Function Point Analysis (FPA) is a widely used technique for measuring software size. It measures software functionality from the user's perspective, usually based on a requirements description. In many software processes, these requirements are represented by UML models. Although there have been attempts to automate the measurement process, FPA counting requires a considerable amount of interpretation which, to be reliable, should be made by experts. On the other hand, fully manual counting methods usually fail to keep synchronized with the requirements model, since requirements frequently change during the development cycle. This paper describes an approach for counting FPA and a compliant tool. This approach makes use of UML requirement models. The tool, called ReMoFP (Requirement Model Function Point counter), leaves all the counting decisions to the analyst, but supports him by ensuring consistency with the requirements represented in the models. The ReMoFP was developed by a software development laboratory in Brazil, and helped it to improve counting productivity, consistency, and maintainability. 1. Introduction Due to complexity and size issues, developing high-quality, cost-effective software is a huge challenge, usually constrained by short schedules. To achieve effective and realistic planning for software development projects, project managers should estimate correctly many software metrics, such as those that reflect size, effort, quality, and risks. Size estimates are used both as input and normalization factor for other relevant metrics. Several well-known methods are available for size measurement; counting LOC (Lines of Code) and Function Point Analysis (FPA) are widely used. This work focuses on FPA, a method that estimates functional complexity, seen from the user viewpoint [1]. Early approaches to FPA were proposed by Albrecht [2]; it has become widely used, despite some weaknesses [3–5]. The method has been maintained and updated by IFPUG (the International Function Point Users Group) [1], and was adopted as an ISO and IEEE standard [6]. FPA details are given in many sources, and we omit them, for the sake of brevity. In recent years, Object-Oriented (OO) technology has emerged as a dominant practice in Software Engineering domain. Therefore, it is useful to match traditional Function Point (FP) measurement to new OO approaches [7], including models based on Unified Modeling Language (UML) [8]. In this work, we defined a tool, called ReMoFP (Requirement Model Function Point counter), to help FP experts during FPA counting based on
References
[1]
IFPUG, IFPUG Counting Practices Manual—Release. 4.2.1, International Function Point Users Group, Westerville, Ohio, USA, 2005.
[2]
A. J. Albrecht, Function Point Analysis, John Wiley & Sons, New York, NY, USA, 1994.
[3]
B. Kitchenham, “Counterpoint: the problem with function points,” IEEE Software, vol. 14, no. 2, p. 29, 1997.
[4]
T. Uemura, S. Kusumoto, and K. Inoue, “Function point measurement tool for UML design specification,” in Proceedings of the 6th International Software Metrics Symposium, pp. 62–69, November 1999.
[5]
D. Gupta, S. J. Kaushal, and M. Sadiq, “Software estimation tool based on three -layer model for software engineering metrics,” in Proceedings of the 4th IEEE International Conference on Management of Innovation and Technology (ICMIT '08), pp. 623–628, September 2008.
[6]
IEEE, “IEEE std 14143.1-2000: implementation note for IEEE adoption of ISO/IEC 14143-1:1998 Information Technology—Software Measurement—Functional Size Measurement—part 1: definition of concepts,” Tech. Rep., Software Engineering, IEEE, New York, NY, USA, 2003.
[7]
J. Ram and S. V. G. K. Raju, “Object oriented design function points,” in Proceedings of the 1st Asia-Pacific Conference on Quality Software (APAQS '00), pp. 121–126, 2000.
[8]
OMG, Unified Modeling Language Specification, Version 2.0, Object Management Group (OMG), 2005.
[9]
V. Harput, H. Kaindl, and S. Kramer, “Extending function point analysis to object-oriented requirements specifications,” in Proceedings of the 11th IEEE International Software Metrics Symposium (METRICS '05), pp. 361–370.
[10]
B. Pimentel, W. P. P. Filho, C. Pádua, and F. T. Machado, “Synergia: a software engineering laboratory to bridge the gap between university and industry,” in Proceedings of the International Workshop on Summit on Software Engineering Education, pp. 21–24, ACM, New York, NY, USA, 2006.
[11]
W. de Pádua, “Quality gates in use-case driven development,” in Proceedings of the International Workshop on Software Quality (WoSQ '06), pp. 33–38, ACM Press, New York, NY, USA, 2006.
[12]
W. de Pádua, Engenharia de Software: Fundamentos, Métodos e Padr？es, LTC, 3rd edition, 2008.
[13]
P. Kruchten, The Rational Unified Process: An Introduction, Addison-Wesley, Reading, Mass, USA, 3rd edition, 2003.
[14]
IEEE, “IEEE recommended practice for software requirements specifications,” Tech. Rep. IEEE Std 830-1998, 1998.
[15]
OMG, Object Constraint Language 2.0 Specification, Object Management Group (OMG), 2005.
[16]
G. Caldiera, G. Antoniol, R. Fiutem, and C. Lokan, “Definition and experimental evaluation of function points for object-oriented systems,” in Proceedings of the 5th International Software Metrics Symposium, pp. 167–178, November 1998.
[17]
G. Cantone, D. Pace, and G. Calavaro, “Applying function point to unified modeling language: conversion model and pilot study,” in Proceedings of the 10th International Symposium on Software Metrics (METRICS '04), pp. 280–291, September 2004.
[18]
B. W. Boehm, C. Abts, A. W. Brown, et al., Software Cost Estimation with COCOMO II, Prentice Hall, New York, NY, USA, 2000.
