The Influence Factors and Improvement of the Accuracy of Software Function Point Count

• Main Authors: Yi-Lin Hu, 胡衣臨
• Other Authors: Sun-Jen Huang
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/00555521243155348941

碩士 === 國立臺灣科技大學 === 資訊管理系 === 93 === How to precisely count software function points is a key to developing the software project planning and accordingly monitoring the project progress. The past research on the accuracy of software function point counts has only addressed on the variation of function point counts. However, hardly any research to date has investigated the factors affecting the accuracy of function point counts and what identification rules are frequently misjudged and the reasons to cause misjudgment. Hence, the objective of this thesis is to explore how the factors of software development technique, software lifecycle stage, counter’s education background, software development experience, and education training of function point analysis method affect the accuracies of function point counts and also to find the solutions to avoid misjudgment of the identification rules of function point counts. By deploying the case study of software function point counts and semi-structured interview with full-time and on-job master students at Department of Information Management, National Taiwan University of Science and Technology, the analysis result from 41 samples and 7 interviews indicates that the education training of function point analysis method has a significant influence on the accuracy of software function point count, but the factors of software development technique, software lifecycle stage, counter’s education background and software development experience do not have significant influences. This thesis also summarizes those frequently misjudged identification rules and the reasons to cause these misjudgments, and accordingly proposes the enhanced interpretations and implication guidelines for these rules. Base on the above findings, we consider the important factors affecting the accuracy of software function point counts are the completeness of software function requirement specification, the counter’s proficiency to function point analysis method and the domain knowledge in the developed software system.