Figure 2 - uploaded by Jose Luis de la Vara
Content may be subject to copyright.
Citations
... There also exist model-based means for other related general areas. Examples of these include prior work on metamodels for artifacts [30], for artifact traceability [31], for processes and the artifacts involved [32], or for safety-critical systems [33]. We do not review these kinds of publications in detail because they do not explicitly tackle the needs of assurance evidence management, e.g., the need to assess and guarantee the suitability of the artifacts used as evidence for compliance and dependability justification purposes. ...
Most safety–critical systems are subject to rigorous assurance processes to justify that the systems satisfy given requirements and are dependable. These processes are typically conducted in compliance with standards and require the provision of assurance evidence in the form of system artifacts, such as system specifications and testing results. The management of assurance evidence is usually a complex process because of the large number of artifacts to deal with, the amount of information to gather about the artifacts, and the need to guarantee evidence quality, among other issues. Our aim is to facilitate assurance evidence management by means of a model-based approach. The approach is based on a metamodel that defines the information to be collected about evidence artifacts during their lifecycle. A process for assurance evidence management and usage guidance are also presented. The approach has been developed in the scope of several industry-academia projects, implemented in the OpenCert tool, and validated by practitioners in 10 industrial case studies. Based on the results of the validation, we argue that the approach is an effective means for assurance evidence management and that it could improve the state of the practice.
... Among about twenty papers that consider the BPMN-based MDSDM, only a few papers [64,63,49,34] present some tool (mainly ATL-and QVT-based transformation programs) enabling the automatic MDSDM, but with very low effectiveness. The semiautomatic BPMN-based MDSDM is presented in [65,12,13,32,35], while the other proposals [76,56,35,45,31] are not implemented at all. Regarding the formalism level of the existing BPMN-based approaches, the formal rules are presented in [23,66], and partially in [30,29,31], while the others give only the informal guidelines. ...
... Regarding the POM-based MDSDM approaches, the case-study based evaluation results are reported in [23,25,55], while the results of controlled experiments are reported in [32,40,7,22,20]. The most complete evaluation results, which are based on the experiment conducted with a significant number of practitioners, are presented in [22,20] (average completeness of the generated models is over 80%). ...
... 31 All manually designed CDMs were analyzed by the teachers and evaluated as adequate. 32 Each manually designed CDM was then compared with the CDM automatically generated based on the reference set of the main BPMs, and the same metrics and measures were used in the evaluation as in the case study and in E-1. ...
The article presents an approach aimed at automatically deriving the initial conceptual database model from a set of business process models. The approach proposes the incremental synthesis of the target model by iteratively composing the partial conceptual database models that are derived from the models contained in the source set. The approach is implemented by the AMADEOS tool, which is the first online web-based tool enabling the automatic derivation of the conceptual database model from a set of business process models. The experimental evaluation proves that the implemented approach enables effective automatic derivation of the initial conceptual database model.
... The BPMN is used in [17,18,26,36,37,39,40,43,44,66,78,87,89,90,[92][93][94]112] as the starting point for MDSDM. There are three QVT-based proposals [66,89,90], but with modest achievements in the automatic generation of analysis level class diagrams. ...
... There are three QVT-based proposals [66,89,90], but with modest achievements in the automatic generation of analysis level class diagrams. There is an XSLT-based proposal [43] for automatic generation, as well as several proposals [17,18,39,44,93] for semiautomatic generation of class diagrams. A MDSDM based on BPMN is also considered in [44,78,112], but without implementation. ...
... Regarding the evaluation of the POM-based approaches, three papers [26,28,75] report case-study-based evaluation, while the results of controlled experiments are reported in [10,39,48]. ...
The article presents the results of an experiment we conducted with database professionals in order to evaluate an approach to automatic design of the initial conceptual database model based on collaborative business process models. The source business process model is represented by BPMN, while the target conceptual model is represented by the UML class diagram. The results confirm those already obtained in a case-study-based evaluation, as well as those of an earlier controlled experiment conducted with undergraduate students. The evaluation implies that the proposed approach and implemented generator enable automatic generation of the target conceptual model with a high percentage of completeness and precision. The experiment also confirms that the automatically generated model can be efficiently used as a starting point for manual design of the target model, since it significantly shortens the estimated efforts and actual time spent to obtain the target model in contrast to the manual design from scratch.
... According to [38], the model describes a set of organized and structured activities with defined inputs and defined outputs. Moreover, design or modeling of the business process is described as the time period during which manual and/or automated workflow descriptions of a process are described [40]. The workflow model typically contains a collection of concepts that are useful in the description of the business process, its activities, the dependencies among activities, and the requirements that will implement the required activities [41]. ...
In recent years, the process orientation requirements engineering field has received significant interest. However, dealing with ordinary users within an enterprise for developing as-is business process is very complex because this requires skills, responsibility, knowledge, and expertise of business processes. This research answers the following questions: 1) how to systematically gather information regarding as-is business process requirements in an informal environment and by non-expert users? and 2) how can an enterprise refine the identified business process with a semantic process model? The primary purpose of this research is to develop a supporting framework that is suitable for the definition of an as-is business process to be deployed within the enterprise environment. Hence, the focus is on gathering requirements to define the as-is business process. This framework concentrates on three significant aspects of this process: development, refinement, and serialization the semantics of the process model. To accomplish this objective, the LORS framework has been proposed, which consists of four phases (List, Order, Refinement, and Serialization). The framework presents a new unique technique to identify the business process, refine the process, and generate the model semantics. This study assumes that a simple and complete framework will help non-expert users define a high-quality as-is process, such that enhance and facilitate the matching process with existing reference models. We present a case study, evaluate the case study relative to specified criteria, and research the limitations and implications discovered from our research. This research concludes that the LORS framework is simple, flexible, visible, interactive, dynamic, and effective.
... • conhecimentos sobre a norma ISO/IEC 25000 para a qualidade do produto de software (Capítulo 5 Vara, 2011]. ...
O conteúdo deste livro aborda conceitos, teorias e, principalmente,
as principais abordagens sobre gestão, qualidade e
processos de software tratadas em âmbito nacional pela indústria
de software. Seu objetivo é apresentar e ilustrar como
estas abordagens devem ser tratadas e usadas no cenário de
desenvolvimento de software para resolver problemas de
gestão, qualidade e processos de software. O livro pode atuar
como como um complemento ao ensino-aprendizagem sobre
as melhores práticas usadas para atingir um programa de
melhoria organizacional.
Este livro é fruto dos resultados obtidos em uma disciplina
ministrada no ano de 2015 no Centro de Informática da UFPE
para os alunos de Pós-Graduação (Mestrado e Doutorado) em
Ciência da Computação, que tinham o objetivo de: apresentar
seminários sobre temas no assunto do livro; discutir em
sala de aula sobre as informações tratadas nos seminários;
apresentar uma revisão sintática e semântica (com críticas
Prólogo e sugestões de melhoria) sobre o material apresentado; e
desenvolver/refinar as informações tratadas nos seminários
na forma de um artigo, sendo este acompanhado pelos
professores responsáveis pela disciplina, Prof. Alexandre
Vasconcelos (Professor Doutor do Centro de Informática da
UFPE) e Prof. Sandro Bezerra (Professor Doutor do Programa
de Pós-Graduação em Ciência da Computação da UFPA). Estes
artigos, obrigatoriamente, tinham como meta apresentar os
resultados da experiência da aplicação e/ou da utilização na
indústria de software dos temas tratados neste livro. É direcionado
a alunos, gestores, implementadores, avaliadores e
outros interessados no tema gestão, qualidade e processos
de software.
Por último, este livro não é autossuficiente. Ele exige que o
leitor utilize as referências bibliográficas que apresentamos
de materiais superlativos. Se algo destacamos deste livro é
que a bibliografia vale a pena por si só.
... O processo de negócio é um conjunto de uma ou mais atividades executadas que seguem uma ordem pré-definida para realizar uma meta objetiva de negócios, normalmente, dentro do contexto de uma estrutura organizacional que define papéis funcionais ou relacionamentos [1,8,11,20,21]. Além disso, um processo pode ser inteiramente contido dentro de uma única organização/unidade, assim como ele pode se estender por várias organizações diferentes. ...
... Para a fase de formalização do processo de negocio a linguagem comumente usada tem sido a linguagem de notação visual BPMN (Business Process Model and Notation) [5]. A notação e o modelo de processo de negócio BPMN é um padrão definido pela OMG [5] para representar, de uma maneira gráfica, os processos que ocorrem dentro de uma organização [6][7] [8]. A notação gráfica facilita o entendimento dos envolvidos, incentivando as colaborações e as transações de negócios na organização [6]. ...
Business process modelling is nowadays an important phase for requirements engineering. The description of the business process aims to reduce the distance between the users of the system and its developers. This paper presents a study case regarding the application of business process mapping using BPMN (Business Process Model and Notation), to develop a system for human resources management within the Ministry of Planning, Budget and Management in Brazil. The system to be developed has an estimated 60,000 function points effort and involves more than 500 people in the requirements phase and more in the subsequent phases of process mapping, development, validation and test.
... The resulted models (As-Is) describe how the business is working today at any level of abstraction from enterprise-wide processes to single tasks performed by single people. For various guidelines that can support business process modeling, the reader is referred to (Vara, 2011). ...
The hardest single part of building a software system is deciding precisely what to build " (Brooks, 1987). Faulty requirements analysis is a major reason for project failure or unsatisfactory information systems that do not fulfill business needs. Although it has been long recognized that system requirements can only be understood within the context of the business processes and business modeling has become the initial phase of most software processes, the transition between business modeling and requirements gathering is still a challenge for research. Moreover, existing work in this area tends to introduce accidental difficulties. This paper reports the results of an action research conducted for elaborating a Business-Oriented approach to Requirements Elicitation. Our approach integrates Requirements Engineering with Business Process Engineering and derives system requirements based on business process models. This ensures that system requirements meet real business needs. The proposed approach is illustrated by a real-world example.
... Although the BPMN is used in 12 papers [47]- [58], there is no paper presenting an automatic generator of the data model based on the complete source model. There are two QVT-based proposals [50], [53], but with modest achievements in the automated generation of the analysis level class diagram, as well as several proposals [48], [49], [52], [56], [58] for the semiautomated generation. ...
... • the majority of proposed approaches are based on guidelines (∼ 30%) and informal rules (∼ 40%), but the development of transformation languages (ATL and QVT) has made an important contribution to the formalization and automatization of the MDSDM approaches in the recent years, so the participation of (semi)automatic techniques is significant (∼ 60%); • a small percentage (∼ 10%) of approaches were evaluated based on a controlled experiment [56], [67] or a single case study [45], [46], while all others were not evaluated at all, or just illustrated by some examples without any quantitative/qualitative evaluation; • only one paper [46] presents an automatic data model generator based on the complete source model, but the generated data model can be considered just as an initial data model, i.e. an analysis level data model; • the semantic capacity of POMs has not yet been sufficiently identified to enable the automatic synthesis of the complete target data model, since the existing approaches still do not have a significant recall in the automated generation of some types of associations and class members. ...
This article presents a survey of model-driven techniques for data model synthesis. During an extensive research, we identified more than 70 research papers in the field and more than 15 different graphical notations used for the source model representation. We have classified the proposed approaches into four distinct groups: function-oriented, process-oriented, communication-oriented and goal-oriented. Their contributions are presented in chronological order and evaluated based on several main criteria. Although the idea of model-driven design of the data model is more than 25 years old, the survey shows the richness and diversity of ideas, but only a small number of implemented automatic generators.
... Software model quality has been the focus of much research during the last two decades. Starting from quality needs for conceptual modelling [31], many researchers have studied and provided specific means to deal with model quality for business process management (e.g., [4]), modeldriven development (e.g., [33]), and requirements engineering (e.g., [15]), among other fields. Other authors have presented general, abstract frameworks (e.g., [36]). ...
... Examples of software-related modelling techniques for which insights into model quality in real projects have been provided are the BPMN notation for business process modelling [15][61], the EKD framework for organizational modelling [56][57], the EPC notation for business process modelling [9], the i* goal-oriented notation [18], the UML language for software modelling [11] [27], UML profiles for modelling of real-time systems [28] [29], and use cases for requirements modelling [13]. Weaknesses identified in the techniques in these works that can lead to a lower model quality are concept redundancy (e.g., in BPMN), understandability difficulties (e.g., in i*), and concept overload (e.g., in UML). ...
... Although they have not provided many insights into the state of the practice, other works have provided valuable guidance for software model quality for activities such as specification of requirements (e.g., [1][52] [55]), identification of classes (e.g., [15] [58]), creation of conceptual schemas (e.g., [40][49] [50]), or modelling of business processes (e.g., [34] [53] [54]). ...
Although past research has resulted in different means to deal with software model quality, creation of adequate software models remains challenging. Any modelling effort must be carefully analysed and planned before it starts, and definition or adoption of modelling guidelines is usually necessary. In addition, the amount of publications addressing model quality in practice is low, and the knowledge about others' experience regarding model quality is limited. This paper reports on our experience in dealing with software model quality in the context of a project between industry and academia. Such a project corresponds to a large-scale research project in which modelling has been used both as part of the necessary work for executing the project and for creating project results. We present how we have dealt with model quality in requirements modelling and in conceptual model specification, as well as a set of lessons learned. The insights provided can help both researchers and practitioners when having to deal with software model quality.
... We conceive business process models as an important artefact for representing organisational behaviour. Since, there are already several works that cover the derivation of information systems from/to business processes (reverse engineering or forward engineering)789. We focus on developing specific artefacts for evolving business process models. ...
Enterprise systems need to evolve to be adapted with changes in their context. Model-driven reengineering frameworks have been developed in order to support post-delivery life cycle activities and software system maintenance (for instance, architecture-driven modernisation is an object management group initiative). However, existing proposals are general purpose and full support for an organisational reengineering framework is still needed. This paper presents a research effort aiming at designing and developing guidelines and tools for supporting model-driven organisational improvement. We propose to support model evolution by means of two artefacts: a pattern definition metamodel and an evolution metamodel. As a proof of concept, this paper applies the proposal to support business process evolution. We illustrate an instance of these metamodels by means of a laboratory-demo. We also discuss benefits and limitations of our proposal.
