A Goal-Oriented Software Testing Methodology

JUITA: Jurnal Informatika

The Goal-Oriented Requirements Engineering (GORE) is one approach that is widely used for the early stages of software development. This method continues to develop in the last three decades. In this paper, a literature study is conducted to determine the GORE state of the art. The study begins with a Systematic Literature Review (SLR) was conducted to determine the research trend in the last five years. This study reviewed 126 papers published from 2016 to 2020. The research continues with the author's search for scientific articles about GORE. There are 26 authors who actively publish GORE research results. Twenty-six authors were grouped into seven groups based on their relation or co-authoring scientific articles. An in-depth study of each group resulted in a holistic mapping of GORE research. Based on the analysis, it is known that most research focuses on improving GORE for an automated and reliable RE process, developing new models/frameworks/methods originating from GO...

International Journal of Intelligent Systems, 2010

In this paper, we introduce the application of rigorous analysis procedures to goal models in order to provide several benefits beyond the initial act of modelling. Such analysis can allow modellers to assess the satisfaction of goals, facilitate evaluation of high-level design alternatives, help analysts decide on the high-level requirements and design of the system, test the sanity of a model, and support communication and learning. The analysis of goal models can be done in very different ways depending on the nature of the model and the purpose of the analysis. In our work, we use the Goal-oriented Requirement Language (GRL), which is part of the User Requirements Notation (URN). URN, a new Recommendation of the International Telecommunications Union, provides the first standard goal-oriented language. Using GRL, we develop an approach to analysis that can be done by evaluating qualitative or quantitative satisfaction levels of the actors and intentional elements (e.g., goals and tasks) composing the model. Initial satisfaction levels for some of the intentional elements are provided in a strategy and then propagated to the other intentional elements of the model through the various links that connect them. The results allow for an assessment of the relative effectiveness of design alternatives at the requirements level.

Fourth International Workshop on Comparative Evaluation in Requirements Engineering (CERE'06 - RE'06 Workshop), 2006

As an attempt to answer the need for methods and tools in requirements engineering (RE) which are domain specific and can address the main RE objectives (REOs), and the growing interest in the goal oriented requirements engineering (GORE) approach that overcomes the inadequacy of the traditional systems analysis approaches, we systematically evaluate the KAOS method, and the Objectiver tool, using the major REOs widely accepted as being important attributes of requirements specifications. In addition, we examine whether KAOS and Objectiver meet their own selfdefined objectives. We use two target problems as a basis for the evaluation. The result of the target problems is raw data consisting of error reports and observations that support the evaluator's judgment. The evaluation itself is qualitative, not a statistical experimental evaluation. Its result will help to answer the research questions: (i) How well do KAOS and Objectiver meet the criteria established in the discipline of RE; and (ii) How well do KAOS and Objectiver achieve their own self-defined objectives.

Requirements Engineering, 2002

The study of contemporary Requirements Engineering (RE) methodologies indicates that modeling of organizational goals constitutes a central activity of the RE process. In particular, goals provide the rationale and drive the elaboration of the requirements that operationalise them. They also provide the criteria against which the completeness and correctness of the requirements specification is validated. In other words, requirements implement goals in the same way that programs implement design specifications. Despite the significance of goals in RE, research in the field is fragmented. No research has so far taken place in order to define the overall role that goals play in RE. This paper puts forward a unifying view of goal analysis in the context of RE. This allows the identification of similarities and differences between the different conceptions of goal used by different approaches and promotes the understanding of the overall role of goal analysis in RE. Based on this understanding the various approaches can be put together thus leading to a stronger goal-driven RE framework that takes advantage of the contributions from the many streams of goal-oriented research.

2006

Abstract→ We briefly review the history and key ideas in Goal-Oriented Requirements Engineering research. We then sketch two applications of these ideas. The first involves establishing an Agent-Oriented Software Development method called Tropos which covers not only requirements but also design phases. The second addresses the design of high-variability software for applications such as home care software and business process design.

1999

The study of contemporary requirements engineering (RE) methodologies indicates that modelling of organisational goals constitutes a central activity of the RE process. In particular, goals provide the rationale and drive the elaboration of the requirements that operationalise them. They also provide the criteria against which the completeness and correctness of the requirements specification is validated. In other words, requirements implement goals in the same way that programs implement design specifications. Despite the significance of goals in RE, research in the field is fragmented. No research has so far taken place in order to define the overall role that goals play in RE. This paper puts forward a unifying view of goal analysis in the context of RE. This allows the identification of similarities and differences between the different conceptions of goal used by different approaches and promotes the understanding of the overall role of goal analysis in RE. Based on this understanding the various approaches can be put together, thus leading to a stronger goal-driven RE framework that takes advantage of the contributions from the many streams of goaloriented research.

The term " Goal " is increasingly being used in Requirement Engineering. Goal-Oriented requirement engineering (GORE) provides an incremental approach for elicitation, analysis, elaboration & refinement, specification and modeling of requirements. Various Goal Oriented Requirement Engineering (GORE) methods exist for these requirement engineering processes like KAOS, GBRAM etc. GORE techniques are based on certain underlying concepts and principles. This paper presents and synthesizes the underlying concepts of GORE with respect to coverage of requirement engineering activities. The advantages of GORE claimed in the literature are presented. This paper evaluates GORE techniques on the basis of concepts, process and claimed advantages.

21st IEEE/ACM International Conference on Automated Software Engineering (ASE'06), 2006

This paper presents a specification-based approach and implementation architecture that addresses several known challenges including false positives and domain knowledge errors. Our approach begins with a system goal graph and functional goal plans. Source code is annotated with goals from plans the program is attempting to achieve; code is then precompiled to emit annotations at run time. Plans are automatically translated into a rule-based recognizer. An oracle is produced from the pre-and post-conditions associated with the plan's goals. When the program is executed, goals and events are emitted and automatically tested against plans and expected results. This allows more efficient testing, including better recognition of false positives-correct results not matching plans-and domain knowledge errors-incorrect results from following intended plans. The concept is demonstrated for a small example and a larger publicly available case study in which we found a mismatch between stated requirements and actual program behavior.

Goal-Oriented Requirements Engineering (GORE) methodologies are gaining prominence because of their ability to overcome the drawbacks of traditional software development techniques. The aim objective of this paper is to know the intricacies of existing GORE literature which will help us in identifying specific areas of scope for improvement. We discuss the set of generic tasks which are part of a GORE methodology. These tasks include goal elicitation, goal analysis, goal refinement and goal validation. Numerous methods have been proposed under these tasks. We also make an attempt to classify different methods proposed under GORE literature based on certain parameters or criteria. The report concludes by identifying gaps in specific areas of GORE literature and suggests possible ways to enhance the effectiveness of this methodology.

2005

We describe a new verification and validation (V&V) approach based on comparing actual system behavior in the form of captured goalannotated event traces, with expected behavior expressed by requirements scenarios with system requirements goals. We believe our proposed V&V approach can leverage requirements engineering work fruitfully and lead to improved software quality, by offering these six potential benefits: (1) higheryield testing, (2) distinguishing false positives, (3) defining test coverage metrics, (4) detecting domain-analysis errors, (5) validating top-level requirements, and (6) efficient control of retesting. We use examples to explain how our method can attain these benefits. If our goal/requirements-based V&V approach succeeds in realizing these six benefits, we believe it will lead to improved requirements that, in turn, can successfully improve software quality. top-level goal: find lcm(6, 4) 1. find lcm of 6 and 4 1.1 find the product p of 6 and 4 * p = 24 1.2 find g, the gcd of 6 and 4 1.2.1 second member of pair (6, 4) is not zero * 4 ≠ 0 1.2.2 replace the pair (m, n) with (n, m mod n) * (m, n) replaced by (4, 2) 1.2.3 second member of pair (4, 2) is not zero * 2 ≠ 0 1.2.4 replace the pair (m, n) with (n, m mod n) * (m, n) replaced by (2, 0) 1.2.5 second member of pair (2, 0) is zero * 0 = 0 1.2.6 return first member of pair (2, 0) as the gcd * 2 returned as value of gcd(6, 4) 1.2.7 assign value 2 = gcd(6, 4) to be value of g * g = 2 1.3 return p / g as the lcm of 6 and 4 * return 24 / 2 * 12 returned as the value of lcm(6, 4) New game began at 2:21:46 PM Machine plays first with X * Machine moved with X in upper left corner (1,1) * You moved with O in lower right corner (3,3) * Machine moved with X in lower left corner (3,1) * You moved with O on left side (2,1) * Machine moved with X in upper right corner (1,3) * You moved with O in center (2,2) * Machine moved with X on top side (1,2) Result: X won the game Transcript ended at 2:22:16 New game began at 2:21:46 PM Machine plays first with X at Novice level Goal: to play with variety make a random first move * Machine moved with X in upper left corner (1,1) * You moved with O in lower right corner (3,3) Goal: make a random move * Machine moved with X in lower left corner (3,1) * You moved with O on left side (2,1) Goal: make a random move * Machine moved with X in upper right corner (1,3) * You moved with O in center (2,2) Goal: make a winning move * Machine moved with X on top side (1,2) Result: X won the game Transcript ended at 2:22:16 PM New game began at 2:21:46 PM Machine plays first with X at Expert level Goal: to play with variety make random first move * Machine moved with X in upper left corner (1,1) * You moved with O in lower right corner (3,3) Goal: make forcing move that will lead to a fork and a win * Machine moved with X in lower left corner (3,1) * You moved with O on left side (2,1) Goal: create a fork * Machine moved with X in upper right corner (1,3) * You moved with O in center (2,2) Goal: make winning move * Machine moved with X on top side (1,2) Result: X won the game Transcript ended at 2:22:16 PM