Workshop em Engenharia de Requisitos
Workshop en Ingeniería de Requerimientos
Workshop on Requirements Engineering

Anais do WER V. 10, 2007. Toronto - Canada


Expediente    Contato

@proceedings{WERpapers: WER07,
  editores = {Carina Alves Frota, Vera Werneck, Luiz Marcio Cysneiros},
  title = {Anais do WER07 - Workshop em Engenharia de Requisitos, Toronto, Canada, Maio 17-18, 2007},
  publisher = {},
  ISBN = {978-1-55014-483-3},
  ISSN = {2675-0066},
  year = {2007}
}

Introduction

1 - Introduction. Carina Alves Frota, Vera Werneck, Luiz Marcio Cysneiros. pp. i-xii, DOI . [pdf] [scholar]

Abstract: We welcome you to the 10th Workshop on Requirements Engineering. This year, WER takes place in Toronto, Canada. This is the first time of the WER workshop series to be held in North America. The workshop started ten years ago as a meeting of the Ibero American requirements engineering community, but over the years has attracted attention of researchers from other parts of the world as well.

Invited Speaker

2 - Requirements Engineering Lessons from House Building. Daniel M. Berry, . pp. 1-1, DOI . [pdf] [scholar]

Abstract: Anyone who has built or remodeled a house and has developed or enhanced SW must have noticed the similarity of these activities. This talk describes some lessons about requirements engineering I learned while being a customer in a house building and two house remodeling. The biggest problem is to avoid very expensive requirements creep. The main lesson is the importance of the customer insisting on following a full requirements engineering process, including goal identification, requirements elicitation, analysis, and specification, and validation of the specification. A secondary lesson is that a customer has an important role in requirements engineering and he or she sometimes needs to learn that role.

Papers

3 - Understanding the Strategic Actor Diagram: an Exercise of Meta Modeling. Julio Cesar Sampaio do Prado Leite, Vera Maria B. Werneck, Antonio de Padua Albuquerque Oliveira , Claudia Cappelli, Ana Luiza A. Cerqueira, Herbet de Souza Cunha , Bruno Gonzalez-Baixauli. pp. 2-12, DOI . [pdf] [scholar]

Abstract: i-star (i*) modeling uses the actor concept to ground the intentions of a given Universe of Discourse. Our work contributes to the understanding of the actor concept as used in i*. We have used a collaborative approach to better understand the actor concept. The authors met 9 times to discuss the topic. The goal was to discuss i* meta-models, which was later specialized to discuss actor modeling. After the meetings and after one week of collaborative work using a collaboration based editor, “Writely”, we have agreed on presenting our model from two different perspectives, but both using UML as the meta language. We understand that these models, designed by consensus, represent what we have labeled the SA Diagram or the Strategic Actor Diagram. The article presents the models we have arrived as well as the process we have used. We believe that making this process transparent will help to shed light not only on the concept of actor, but on the process of meta-modeling as well.

4 - Evaluating ADELFE Methodology in the Requirements Identification. V.M. B. Werneck, A. Y. Kano , L. M. Cysneiros. pp. 13 - 24, DOI . [pdf] [scholar]

Abstract: The increasing use of multi-agent systems brings challenges that have not been studied yet, such as how should we adapt requirements elicitation to cope with agent properties like autonomy, sociability and proactiveness. Many methodologies were proposed adopting this new paradigm. However, most of them are still in their early phases and therefore need to be adapted. In this work ADELFE, an agent-oriented methodology is evaluated. We use an exemplar proposed in 2001 by Yu and Cysneiros [1] to evaluate both agent/goal orientation and object orientation. This evaluation aims at analysing the strengths and weaknesses of ADELFE through the methodologies questions proposed in the exemplar.

5 - A Strategy for Information Source Identification. Julio Cesar Sampaio do Prado Leite, Edson Andrade de Moraes, Carlos Eduardo Portela Serra de Castro . pp. 25 - 34, DOI . [pdf] [scholar]

Abstract: Eliciting requirements is well known to be a difficult task. Several strategies exist that makes it possible to elicit information from different information sources. These strategies vary and should be selected in accordance with the situation at hand. For instance, selecting a strategy to a situation where there are available COTS solutions is different from a tailored and specialized software application. This article reports on a strategy to be used before the selection of an elicitation strategy. We are concerned in how to select the information sources that should be taken in consideration for the requirements elicitation. We consider this activity of fundamental importance to the elicitation process. Selecting the appropriate sources is fundamental as to deliver quality requirements. The article describes the proposed strategy and exemplifies it with a well known and published case, The London Ambulance Service.

6 - Descomposición de árboles de metas a partir de modelos de procesos. Jose Luis De la Vara González, David Anes Alcolea, Juan Sánchez Díaz. pp. 35 - 46, DOI . [pdf] [scholar]

Abstract: Los Sistemas de Información (SI) utilizados dentro de una organización no son algo separado de la organización empresarial a la que le dan soporte, y por tanto la Ingeniería de Requisitos (IR) debe considerar las necesidades de negocio de una organización. Aunque se reconoce que la IR es el puente natural que conecta el mundo empresarial y el mundo de los SI, la mayor parte de la investigación en IR continúa estando orientada a la solución, evitando considerar los problemas reales del mundo empresarial. Las necesidades de negocio pueden ser descritas mediante el alineamiento de los SI con la estrategia del negocio, los procesos de negocio, las infraestructuras organizacionales y las metas organizacionales. Una de las consecuencias del alineamiento entre negocio y los SI es el “mapeado” de las metas organizacionales y los procesos a la especificación del sistema En este trabajo se presenta una aproximación que utiliza una especificación en la forma de modelo de metas, construida mediante heurísticas en base a modelos de procesos en la forma de BPMN. A partir del modelo de metas, mediante un proceso de refinamiento y de etiquetado, se obtiene un modelo de requisitos en la forma de casos de uso. La especificación así obtenida permite reflejar de manera más cercana las necesidades del negocio y asegura el alineamiento de las mismas con el futuro SI.

7 - Guidelines for Business Modeling Elaboration based on Views from Domain Information. Silvia Angelica Zanco Ladeira, Maria Istela Cagnin. pp. 47 - 55, DOI . [pdf] [scholar]

Abstract: Business modeling is an activity of the Requirements Engineering that involves a knowledge process of the organization and provides a specific business domain view. On this process, models that can approach the context in which the system domain is inserted in different views (roles, business process and business rules) are built. In spite of the importance of these models, in general, they are not part of the software documentation due to the additional cost and time to build them. One solution for this is the business modeling reuse. In this context, this paper presents a set of guidelines for elaboration of business modeling based on views starting from the documentation of a certain domain, aiming to make available such modeling for reuse on the software development, maintenance and reengineering. A case study to evaluate the applicability of these guidelines was performed and it is also presented on this paper.

8 - Evolving Use Case Maps as a Scenario and Workflow Description Language. Gunter Mussbacher, . pp. 56 - 67, DOI . [pdf] [scholar]

Abstract: Since 1996, the core Use Case Map (UCM) nota-tion has remained remarkably stable. As the structure and intent of workflow and scenario languages are very similar, UCMs have been applied to scenario, workflow, and business process modeling. The recent rise of workflow languages for the description of busi-ness processes and web services resulted in a more formal assessment method for such languages based on generic workflow and communication patterns. We present such an assessment for UCMs, thereby meas-uring the applicability of UCMs for workflow descrip-tion in particular and scenario descriptions in general and gathering evidence on how to evolve the UCM notation. The results are compared to similar assess-ments which were carried out for current standards for workflow, business process design, and business proc-ess execution languages such as the Business Process Modeling Notation (BPMN), the Business Process Execution Language for Web Services (BPEL4WS), and UML 2.0 Activity Diagrams.

9 - Integración de un entorno de producción automática de software en un marco de alineamiento estratégico. José Luís De la Vara González, Juan Sánchez Díaz, Óscar Pastor López. pp. 68 - 79, DOI . [pdf] [scholar]

Abstract: El alineamiento entre la estrategia de negocio y las tecnologías de información (TI) es reconocido como uno de los principales problemas a los que se enfrentan los responsables de una organización. Pese a que han sido presentadas numerosas propuestas para alcanzar este objetivo, sigue siendo frecuente que las inversiones que se realizan en TI no reporten todas las ventajas que se esperan de ellas porque no encajan con las necesidades reales de la organización. En este trabajo se presenta una aproximación para afrontar el alineamiento estratégico basada en la definición de un marco en el que se integre un entorno de producción automática de software. Para ello se usan el modelo de alineamiento estratégico SAM, el framework Zachman, el modelo de gestión empresarial Balanced Scorecard (BSC) y OO-Method, cuya unión tiene como fin la generación automática de infraestructuras TI a partir de infraestructuras organizacionales acordes a la estrategia de negocio, posibilitando así el alineamiento.

10 - Business Process Monitoring and Alignment: An Approach Based on the User Requirements Notation and Business Intelligence Tools. Alireza Pourshahid, Daniel Amyot , Pengfei Chen Michael Weiss , Alan J. Forster. pp. 8- - 91, DOI . [pdf] [scholar]

Abstract: Monitoring business activities using Business Intel-ligence (BI) tools is a well-established concept. How-ever, online process monitoring is an emerging area which helps organizations not only plan for future im-provements but also change and alter their current ongoing processes before problems happen. In this paper, we explore how monitoring process perfor-mance can help evolve process goals and require-ments. We elaborate an approach that uses the User Requirements Notation (URN) to model the goals and processes of the organization, and to monitor and align processes against their goals. A BI tool exploit-ing an underlying data warehouse provides the Key Performance Indicators (KPI) used to measure the satisfaction of goals and process requirements. Feed-ing this information into the URN modeling tool, we can analyze the consequences of current business ac-tivities on desired business goals, which can be used for process and business activity alignment thereafter. We illustrate the approach with a case study from the healthcare sector: a hospital discharge process.

11 - An Approach to Requirements Encapsulation with Clustering. Zude Li, Quazi A. Rahman, Nazim H. Madhavji. pp. 92 - 96, DOI . [pdf] [scholar]

Abstract: Requirements encapsulation means organizing software requirements into a set of requirements clusters with tight cohesion along with external interfaces such that each cluster can be ultimately implemented by a functionality module. We propose an approach to encapsulating requirements which includes two steps: clustering requirements based on the similarity and associativity relations and then encapsulating each cluster by defining its external interface as stimulus-response pairs. The potential benefits of encapsulating requirements are reduced software development and maintenance costs.

12 - Extended Disambiguation Rules for Requirements Specifications. Sri Fatimah Tjong, Michael Hartley, Daniel M. Berry. pp. 97 - 106, DOI . [pdf] [scholar]

Abstract: This paper extends earlier work by the authors in identifying guiding rules for natural language (NL) requirements specifications (RSs) by analysing a few sets of requirements documents from different domains. It presents guiding rules that help reduce ambiguities and imprecision in NL RSs. It validates these rules by applying them to sentences in several industrial strength NL RSs.

13 - Evaluating the Effectiveness of Using Catalogues to Elicit Non-Functional Requirements. Luiz Marcio Cysneiros, . pp. 107 - 115, DOI . [pdf] [scholar]

Abstract: Non-Functional Requirements (NFR) are subjective, interactive and relative, thus realizing the need for particular NFR is by itself a challenge. Furthermore understanding what the software must implement in order to cope with these needs may prove to be an even more challenging task. One way of addressing the need for help on NFR elicitation is the use of catalogues. However, it is not clear how effective it is to use them. This work investigates it through an empirical study where different teams will model the same problem. Two teams will use catalogues with a systematic method, another two teams will use catalogs in an ad hoc manner and yet another two teams will not use catalogues. We show at the end of this work that teams using catalogues performed significantly better.

14 - Test-case Driven versus Checklist-based Inspections of Software Requirements–An Experimental Evaluation. Nina D. Fogelström, Tony Gorschek. pp. 116 - 126, DOI . [pdf] [scholar]

Abstract: Software inspections have proved to be an effective means to find faults in different software artifacts, and the application of software inspections on requirements specifications is believed to give a high return on investment as problems are caught early. However, despite the existing evidence of positive effects requirements inspections are not a common practice in industry. The reason is believed to be the cost associated with inspections as a technology. This paper presents an evaluation of test-case driven inspections (TCD) - an emerging inspection technique that aims to cut costs associated with traditional requirements inspections. To formally test the efficiency and effectiveness of TCD inspections an experiment was conducted, in a controlled environment, where checklist based inspections was used as a point of reference. The experiment results indicate that TCD inspections perform better when it comes to effectiveness in finding major faults in a requirements specification.

15 - Preliminary Results from an Empirical Study in Market-Driven Software Companies. Carina Alves, Silvia Pereira, George Valença, João Pimentel , Rodolfo V. C. L. de Andrade. pp. 127 - 134, DOI . [pdf] [scholar]

Abstract: In this paper we present initial findings from an empirical study that has been conducted with 13 marketdriven software companies based in Recife, Brazil. The objective of the study is to explore the state of the practice in requirements engineering (RE) for commercial off-theshelf software products. The study followed a qualitative research method using questionnaire and semi-structured interviews. The empirical study investigates key challenges faced by market-driven companies previously identified in a literature survey. Our goal was to gather empirical evidence from studied companies to evaluate specific facts concerning RE for software products identified in the literature.

16 - Towards Variability Design as Decision Boundary Placement. Catalin Bidian, Eric S.K. Yu. pp. 139 - 148, DOI . [pdf] [scholar]

Abstract: Complex information systems have numerous design variables that are systematically decided upon during the design process. In high-variability systems, some of these decisions are left open and deferred to later stages. For example, in product line architectures, some decision variables are used to generate families of products with variations in features. In user-adaptive systems, the behavior of the system is determined at runtime, based on user characteristics and preferences. In this paper, we propose to characterize variability in terms of boundaries in design decision graphs which depict the space of alternatives. A design decision about variability, such as what choices should be left to the user and which ones should be fixed at which stage in the design process, is then a question of where to place that decision boundary along some path in the relevant decision graph.

17 - A Requirements Management Framework for Privacy Compliance. Sepideh Ghanavati, Daniel Amyot, Liam Peyton. pp. 149 - 159, DOI . [pdf] [scholar]

Abstract: Compliance with privacy legislation is a primary concern for health care institutions that are building information systems support for their business processes. This paper describes a requirements management framework that enables health information custodians (HIC) to document and track compliance with privacy legislation. A metamodel is defined for our framework to define compliance tracking links between separate User Requirements Notation models of the HIC and privacy legislation. Using examples from a case study at a major teaching hospital, we show how this framework can be used to manage change and ensure compliance when privacy legislation is amended or the business processes evolved.

18 - O Gerenciamento de Requisitos no Ambiente COCAR. André Di Thommazo, Marcos Danilo Martins, Sandra Fabbri. pp. 160 - 170, DOI . [pdf] [scholar]

Abstract: This research group has been working towards building an environment which helps software development, providing support to some activities mainly based on use case model. The reading technique TUCCA and a template to specify requirements were previously defined in another work of this group and are the basis of this environment. Aim. The aim of this paper is to give an overview of COCAR environment, emphasizing the requirement management activities supported by it. Method. Based on literature and several proposals of other authors, we introduced in the environment some functionalities that support requirement management and provide traceability between the requirements and the use case model. Results. The environment was used in a simple example of a real company and the results motivated the improvement of the environment and the continuity of this research. Conclusions. The information provided by the requirement functionalities give relevant support to maintain the requirement exchange controls, consistence and the development planning, supporting and facilitating some activities related to requirement management.

Short Presentation Papers

19 - i* with Aspects: Evaluating Understandability. Ricardo A. Ramos, Fernanda Alencar , João Araújo , Ana Moreira, Jaelson Castro, Rosangela Penteado. pp. 171 - 178, DOI . [pdf] [scholar]

Abstract: The applicability of the i* approach for organizational modeling has been compromised by the complexity of the resulting models. To solve this problem an approach that use i* with Aspect Orientation is proposed. In this work we evaluate the understandability of five selected concerns of two requirements documents. The first requirements document was modeled by the i* approach and the other one by i* extended with aspects. To do this evaluation it was used abstract metrics that needed to be instantiated in both requirement documents.

20 - Analyzing Problem Frames together with Solution Patterns. Ellen Souza, Maria Lencastre , Renata Melo, Lilian Ramires, Keldjan Alves. pp. 179 - 189, DOI . [pdf] [scholar]

Abstract: The Problem Frames approach defines identifiable problem classes based on, among other things, their context and the characteristics of their domains, interfaces and requirements, without going deeply into the solution. Other software engineering approaches deal with the concept of patterns that present wellknown solutions, such as archetype, analysis and design patterns. We can say, for instance, that patterns are about solutions and problem frames are about problems. This paper attempts to make an analysis of the integration of problem classes, that is problem frames, and solutions, by analyzing a set of different kinds of patterns together within problem frames. The relationship, between these approaches, seems to have a good chance of improving software development.

21 - Exploiting a Goal-Decomposition Technique to Prioritize Non-functional Requirements. M. Daneva, M. Kassab, M. L. Ponisio, R. J. Wieringa , O. Ormandjieva. pp. 190 - 196, DOI . [pdf] [scholar]

Abstract: Business stakeholders need to have clear and realistic goals if they want to meet commitments in application development. As a consequence, at early stages they prioritize requirements. However, requirements do change. The effect of change forces the stakeholders to balance alternatives and reprioritize requirements accordingly. In this paper we discuss the problem of priorities to non-functional requirements subjected to change. We, then, propose an approach to help smooth the impact of such changes. Our approach favors the translation of nonoperational specifications into operational definitions that can be evaluated once the system is developed. It uses the goal-question-metric method as the major support to decompose non-operational specifications into operational ones. We claim that the effort invested in operationalizing NFRs helps dealing with changing requirements during system development. Based on this transformation and in our experience, we provide guidelines to prioritize volatile non-functional requirements.

22 - Requirements for Tools for Ambiguity Identification and Measurement in Natural Language Requirements Specifications. Nadzeya Kiyavitskaya, Nicola Zeni, Luisa Mich, Daniel M. Berry. pp. 197 - 206, DOI . [pdf] [scholar]

Abstract: This paper proposes a two-step approach to identifying ambiguities in natural language (NL) requirements specifications (RSs). In the first step, a tool would apply a set of ambiguity measures to a RS in order to identify potentially ambiguous sentences in the RS. In the second step, another tool would show what specifically is potentially ambiguous about each potentially ambiguous sentence. The final decision of ambiguity remains with the human users of the tools. The paper describes two requirements-identification case studies with one small NL RS using a prototype of the first tool based on an existing NL processing system and a manual simulation of the second tool.

23 - Role-Based Access Control Requirements Model with Purpose Extension. Faranak Farzad, Eric Yu, Patrick C. K. Hung. pp. 207- 216, DOI . [pdf] [scholar]

Abstract: Role-Based Access Control (RBAC) is increasingly used for ensuring security and privacy in complex organizations such as healthcare institutions. In RBAC, access permissions are granted to an individual based on her defined roles. Much work has been done on the specification of RBAC models for enforcing access control; however, in order to arrive at appropriate choices of access control for particular roles and individuals in an organization, we need models at the requirements level to support elicitation and analysis. Crook et al. [3] have provided a requirements level model for RBAC, defining access to an information asset based on role, responsibility, operation, and context. We extend the Crook model to include a purpose hierarchy in order to meet the needs of privacy requirements. Access to health records is used as the example domain.

  voltar