújo.pdf: 1742206 bytes, checksum: c2b6f413666e50be58b963e8624a1386 (MD5) / Made available in DSpace on 2016-07-01T12:38:10Z (GMT). No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Dissertação-JoelsonIsidroDaSilvaAraújo.pdf: 1742206 bytes, checksum: c2b6f413666e50be58b963e8624a1386 (MD5) Previous issue date: 2015-09-04 / CAPEs / Contexto: a mudança faz parte da evolução e durante o ciclo de vida do software a maior parte dos custos está associada a esta tarefa. Poder fazer previsões sobre os potenciais efeitos causados através de uma mudança é uma forma de minimizar esses custos. Neste contexto, surge então a Análise de Impacto (AI) para medir o esforço que será necessário à mudança e para nortear como realizar a mesma da maneira mais adequada, entretanto o resultado gerado pode ser insuficiente, pois é possível existirem erros na identificação dos elementos possivelmente impactados, não contemplando todos os problemas existentes. Objetivo: este trabalho tem como objetivo investigar o que se tem feito para permitir um resultado mais preciso na AI, gerando um catálogo de benefícios e limitações e propondo um guia de boas práticas, respondendo as perguntas de pesquisa – O que se sabe atualmente sobre os benefícios e limitações da AI em mudança de software? O que se tem feito para minimizar os erros gerados na análise? Método: para a condução da pesquisa fez-se necessário a busca de dados na literatura, através de uma pesquisa exploratória, por meio de uma revisão sistemática com o intuito de investigar as técnicas de AI relatadas em pesquisas dos últimos anos. Resultados: de posse dos dados resultantes da extração e análise dos dados, os resultados são: (1) evidências de técnicas existentes que conseguiram minimizar imprecisões nos resultados da análise, (2) geração de catálogo de benefícios e limitações em seu uso e ainda, um guia de propostas de boas práticas a serem adotadas para permitir que a análise apresente melhores resultados. Conclusão: os resultados fornecem uma melhor visão dos fatores que precisam ser melhorados e, além disso, possibilitaram a criação de um guia de boas práticas. Com isto, pretendemos contribuir fornecendo uma melhor compreensão sobre as técnicas existentes, de que forma melhorias vêm sendo propostas e quais práticas permitem a maximização dos resultados gerados através da análise de impacto. / Context: Changing is part of the evolution and during the software lifecycle most cost is associated with this task. Being able to make predictions about the potential effects caused by a change is a way to minimize these costs. In this context, the Impact Analisys (IA) can be used to measure the effort it will take to change and to guide how to do the same in the most appropriate way, however the results generated may be insufficient, it is possible to detect errors on the elements identification possibly impacted, not including all the existing problems. Objective: This study has the objetive to investigate what has been done to allow more accurate result in IA , generating a catalog of benefits and limitations and proposing a guide of good practice by answering the research questions - What is currently known about the benefits and IA limitations on software changes? What has been done to minimize errors generated in the analysis? Methodology: To conduct this research it is necessary to search data in the literature, through an exploratory research using a systematic review that will allow an investigation about the most IA techniques used in the last years. Results: With the data generated through the extration and analisys of data, the results are: evidences of techniques which can be used to minimize inaccuracies in test results, (2) generation of catalog of benefits and limitations in its use and also a good practice guide to be adopted to allow the analysis present better results. Conclusion: the expected results will provide a better view of the factors that need to be improved and, besides, will enable the creation of a good practice guide. With this, we intend to contribute by providing a better understanding of existing techniques, how improvements have been proposed and what practices has been used to improve the results generated by impact analysis.

Model Driven Development (MDD) is considered as a powerful emerging paradigm for developing system software and services. MDD relies heavily on the models to generate part of the code for implementation. It is virtually inevitable that changes at the model level result in breaking the compatibility with the existing code base. The study presented in this thesis is performed in an industrial setting. Our industrial partner is involved in MDD and has experienced compatibility problems described above. The aim of this thesis is to identify the changes that can be made to UML class diagram and investigate what kind of negative impact they pose on the compatibility with already existing Java implementations. Apart from identifying possible model changes and assessing their negative impact the thesis provides suggestions and guidelines of how to perform such modifications so that their negative impact will be minimized. / +46 455-395070

Software development is prone to software faults due to the involvement of multiple stakeholders especially during the fuzzy phases (requirements and design). Software inspections are commonly used in industry to detect and fix problems in requirements and design artifacts, thereby mitigating the fault propagation to later phases where the same faults are harder to find and fix. The output of an inspection process is list of faults that are present in software requirements specification document (SRS). The artifact author must manually read through the reviews and differentiate between true-faults and false-positives before fixing the faults. The first goal of this research is to automate the detection of useful vs. non-useful reviews. Next, post-inspection, requirements author has to manually extract key problematic topics from useful reviews that can be mapped to individual requirements in an SRS to identify fault-prone requirements. The second goal of this research is to automate this mapping by employing Key phrase extraction (KPE) algorithms and semantic analysis (SA) approaches to identify fault-prone requirements. During fault-fixations, the author has to manually verify the requirements that could have been impacted by a fix. The third goal of my research is to assist the authors post-inspection to handle change impact analysis (CIA) during fault fixation using NL processing with semantic analysis and mining solutions from graph theory. The selection of quality inspectors during inspections is pertinent to be able to carry out post-inspection tasks accurately. The fourth goal of this research is to identify skilled inspectors using various classification and feature selection approaches. The dissertation has led to the development of automated solution that can identify useful reviews, help identify skilled inspectors, extract most prominent topics/keyphrases from fault logs; and help RE author during the fault-fixation post inspection.

There is now a general consensus among scientists on occurrences of more and intense climatic disasters, floods and droughts, everywhere in the future. To act sooner and smarter against these negative impacts, we must shift our focus in better understanding of the future climate change and possible implications of that to better manage our water resources. Certainly, there is a link between the future effects of climate change on water resources and trends of climatic variables. In this study, by using Mann-Kendall trend analysis method, it is concluded that from 1961 to 2010 only temperature has an upward trend, in all the seasons and yearly, in all the 16 studied stations from north, middle and south of Sweden. Furthermore, runoff simulated by HBV model shows increasing trend in summer and winter which is in partially agreement with the recorded one that discerns a growth in the mentioned seasons, as well as yearly. What is more, potential evapotranspiration estimated by FAO Penman-Monteith equation and actual one simulated by CoupModel reveal a rise in spring for the former and both spring and winter for the latter. Other meteorological variables do not show any significant trend, while intensive precipitation increased in winter and summer in the majority of the stations. Comparing the runoffs simulated by CoupModel and HBV model shows that HBV works better for three selected stations which can enforce the claim that HBV works better for smaller fields.

京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第17876号 / 工博第3785号 / 新制||工||1579(附属図書館) / 30696 / 京都大学大学院工学研究科都市社会工学専攻 / (主査)教授 角 哲也, 教授 堀 智晴, 准教授 田中 賢治, 准教授 Sameh Ahmed Kantoush / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DGAM