Dauguma programinę įrangą kuriančių įmonių susiduria su problemomis: projektai vėluoja, viršijamas biudžetas, klientai nepatenkinti produktų kokybe. Daugelis problemų kyla dėl nebrandaus programų kūrimo proceso, pagal kurį įmonė įgyvendina projektus. Populiariausi pasaulyje programų kūrimo procesų vertinimo modeliai yra ISO/IEC 15504 ir CMMI. Įmonės, norėdamos būti oficialiai pripažintos platesnėje aplinkoje, renkasi vieną iš šių dviejų modelių. Įmonės susiduria su problema, kad skirtingi užsakovai reikalauja skirtingų procesų vertinimo modelių. Todėl įmonėms yra aktualu turėti instrumentą, kuris atvaizduotų jos vertinimo rezultatus iš vieno procesų vertinimo modelio į kitą, neatliekant kaskart realaus įmonės procesų vertinimo. Disertacijoje pateiktas tarpinis programų kūrimo procesų vertinimo modelis yra instrumentas įmonėms, siekiančioms kurti programų sistemas automobilių, krašto apsaugos, kosmoso, medicinos ir kitose pramonės šakose, gauti vertinimo rezultatus pagal įvairius procesų vertinimo modelius, atliekant vieną procesų vertinimą. Pagal pasiūlytą tarpinio programų kūrimo procesų vertinimo modelio kūrimo metodą galima kurti tarpinius modelius ir kitų sričių procesams vertinti. Pavyzdžiui, galima konstruoti tarpinį modelį paslaugas teikiančioms organizacijoms, apimantį CMMI-SVC ir ISO/IEC 15504-8 vertinimo modelius, arba tarpinį modelį bet kokia veikla užsiimančioms organizacijoms, apimantį Enterprise SPICE ir FAA-iCMM vertinimo modelius. / Many software companies face such problems as projects being behind schedule, exceeding the budget, customer dissatisfaction with product quality. Most of the problems arise due to immature software process of the company. The most popular process assessment models worldwide are ISO/IEC 15504 and CMMI. Companies seeking wider official recognition choose between these two models. Companies face the problem that different customers require process assessment according to different models. Therefore, it is important for companies to have an instrument that would transform company’s assessment results according to one process assessment model to another model without actual reassessment each time. The transitional process assessment model proposed in the dissertation allows the companies in the automotive, defence, space, medical, and other industries to have assessment results according to different process assessment models by performing only single process assessment. The presented method for TPAM development could be used developing transitional process assessment models in other domains. For example, service companies could implement a transitional model covering CMMI-SVC and ISO/IEC 15504-8 assessment models, and companies working in other areas could implement a transitional model covering Enterprise SPICE and FAA-iCMM assessment models.

A avaliação e melhoria de um processo de software requer um grande esforço por parte de gerentes e desenvolvedores, uma vez que a obtenção de dados para tomada de decisão, acerca da evolução do processo nem sempre é trivial. O desenvolvimento de software, como qualquer disciplina de Engenharia, necessita de mecanismos de mensuração, para que através de uma avaliação de seus resultados o processo possa ser melhorado. Mas a mensuração pode ser uma tarefa bastante trabalhosa sem o auxílio de metodologias ou ferramentas de apoio. Neste sentido, este trabalho apresenta um modelo para mensuração em processos de software, no contexto do projeto APSEE-Prosoft. O objetivo do modelo proposto é prover suporte a realização de mensuração em processos de software, através do desenvolvimento e implantação de programas de mensuração, visando auxiliar aos gerentes na realização de atividades como planejamento, coleta e análise de métricas em um processo de software. Neste trabalho são combinados aspectos relacionados a importantes áreas da Engenharia de Software, como mensuração, processos de software, ambientes de engenharia de software centrados em processos (PSEE), bem como a utilização de métodos formais para a definição e especificação de um modelo de mensuração integrado a um PSEE. / The software process evaluation and improvement requests great effort of managers and developers, especially because obtaining data for decision support (about process evolution) is not a simple task. The software development, as any discipline of Engineering, needs measurement mechanisms, so that through an evaluation of its results, the process can be improved. But measurement can be a difficult task without the aid of methodologies or support tools. In this sense, this paper presents a model for software processes measurement, on the context of APSEE-Prosoft project. The model's objective is to provide support for the measurement in software processes, through the development and implantation of measurement programs, to aid managers in the accomplishment of activities such as planning, collecting and analysis of metrics in a software process. This paper combines aspects related to important areas of Software Engineering, as measurement, software process, process-centered engineering environment (PSEE), besides the use of formal methods for a definition and specification of a measurement model integrated to a PSEE.

A avaliação e melhoria de um processo de software requer um grande esforço por parte de gerentes e desenvolvedores, uma vez que a obtenção de dados para tomada de decisão, acerca da evolução do processo nem sempre é trivial. O desenvolvimento de software, como qualquer disciplina de Engenharia, necessita de mecanismos de mensuração, para que através de uma avaliação de seus resultados o processo possa ser melhorado. Mas a mensuração pode ser uma tarefa bastante trabalhosa sem o auxílio de metodologias ou ferramentas de apoio. Neste sentido, este trabalho apresenta um modelo para mensuração em processos de software, no contexto do projeto APSEE-Prosoft. O objetivo do modelo proposto é prover suporte a realização de mensuração em processos de software, através do desenvolvimento e implantação de programas de mensuração, visando auxiliar aos gerentes na realização de atividades como planejamento, coleta e análise de métricas em um processo de software. Neste trabalho são combinados aspectos relacionados a importantes áreas da Engenharia de Software, como mensuração, processos de software, ambientes de engenharia de software centrados em processos (PSEE), bem como a utilização de métodos formais para a definição e especificação de um modelo de mensuração integrado a um PSEE. / The software process evaluation and improvement requests great effort of managers and developers, especially because obtaining data for decision support (about process evolution) is not a simple task. The software development, as any discipline of Engineering, needs measurement mechanisms, so that through an evaluation of its results, the process can be improved. But measurement can be a difficult task without the aid of methodologies or support tools. In this sense, this paper presents a model for software processes measurement, on the context of APSEE-Prosoft project. The model's objective is to provide support for the measurement in software processes, through the development and implantation of measurement programs, to aid managers in the accomplishment of activities such as planning, collecting and analysis of metrics in a software process. This paper combines aspects related to important areas of Software Engineering, as measurement, software process, process-centered engineering environment (PSEE), besides the use of formal methods for a definition and specification of a measurement model integrated to a PSEE.

Context: There has been a substantial growth of organizations adoptingAgile software development methodologies for various reasons. The requirementof improving the software processes with respect to traditional softwaredevelopment was clear and evident for different reasons. But the need forSoftware Process Improvements (SPI) in Agile context is unclear and thechallenges faced during the implementation of SPI in Agile software developmentare quite ambiguous. These two issues lie as a motivation for theobjectives of our study. Agile being a flexible way of software development,having a non-flexible framework is almost incompatible for implementingSPI in Agile software development. This acts as an inducement for buildingup our final objective. Objectives: The objectives of this research is to identify the need of Agile-SPI in software industry, challenges faced in implementing Agile-SPI atorganizational level and at team level and finally propose an approach forimplementing Agile-SPI based on improving practices. Methods: In order to achieve the objectives of our research, we initiallycarried out a survey, later cross verified and validated the data obtained inthe surveys through interviews. Literature review was performed to gainknowledge over the background and related work. Results: A total of 34 responses were obtained through survey. Theseresponses obtained through survey are further cross verified and validatedthrough 9 interviews. The data obtained through survey has been analyzedthrough descriptive statistics and the data obtained through interviews wasanalyzed using thematic coding. Conclusions: The need of Agile-SPI and the challenges faced by the organizationsand teams while implementing SPI in Agile software developmentwere identified. A total of 16 needs of Agile-SPI, 30 challenges faced byorganization and 37 challenges faced by team were drawn from survey andinterviews conducted. Finally, a conceptual framework has been proposedto implement SPI in Agile environment based on improving practices.

A avaliação e melhoria de um processo de software requer um grande esforço por parte de gerentes e desenvolvedores, uma vez que a obtenção de dados para tomada de decisão, acerca da evolução do processo nem sempre é trivial. O desenvolvimento de software, como qualquer disciplina de Engenharia, necessita de mecanismos de mensuração, para que através de uma avaliação de seus resultados o processo possa ser melhorado. Mas a mensuração pode ser uma tarefa bastante trabalhosa sem o auxílio de metodologias ou ferramentas de apoio. Neste sentido, este trabalho apresenta um modelo para mensuração em processos de software, no contexto do projeto APSEE-Prosoft. O objetivo do modelo proposto é prover suporte a realização de mensuração em processos de software, através do desenvolvimento e implantação de programas de mensuração, visando auxiliar aos gerentes na realização de atividades como planejamento, coleta e análise de métricas em um processo de software. Neste trabalho são combinados aspectos relacionados a importantes áreas da Engenharia de Software, como mensuração, processos de software, ambientes de engenharia de software centrados em processos (PSEE), bem como a utilização de métodos formais para a definição e especificação de um modelo de mensuração integrado a um PSEE. / The software process evaluation and improvement requests great effort of managers and developers, especially because obtaining data for decision support (about process evolution) is not a simple task. The software development, as any discipline of Engineering, needs measurement mechanisms, so that through an evaluation of its results, the process can be improved. But measurement can be a difficult task without the aid of methodologies or support tools. In this sense, this paper presents a model for software processes measurement, on the context of APSEE-Prosoft project. The model's objective is to provide support for the measurement in software processes, through the development and implantation of measurement programs, to aid managers in the accomplishment of activities such as planning, collecting and analysis of metrics in a software process. This paper combines aspects related to important areas of Software Engineering, as measurement, software process, process-centered engineering environment (PSEE), besides the use of formal methods for a definition and specification of a measurement model integrated to a PSEE.

The Information Technology (IT) market is growing. The number of companies and the volume of contracting in IT has increased. Because of the increasing competition, organizations need to improve their market presence and mainly seek increased efficiency and effectiveness through the implementation of software process improvement (SPI). Passos, Dias-Neto e Barreto (2012a) to believe that the results of this initiative appear when the achieved benefits are aligned with the organizational values. This study aims to verify the benefits achieved by organizations with the implementation of MPS.BR; to identify if organizational values are present in these organizations and to verify the existence of relationship between the benefits achieved by the implementation of software process models and organizational values. Data collection was conducted through questionnaires in organizations which implemented MPS.BR. The participants were 9 organizations, 47 questionnaires were answered. Benefits surveyed "clearer processes" and "processes with more quality" were the most perceived as "reduction in development time" and "reduction in the cost of development", the less checked by employees. After completing statistical analyzes, it was possible to reduce to four factors the 18 benefits studied, identifying the categories: quality, control, process and staff. Of the 10 organizational values studied, the most noted were "teamwork" and "responsibility" and the least observed were "strategy" and "information". The statistical analysis showed the relationship between the benefits and the organizational values, showed that organizational commitment and supervision values are strongly correlated of impactful on the benefits, suggesting that software development organizations that wish to achieve the benefits from the implementation of MPS.BR should focus their efforts on strengthening the supervision in their organization and develop a commitment to their team. / O mercado de Tecnologia da Informação está em ritmo crescente. Tem aumentado a quantidade de empresas e o volume de aquisições e contratações na área. Com a concorrência crescente, as organizações precisam melhorar sua presença no mercado e, principalmente, buscar aumento de eficiência e eficácia por meio da implementação de melhoria de processo de software. Segundo Passos, Dias-neto e Barreto (2012a), para que essa iniciativa traga resultados, um dos principais fatores é que os benefícios provenientes dessa implantação estejam alinhados com os valores organizacionais. Diante do exposto, este estudo pretende verificar quais os benefícios conquistados pelas organizações com a implementação da Melhoria de Processo de Software Brasileiro - MPS.BR; identificar quais valores organizacionais estão presentes nessas organizações; e verificar a existência de relações entre os benefícios alcançados pela implementação de modelos de processo de software e os valores organizacionais. O levantamento de dados foi realizado por meio de aplicação de questionários em organizações que possuem o Modelo de Processo de Software Brasileiro – MPS.BR implementado. Participaram da pesquisa 9 organizações, foram respondidos 47 questionários. Entre os benefícios pesquisados, ―processos mais claros‖ e ―processos com mais qualidade‖ foram os mais percebidos e ―redução no tempo de desenvolvimento‖ e ―redução no custo de desenvolvimento‖, os menos verificados pelos empregados. Depois de realizadas análises estatísticas, foi possível reduzir para 4 fatores os 18 benefícios estudados, identificando as categorias qualidade, controle, processo e equipe. Dos 10 valores organizacionais estudados, os mais constatados foram ―trabalho em equipe‖ e ―responsabilidade‖ e os menos observados foram ―estratégia‖ e ―informação‖. As análises estatísticas para verificar as correlações entre os benefícios e os valores organizacionais evidenciaram que os valores organizacionais, comprometimento e supervisão, se correlacionam, de forma impactante nos benefícios, sugerindo que as organizações de desenvolvimento de software que desejam alcançar os benefícios com a implementação do MPS.BR devem focar seus esforços em fortalecer a supervisão em sua organização e desenvolver o comprometimento em sua equipe. / São Cristóvão, SE

Made available in DSpace on 2015-04-11T14:03:21Z (GMT). No. of bitstreams: 1 Davi Viana dos Santos.pdf: 1879718 bytes, checksum: c5d0d7175f9d9c94ab6c07ac4434fa41 (MD5) Previous issue date: 2011-07-21 / CNPq - Conselho Nacional de Desenvolvimento Científico e Tecnológico / Several factors can influence the success of Software Process Improvement (SPI) programs. Results from previous research have shown that human and social aspects influence these programs in different ways. For this reason, it is important to analyze the human aspects involved on SPI programs to gain a better understanding about their influence on such programs. This thesis presents a research about human aspects influences in software process improvement. In this work, we analyze the major findings of a qualitative research conducted in three different phases to analyze which factors were most influential from view-point of the professionals working in the organizations. Our ultimate goal is to deepen the understanding about how human aspects can influence a SPI program in order to help in the success of future SPI implementation. / Diversos fatores podem influenciar o sucesso de um programa de Melhoria de Processo de Software (MPS). Os resultados de pesquisas anteriores mostram que vários desses fatores são de natureza humana e social. Por esta razão, é importante analisar os aspectos humanos envolvidos a fim de obter uma maior compreensão da influência destes em um programa MPS. Esta dissertação apresenta uma pesquisa sobre a influência dos aspectos humanos em Programas de MPS. Neste trabalho são analisados os principais resultados de uma pesquisa qualitativa conduzida em três diferentes fases para analisar quais fatores tiveram maior influência do ponto de vista dos colaboradores das organizações. O objetivo final é aprofundar a compreensão dos aspectos humanos que podem influenciar um programa de MPS com a finalidade de auxiliar no sucesso de futuros programas de melhoria.

Context: Software Process Improvement uses lean principles for eliminating wastes in the software development process. Waste is defined as anything that does not add value to the customer and product. The seven traditional wastes in software engineering are partially done work, extra process, extra features, waiting, motion, task switching and defects. Using the lean principles and practices, the wastes can be reduced or eliminated. It is important to know the lean practices that are widely used in software development and to know the practices, which captures the seven wastes. From the literature, the ability of waste reduction is theoretically discussed [2], but practically only little empirical evidence is available on ‘which practice is best capable in reducing specific kinds of waste.’ Objectives: Many software development organizations have adopted lean practices and agile practices for eliminating wastes of different kinds. Therefore, this study focuses on evaluating the effectiveness of lean practices in their ability to reduce the seven types of wastes associated with Lean Software Engineering. Methodology: The methodology that is used in this study is systematic literature review and industrial survey. In order to achieve the objective on evaluating the lean practices and agile practices in their ability to reduce the seven types of wastes that have more attention in research, a systematic literature review is conducted. Thereafter, to capture the effectiveness of lean practices in waste removal, a survey is designed to capture the perception of practitioners. Results: The systematic literature review has identified 53 relevant studies to the research topic. From these primary studies, the lean practices/principles, hybrid agile and lean practices, and the efficiency of agile practices in eliminating the seven wastes were identified. In addition to that, wastes that are captured by using lean practices; hybrid lean and agile practices were also identified. The reason for considering agile practices is that, agile and lean have similarities in eliminating the wastes and creating value to the customer [2][3]. Through the systematic literature review, it can be observed that researchers have not investigated all the seven wastes captured by lean practices. Thereafter, survey is the main contribution to this research where, the responses of 55 respondents from different countries were recorded. Most of the respondents are from India with 42% of the responses. We have asked the role of every respondent in their companies, and 19 respondents are Team leaders, 16 respondents are project managers and remaining people perform various other roles in software development. 38 respondents are from large-scale industry, which constitutes the majority part of the survey. The main contribution of the survey is identifying the ability of lean practices in eliminating different wastes. The respondents were asked to provide their level of agreement related to the effectiveness of each lean practices and agile practice in waste removal. A comparison between the results of SLR and survey reveals that there is a lot of variance in the perception of researchers and practitioners regarding the lean practices in eliminating wastes. Survey captures more wastes than SLR. From the results of survey, it was also identified that using the lean practices i.e. Kanban and VSM can eliminate most of the wastes. Kanban (41) and VSM (33) are mostly used and more efficient in industries. Conclusions: To conclude, the obtained results from this study will be quite useful for the real-time execution of lean practices. Team leaders, project managers, and organizations can adopt lean by choosing lean practices in accordance to the wastes that have to be eliminated from their software development process. This study has also identified the benefits and limitations of lean practices implemented in industry. This study helps researchers in providing necessary information that is very useful for further research in lean practices. The combinations of lean practices were also presented, which in terms one lean practice can compensate another in capturing all the seven wastes. In the survey, the additional wastes were identified when compared to SLR and this complements the literature. There is a considerable reach gap between the state of art and state of practice. It has been identified that VSM and Kanban practices have much attention in the literature. The remaining practices like Kotter and Kaizen are less concentrated in most of the research literature. From the literature, it is evident that none of the practices is capable of eliminating all the seven wastes in software development. VSM is capable of capturing wastes like waiting, extra process and motion. In addition, Kanban captures and eliminates wastes like partially done work, defects, task switching and extra features. With respect to the survey, Kanban and VSM practices are efficient in eliminating wastes. When the practitioners consider hybrid lean and agile practices, the combination of Kanban and Scrum, Scrum and VSM are efficient in eliminating wastes. The practitioners can consider the benefits of lean practices that are identified in this research. / V S S N R Ram Nanduri vssnrram@gmail.com

Abstract—Background: Software Process Improvement (SPI) maturity models have been developed to assist organizations to enhance software quality. Agile methodologies are used to ensure productivity and quality of a software product. Amongst others they are applied in Small and Medium – sized Enterprises (SMEs). However, little is known about the combination of Agile methodologies and SPI maturity models regarding SMEs and the results that could emerge, as all the current SPI models are addressed to larger organizations and all these improvement models are difficult to be used by Small and Medium – sized firms. Combinations of these methodologies could lead to improvement in the quality of the software products, better project management methodologies and organized software development framework. Objectives: The aim of this study is to identify the main Agile methodologies and SPI maturity models applied in SMEs, the combinations of these methodologies, and the results that could emerge. Through these combinations, new software development frameworks are proposed. What is more, the results of this study can be used as a guide with the appropriate combination for each SME, as a better project management methodology or as improvement in the current software engineering practices. Methods: A Systematic Literature Review was conducted, resulting in 71 selected relevant papers ranging from 2001 to 2013. Besides, a survey has been performed from June 2013 to October 2013, including 49 participants. Results: Seven Agile methodologies and six different SPI maturity models were identified and discussed. Furthermore, the combination of eight different Agile methodologies and Software Process Improvement maturity models is presented, and as well as their benefits and drawbacks that could emerge in Small and Medium – sized firms. Conclusion: The majority of the Agile methodologies and SPI maturity models are addressed to large or very large enterprises. Thus, little research has been conducted for SMEs. The combinations of the Agile methodologies and SPI maturity models are usually performed in experimental stages. However, it has been observed that such type of combination could present numerous benefits, which can also be applicable in SMEs as well. The combinations that are most common are the CMMI and XP, CMMI and Scrum, CMMI and Six Sigma, and the PRINCE2 and DSDM. / 0034-636835645

Aplicações sensíveis a contexto utilizam informações de contexto para fornecer serviços adaptados a usuários na realização de suas tarefas. Informação de contexto é qualquer informação considerada relevante para caracterizar entidades de uma interação usuário-computador, como a identidade e a localização de usuários. Esta tese trata a carência de uma abordagem que considere, em termos de processo de software, a complexidade de desenvolvimento de software sensível a contexto. O problema em questão é tratado por meio de três linhas de investigação: modelagem de informação contextual, serviços para tratamento de informação contextual e processo de software para computação sensível a contexto. As contribuições desta tese incluem: (i) o processo de software POCAp (Process for Ontological Context-aware Applications) para apoiar a construção de aplicações sensíveis a contexto baseadas em ontologias; (ii) o modelo de informações de contexto SeCoM (Semantic Context Model) baseado em ontologias e em padrões da Web Semântica; (iii) a infra-estrutura de serviços configuráveis SCK (Semantic Context Kernel) para interpretar informações de contexto apoiadas por modelos ontológicos de informação contextual, como o modelo SeCoM; (iv) uma instanciação do processo POCAp correspondente à extensão de uma aplicação com informações de contexto apoiadas pelo modelo SeCoM, e sua integração com serviços da infra-estrutura SCK; e (v) a identificação de questões de projeto relacionadas à inferência sobre informação contextual ontológica / In order to provide adaptive services according to users? tasks, context-aware applications exploit context information, which is any information used to characterize entities of a user-computer interaction such as user identity or user location. This thesis deals with the lack of a software process-based approach to supporting the inherent complexity of developing context-aware systems. The work reported in this thesis followed three main lines of investigation: context information modeling, services for processing context information, and a software process for context-aware computing. The contributions of this thesis include: (i) the Process for Ontological Context-aware Applications (POCAp) to support the development of context-aware applications based on ontologies; (ii) the Semantic Context Model (SeCoM) based on Semantic Web standards and ontologies; (iii) the Semantic Context Kernel (SCK) services infrastructure for interpreting ontological context information models such as the SeCoM model; (iv) an implementation of the POCAp process for the extension of an application with context information based on the SeCoM model, and its integration with services of the SCK infrastructure; and (v) the identification of design issues related to the inference over ontology-based context information