Measuring the Functional Size of Real-Time and Embedded Software: a Comparison of Function Point Analysis and COSMIC

The most widely used methods and tools for estimating the cost of software development require that the functional size of the program to be developed be measured, either in \u201ctraditional\u201d Function Points or in COSMIC Function Points. The latter were proposed to solve some shortcomings of the former, including not being well suited for representing the functionality of real-time and embedded software. However, little evidence exists to support the claim that COSMIC Function Points are better suited than traditional Function Points for the measurement of real-time and embedded applications. Our goal is to compare how well the two methods can be used in functional measurement of real-time and embedded systems. We applied both measurement methods to a number of situations that occur quite often in real-time and embedded software. Our results seem to indicate that, overall, COSMIC Function Points are better suited than traditional Function Points for measuring characteristic features of real-time and embedded systems. Our results also provide practitioners with useful indications about the pros and cons of functional size measurement methods when confronted with specific features of real-time and embedded software