A knowledge audit model for requirement elicitation process

Knowledge plays an important role in the success of any organization. Software development organizations are highly knowledge-intensive organizations especially in their requirement elicitation process (REP). There are several problems regarding communicating and using the knowledge in REP such as misunderstanding, being out of scope, conflicting information and changes of requirements. Because there are different people involved in REP and these people are from different backgrounds have different knowledge. Different areas of knowledge lead to different ways of knowledge expression and negatively affect knowledge understandability and completeness. Ambiguity in knowledge results in incorrect interpretation of knowledge and requirements. To allay these problems, it is necessary to identify and assess the knowledge. Knowledge Audit (KA) is the process of knowledge analysis and assessment that aims to answer these questions: what knowledge exists and what knowledge is missing as well as assess the knowledge in terms of completeness, correctness and understandability. Since there is a lack of KA model and knowledge assessment in REP, this research tends to fill this gap. Therefore, this research introduces a knowledge audit (KA) model to support knowledge communication of people who are clients and developers in REP. A survey on 40 respondents of clients and developers consisting of system analysts, system developers, project managers and clients during a focus group is conducted. KA is described through four main knowledge components for the model: 1) knowledge sources,2) requirements knowledge, 3) knowledge inventory, and 4) audited knowledge. The findings revealed confirmatory of the model after some refinements. The results have confirmed the identified KA components and their elements as well as the inter-relationship among them. This research also develops a prototype based on the proposed model to validate the model through post study. The findings of post study also confirmed the effectiveness of KA model for REP on the criteria of completeness, correctness, and understandability. The contribution of this research lies in the KA model that illustrates the KA components with the focus of knowledge assessment in REP to fill the exiting gap in this area. It also provides a prototype to assist software developers in REP. Overall, it introduces a KA model for REP to identify and assess knowledge which leads to the success in the process of requirement elicitation