Topic > Early identification, specification and measurement of…

1. INTRODUCTION Currently, system functional requirements (FUR) are the most focused aspect during the system requirements gathering phase, while non-functional requirements (NFR) are derived and extracted by system engineers at a general level. Those NFRs of the system, traditionally left to be managed later by developers in the design phase of the software system. Whereas, NFR plays a significant responsibility in software development, where the system behavior attributes depend on it, consequently, NFR must be taken into consideration high levels of requirements, therefore they have a significant impact on the whole system as well as the process estimation and benchmarking. In practice, there are no generic standards or models that describe how to specify and measure such system NFRs in a requirements reference model. Software system requirements are initially typically addressed at the system level, either as high-level system functional user requirements (FUR system) or as high-level system non-functional requirements (NFR system) [5-7]. System requirements typically need to be detailed, allocated, and implemented in the intact system as hardware, software, or a specific permutation of hardware and/or software. Figure 1 describes a system of mapping FUR and NFR to the FUR software.Fig. 1. Mapping the system FURs and NFRs to the software FURs [?]To further understand these requirements, the FURs will describe a desired function in the system and typically formulated with a subject and a predicate, while the NFR describes the behavior of the desired functions in the system and typically formulated with an adverb or a modifying clause. The latter, in the stages of the software requirements engineering process, the NFR system cannot...... half of the paper ......ms by Kaiya et al [24] of the existing system contributed its I work in identifying interested parties and their requirements. Recently, some have proposed a solution such as an NFR framework, which suggests a step towards reducing the impact of uncertainty due to NFRs, as proposed by Kassab et al [26]. Kassab et al also extend to another solution that can help in software effort estimation based on the incorporation of NFR and FUR basis [25]. Kassab et al., used the COSMIC standard functional size measurement. Furthermore, Kassab introduces initial work based on soft goals, in order to pave the way for the ability to solve the quantitative evaluation problem of NFR. The most recent, by Buglion, introduces suggested work that helps measure the productivity of NFR, as well as suggested more worthy estimates for projects based on the use of inconsistent NFR and UFR.[42].