Interested in improving your requirements engineering process? Try requirement patterns!
Xavier Franch, Cindy Guerlain, Cristina Palomares, Maria Carme Quer Bosor, Samuel Renault
Abstract
Xavier Franch, Cindy Guerlain, Cristina Palomares, Maria Carme Quer Bosor, Samuel Renault
Abstract
Requirement elicitation is the process of acquiring the system requirements from the system stakeholders. This process is critical in all software projects: if not all the requirements are elicited, or if some elicited requirements do not describe real stakeholder needs, or if the quality of the requirements is poor (e.g., they suffer from ambiguities), the chance of project failure increases. Techniques supporting requirements elicitation (interviews, meetings, storyboards...) are mostly oriented to obtain requirements from scratch and they may hardly take advantage of a fundamental observation: When specifying a system, it is quite usual that a significant proportion of requirements is recurrent and belongs to a relatively small number of categories, especially in the case of non-functional requirements. Our motivation is to consider this observation for improving the effectiveness of the requirement elicitation process. We are using the concept of software requirement pattern [1] (SRP). An SRP basically consists of a template that generates one or more requirements, and some information to identify its need in a particular project, and how it may be tailored to this project. The main benefits of using SRPs may be summarized as: 1) more effective requirement elicitation (requirements are not built from scratch; a process guides the engineer by giving advices, suggesting information, ...); 2) improved quality and consistency of requirements documents (by using a uniform style); 3) improved requirements management (e.g., clear traceability from requirements to goals).
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Requirement elicitation is the process of acquiring the system requirements from the system stakeholders. This process is critical in all software projects: if not all the requirements are elicited, or if some elicited requirements do not describe real stakeholder needs, or if the quality of the requirements is poor (e.g., they suffer from ambiguities), the chance of project failure increases. Techniques supporting requirements elicitation (interviews, meetings, storyboards...) are mostly oriented to obtain requirements from scratch and they may hardly take advantage of a fundamental observation: When specifying a system, it is quite usual that a significant proportion of requirements is recurrent and belongs to a relatively small number of categories, especially in the case of non-functional requirements. Our motivation is to consider this observation for improving the effectiveness of the requirement elicitation process. We are using the concept of software requirement pattern [1] (SRP). An SRP basically consists of a template that generates one or more requirements, and some information to identify its need in a particular project, and how it may be tailored to this project. The main benefits of using SRPs may be summarized as: 1) more effective requirement elicitation (requirements are not built from scratch; a process guides the engineer by giving advices, suggesting information, ...); 2) improved quality and consistency of requirements documents (by using a uniform style); 3) improved requirements management (e.g., clear traceability from requirements to goals).
Key concepts: Requirements elicitation, Requirements management, Requirements traceability, Requirements engineering, Requirement, Requirements analysis, Non-functional requirement, Computer science