Context. Non-functional requirements (NFRs) are important aspects which directly or indirectly determine whether a product is a success or a failure. It becomes essential to incorporate and understand them, before the software product enters the development phase. Despite the increasing emphasis put into NFR studies (namely; models and frameworks etc.) over the past few years, most industries prefer not use these techniques or to deal with NFRs later, in simpler manners. This could limit the efficiency of the development process. Integration of the existing NFR models/frameworks into the earlier phases of the product life cycle can provide a systematic approach to plan and anticipate NFRs for any software product. Objectives. This study aims to provide a generic meta model which acts as a compilation of the best NFR models/frameworks integrated into the early phases of the software product life cycle. This study also provides a real-world example which applies the conceptual meta model. Lastly, the meta model undergoes some limited validation to determine its relevance to what is being used and the extent of its practical use. Methods. Initially, a systematic literature review (snowballing) was conducted, to identify the different types of NFR models/frameworks. A comparative pro-con analysis was performed on the results of the SLR, which was the basis of the inclusion criteria for the meta model. The conceptual meta model was developed based on the International Software Product Management Association’s (ISPMA) definition of a product life cycle. Each phase of this meta model was imbedded with an NFR model/framework associated to the purpose of that phase and the results from the SLR. The application of the meta model was then demonstrated using a mobile phone example. Finally, the meta model was validated limitedly via an exploratory survey and the results were analyzed. Results. The meta model introduced can be used for the constructive inclusion of NFRs from product inception to product development. All phases required for the fulfillment of an NFR, are included. The overall positive feedback of the meta model is at 67%. Validations and assessments by practitioners helped determine to some extent that some industries are open to using the approach. Keeping in mind most of the available models on NFRs have not been validated, the NFR works used in this research have gone through some preliminary validation in this study. Conclusions. The study promotes the use of NFR models in the early phases of the software product life cycle. Some of the best modeling techniques were included based on results of literature analysis and their capability to fit into each phase. This study also analyzed the various insights of practitioners and researchers, justifying the significance of modeling, and the proposed technique. Possible extensions to this research were also mentioned.
Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:bth-13986 |
Date | January 2017 |
Creators | Nanduru, Pavan Kumar |
Publisher | Blekinge Tekniska Högskola, Institutionen för programvaruteknik |
Source Sets | DiVA Archive at Upsalla University |
Language | English |
Detected Language | English |
Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
Format | application/pdf |
Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.1956 seconds