J.-C. Liou (USA)
Software metrics, project management, estimation.
Among all kinds of measures, software size serves as a basis for other measures and plays a major role in better understanding the software project. For software size measuring, there are some methodologies reported in the literatures. These methodologies can be either simply counting the source lines of code (SLOC) or using metric methodology. The Function Point Analysis (FPA) is by far the most popular methodology for sizing software in industry. However, when applying the IFPUG standard to count the Object-Oriented software application, the resulting function points are insignificant or inaccurately low. In this paper, we present a methodology, called Object Point Analysis (OPA), for counting the object points for the Object-Oriented software application. The OPA has been used by some software projects for more than six years. The positive results received from these projects provide a preliminary study on sizing of OO software application. Comparing to the FPA, the OPA not only produces a more accurate size measurement for OO software applications, but also provides a means for sizing automation, which reduces the human errors often found in FPA.
Important Links:
Go Back
