Exploring how to use scenarios to discover requirements

This paper investigates the effectiveness of different uses of scenarios on requirements discovery using results from requirements processes in two projects. The first specified requirements on a new aircraft management system at a regional UK airport to reduce its environmental impact. The second specified new work-based learning tools to be adopted by a consortium of organizations. In both projects scenarios were walked through both in facilitated workshops and in the stakeholders’ workplaces using different forms of a scenario tool. In the second project, scenarios were also walked through with a software prototype and creativity prompts. Results revealed both qualitative and quantitative differences in discovered requirements that have potential implications for models of scenario-based requirements discovery and the design of scenario tools.     

Designing with scenarios: A critical review of current research and practice

There is a great deal of requirements engineering specifically, and information systems development research in general, in the area of scenarios as the ‘vocabulary’ for discussing and characterising designs for new artefacts. This is partly due to opening up the design process to a variety of participants in a project, and making explicit their ways of working, thinking and interactions with the design products. Scenarios, being concrete, and easy to understand and use, provide the means to describe the design vocabulary. They are used to focus on episodic cases, exchange ideas and thoughts about them effectively, and generally describe requirements and designs for a new artefact from the user’s perspective. In this paper, we review the effectiveness of the current state-of-practice in scenario-based approaches. The objective of this evaluation exercise is to define the requirements for improved ‘by scenario’ approaches to cope with requirements and designs for developing new artefacts.     

A software product certification model

Certification of software artifacts offers organizations more certainty and confidence about software. Certification of software helps software sales, acquisition, and can be used to certify legislative compliance or to achieve acceptable deliverables in outsourcing. In this article, we present a software product certification model. This model has evolved from a maturity model for product quality to a more general model with which the conformance of software product artifacts to certain properties can be assessed. Such a conformance assessment we call a ‘software product certificate’. The practical application of the model is demonstrated in concrete software certificates for two software product areas that are on different ends of the software product spectrum (ranging from a requirements definition to an executable). For each certificate, a concrete case study has been performed. We evaluate the use of the model for these certificates. It will be shown that the model can be used satisfactorily for quite different kinds of certificates.     

A Representational Framework for Scenarios of System Use

Scenarios are becoming widely used in three areas of system development: software engineering, human–computer interaction (HCI), and organisational process design. There are many reasons to use scenarios during system design. The one usually advanced in support of the practice is to aid the processes of validating the developers’ understanding of the customers’ or users’ work practices, organisational goals and structures, and system requirements. All three areas identified above deal with these processes, and not surprisingly this has given rise to a profusion of scenario-based practices and representations. Yet there has been little analysis of why scenarios should be useful, let alone whether they are. Only by having such a framework for understanding what scenarios are, and what they are for, can we begin to evaluate different scenario approaches in specific development contexts. This paper is a contribution toward such a framework. We lay out a space of representational possibilities for scenarios and enumerate a set of values or criteria that are important for different uses of scenarios. We then summarise several salient representations drawn from the software engineering, HCI, and organisational process design communities to clarify how these representational choices contribute to or detract from the goals of the respective practices. Finally, we discuss how scenario representations from one area of design may be useful in others, and we discuss the relationship between these representations and other significant early-design and requirements engineering practices.     

An integrative approach to requirements analysis: How task models support requirements reuse in a user-centric design framework

Many software systems fail to address their intended purpose because of a lack of user involvement and requirements deficiencies. This paper discusses the elaboration of a requirements-analysis process that integrates a critical-parameter-based approach to task modeling within a user-centric design framework. On one hand, adapting task models to capture requirements bridges the gap between scenarios and critical parameters which benefits design from the standpoint of user involvement and accurate requirements. On the other hand, using task models as a reusable component leverages requirements reuse which benefits design by increasing quality while simultaneously reducing development costs and time-to-market. First, we present the establishment of both a user-centric and reuse-centric requirements process along with its implementation within an integrated design tool suite. Secondly, we report the design, procedures, and findings of two user studies aimed at assessing the feasibility for novice designers to conduct the process as well as evaluating the resulting benefits upon requirements-analysis deliverables, requirements quality, and requirements reuse.     

Studying the co-evolution of production and test code in open source and industrial developer test processes through repository mining

Many software production processes advocate rigorous development testing alongside functional code writing, which implies that both test code and production code should co-evolve. To gain insight in the nature of this co-evolution, this paper proposes three views (realized by a tool called TeMo) that combine information from a software project’s versioning system, the size of the various artifacts and the test coverage reports. We validate these views against two open source and one industrial software project and evaluate our results both with the help of log messages, code inspections and the original developers of the software system. With these views we could recognize different co-evolution scenarios (i.e., synchronous and phased) and make relevant observations for both developers as well as test engineers.     

Computer-aided Support for Secure Tropos

In earlier work, we have introduced Secure Tropos, a requirements engineering methodology that extends the Tropos methodology and is intended for the design and analysis of security requirements. This paper briefly recaps the concepts proposed for capturing security aspects, and presents an implemented graphical CASE tool that supports the Secure Tropos methodology. Specifically, the tool supports the creation of Secure Tropos models, their translation to formal specifications, as well as the analysis of these specifications to ensure that they comply with specific security properties. Apart from presenting the tool, the paper also presents a two-tier evaluation consisting of two case studies and an experimental evaluation of the tool’s scalability.     

Commitment analysis to operationalize software requirements from privacy policies

Online privacy policies describe organizations’ privacy practices for collecting, storing, using, and protecting consumers’ personal information. Users need to understand these policies in order to know how their personal information is being collected, stored, used, and protected. Organizations need to ensure that the commitments they express in their privacy policies reflect their actual business practices, especially in the United States where the Federal Trade Commission regulates fair business practices. Requirements engineers need to understand the privacy policies to know the privacy practices with which the software must comply and to ensure that the commitments expressed in these privacy policies are incorporated into the software requirements. In this paper, we present a methodology for obtaining requirements from privacy policies based on our theory of commitments, privileges, and rights, which was developed through a grounded theory approach. This methodology was developed from a case study in which we derived software requirements from seventeen healthcare privacy policies. We found that legal-based approaches do not provide sufficient coverage of privacy requirements because privacy policies focus primarily on procedural practices rather than legal practices.     

Scenario Development and Practical Decision Making under Uncertainty: Application to Requirements Engineering

In this paper, we address the question of how flesh and blood decision makers manage the combinatorial explosion in scenario development for decision making under uncertainty. The first assumption is that the decision makers try to undertake ‘robust’ actions. For the decision maker a robust action is an action that has sufficiently good results whatever the events are. We examine the psychological as well as the theoretical problems raised by the notion of robustness. Finally, we address the false feeling of decision makers who talk of ‘risk control’. We argue that ‘risk control’ results from the thinking that one can postpone action after nature moves. This ‘action postponement’ amounts to changing look-ahead reasoning into diagnosis. We illustrate these ideas in the framework of software development and examine some possible implications for requirements analysis.     

一种基于两两组合测试的FIT框架扩展

优秀的软件开发过程需要协作和沟通,集成测试框架FIT(Framework for Integrated Test)就是一个用于增强交流和协作的工具.FIT创建了一个在客户和程序员之间的反馈循环.在集成测试框架FIT的基础上,研究软件测试用例的选择和生成,并结合两两组合覆盖的测试用例自动生成技术,实现了从接口参数边界值的确定,到以HMTL形式显示集成测试结果的半自动化过程,从而提高软件测试的质量和效率.     

Design of test inputs and their sequences in multi-function system testing

This discussion paper addresses combinatorial models in system testing from the perspective of system usage (utilization) and corresponding examination of system functions and their groups. Thus the following aspects of multi-function system testing are under study: analysis of system requirements and revelation of atomic system functions and their relationships, analysis of system function groups (clusters), design of the most important test inputs and sequences of the test inputs. The basic combinatorial problem is: composition of the best (the most important) test input(s) for each group of atomic system functions. Additional combinatorial problems are the following: (a) design of test input sequence for a trail (chain) of function clusters, (b) design of collection of test input sequences as covering of function cluster digraph, (c) structural fusion of unit test results. Numerical and real world examples illustrate the proposed approach. Mark Sh. Levin received the M.S. degree in Radioengineering from Moscow Techn. Univ. for Communication and Informatics (1970), the M.S. degree in Mathematics from ‘Lomonosov’ Moscow State Univ. (1975), the Ph.D. degree in Systems Analysis and Combinatorial Optimization from Inst. for System Analysis of Russian Acad. of Sci. (1982). His research interests include systems engineering, decision making, combinatorial optimization, and applications. Currently Dr. Levin is with Inst. for Inform. Transmission Problems of Russian Acad. of Sci. (Moscow) as a Senior Research Scientist. He is a member of ACM, IEEE, SIAM, Int. Soc. on MCDM, Int. Soc. of Appl. Intel., and OR Society of Israel. Mark Last received his Ph.D. degree in Industrial Engineering from Tel Aviv Univ. (2000). Dr. Last is a faculty member at Dept. of Information Systems Engineering of Ben-Gurion Univ. (Israel) and affiliated as a visiting faculty at National Inst. for Systems Test and Productivity (Univ. of South Florida, USA). His research interests include information systems, knowledge discovery and data mining, fuzzy sets and fuzzy logic, and software testing. He is a member of IEEE Computer Society and ACM.     

基于某型控制盒单板测试仪通用测试系统的研究

为了满足控制盒单板测试通用化、模块化的要求,在自动测试系统的设计中采用了虚拟仪器测控技术,提出新的设计方案;讨论了自动测试系统的功能、硬件配置、软件模块化设计、软件流程和工作原理,设计开发了基于虚拟仪器技术的通用测试系统。该系统容错能力强,通用性好,操作方便,具有良好的扩展性,对提高武器系统的可维护性和可靠性有重要的意义。     

Desktop haptic virtual assembly using physically based modelling

This research investigates the feasibility of using a desktop haptic virtual environment as a design tool for evaluating assembly operations. Bringing virtual reality characteristics to the desktop, such as stereo vision, further promotes the use of this technology into the every day engineering design process. In creating such a system, the affordability and availability of hardware/software tools is taken into consideration. The resulting application combines several software packages including VR Juggler, open dynamics engine (ODE)/open physics abstraction layer (OPAL), OpenHaptics, and OpenGL/GLM/GLUT libraries to explore the benefits and limitations of combining haptics with physically based modelling. The equipment used to display stereo graphics includes a Stereographics emitter, Crystal Eyes shutter glasses, and a high refresh rate CRT Monitor. One or two-handed force feedback is obtained from various PHANTOM haptic devices from SensAble Technologies Inc. The application’s ability to handle complex part interactions is tested using two different computer systems, which approximate the higher and lower end of a typical engineer’s workstation. Different test scenarios are analyzed and results presented.     

Metrics for BPEL process context-independency analysis

BPEL processes are workflow-oriented composite services for service-oriented solutions. Rapidly changing environment and turbulent market conditions require flexible BPEL processes to adapt with several modifications during their life cycles. Such adaptability and flexibility require the low degree of dependency or coupling between a BPEL process and its surrounding environment. In fact, heavy coupling and context dependency with partners provoke several undesirable drawbacks such as poor understandability, inflexibility, inadaptability, and defects. This paper is to propose metrics at the design phase to measure BPEL process context independency. With the aid of these metrics, the architect could analyze and control the context independency of a BPEL process quantitatively. To validate the metrics, authors collected a data set consisting 70 BPEL processes and also gathered the expert’s rating of context independency through conducting a controlled experiment. The obtained results reveal that there exists a high statistical correlation between the proposed metrics and the expert’s judgment of context independency.     

Adaptive Performance Management for SMS Systems

We present a design for performance management of SMS systems. The design takes as input the administrator’s performance objectives, which can be adjusted at run-time. Based on these objectives, the design takes the necessary actions to achieve them and it dynamically adapts to changing networking conditions. It does so by periodically solving a linear optimization problem that computes a new configuration for the SMS system. We have evaluated the design through extensive simulations in various scenarios using traces from a production SMS system. It has proved effective in achieving the administrator’s performance objectives, and efficient in terms of computational cost. Our experiments also show that the design is adaptive, i.e., it effectively adapts the systems’s configuration to changes in the networking conditions, in order to continuously meet the performance objectives. Finally, the feasibility of our design is proved through the development of a prototype on a commercial SMS platform.     

一个支持软件需求测试任务分析方法的工具

需求定义模型测试的任务分析方法，把需求情形实例当作需求模型的测试用例，用任务活动序列描述情形，并提供给任务分析专家和用户进行对照分析。情形实例的活动序列描述以文字的形式描述系统行为的细节，表达清晰，较需求模型更容易被用户所理解。文章介绍支持该方法的一个需求模型测试自动化工具ＳＲＤＭＴｅｓｔ．     

Converting scenarios to CSP traces with Mise en Scene for requirements-based programming

The Requirements-to-Design-to-Code (R2D2C) project of NASA’s Software Engineering Laboratory is based on inferring a formal specification expressed in Communicating Sequential Processes (CSP) from system requirements supplied in the form of CSP traces. The traces, in turn, are to be derived from scenarios, a user-friendly medium used to describe the required behavior of computer systems under development. An extensive survey of the “scenario” concept and an overview of scenario-based approaches to system engineering are presented. This work, called Mise en Scene, defines a new scenario medium (scenario notation language, SNL) suitable for control-dominated systems, coupled with a two-stage process for automatic translation of scenarios to a new trace medium (trace notation language, TNL), which encompasses CSP traces. Notes on progress toward a “smart” scenario authoring tool are provided, as well as a detailed case study. This work was originally presented at the Software Engineering Workshop (SEW-31) in March 2007. It was supported by research grants from Canada’s Natural Sciences and Engineering Research Council (NSERC).     

基于顺序图的FADEC软件需求状态图模型验证

基于模型的系统工程MBSE方法论在全权限数字电子控制(FADEC)软件开发领域的应用越来越普遍,模型作为开发过程中传递信息的介质,其在软件开发过程中的重要性不言而喻.完善的需求模型可以降低FADEC软件研制过程中的风险,提高研制效率.以SysML顺序图为测试用例,对模型进行测试.通过测试用例与需求的追踪关系,保证了测试的有效性.测试用例生成过程中利用了ATG用例自动生成技术.测试结果表明了此方法的实用性,实现了对需求模型的充分测试.     

一个实用的软件质量评估模型

定义了一个实用软件质量评估模型,该模型包括两维：软件产品类型维和软件开发阶段维。该模型用于软件产品需求的定义、软件产品测试目标和计划的定义以及软件产品质量的评估。