Measure Size , Complexity of Algorithms Using Function Points

February 2001 An algorithm is a set of equations that is executed in a logical sequence to produce an external output. In the fields of function point analysis and operations research, an algorithm can be seen as a critical tool to reduce effort needed to solve a complex series of calculations. We have written this paper focusing on intermediate function point analysis theory, and use linear programming to exemplify our points. Some readers may want to refer to the Definition of Terms at the end of this paper, or to [1, 2, and 3] for detailed references describing function point analysis, linear programming, and operations research, respectively. An algorithm we counted recently was one used to compute pay. This required solving a set of equations to calculate such pay subcomponents as base pay rate, holiday pay, temporary duty (travel) expenses, and foreign exchange rates. All subtotals were added to yield the final pay amount. Using the method of the International Function Point Users Group (IFPUG), beginning function point counters would probably recognize the input screen needed to enter a traveler’s pay/expense data and count six or fewer function points. They would probably recognize the resulting earnings/expense statement as an external output and count it as seven or fewer function points. However, they may overlook other, more substantial functionality inherent in this algorithm. Measure Size, Complexity of Algorithms Using Function Points