EFFORT ESTIMATION OF COMPONENT BASED SOFTWARE DEVELOPMENT

Abstract

In this era, no one denies the importance of software reuse because software systems are growing and becoming complex with every passing day. Component Based Software Development (CBSD) emerged as a software creation approach with the concept of reusability. In this approach, Software Components which are common among different software applications are reused rather than being written from scratch for every application. CBSD technique is of keen interest to researchers and practitioners as they hold promise to support the timely and cost effective development of large-scale complex systems. It is becoming imperative that effort involved in CBSD may be accurately estimated to attain maximum benefits of the approach. Effort estimation is one of the major tasks in software project management. The literature shows several efforts estimation models of CBSD but each model does have their own pros and cons. Furthermore, different effort estimation models primarily focuses on the efforts involved in component’s integration activities. Moreover, all phases of CBSD lifecycle are unaddressed by existing effort estimation models. Thus, the need to estimate effort involved in CBSD lifecycle is an ongoing challenge. In this research focus is on the effort estimation of CBSD lifecycle with the help of Fuzzy Logic approach. For the purpose, it was necessary to have a comprehensive CBSD lifecycle model which can be made the basis of effort estimation in CBSD. Thus, first in this study a Circular Process Model (CPM) for CBSD lifecycle is proposed. CPM contains the strengths and weaknesses of the existing CBSD lifecycle models with the focus on rejuvenation of one phase in subsequent phases of the lifecycle. CPM is also validated using the Process Quality Measurement Model (PQMM) [19] and by comparing with the existing CBSD lifecycle process model of Hazleen Iris et al [13]. Then, effort estimation model for CBSD lifecycle is proposed on the basis of CPM. The proposed effort estimation models is also implemented and validated with the help of a case study. Fuzzy logic is used in the implementation as it is more appropriate when the systems are not suitable for analysis by conventional approach or when the available data is uncertain, inaccurate or vague. V Table of Contents