Constructive complexity: a metric for operations research spreadsheet model design

Abstract

This article makes a modest step toward a quantitative measure of the work required to solve a specific kind of problem: a given spreadsheet design. The method relies on a measure that we define called constructive complexity, which is the number of keystrokes required to write a spreadsheet for a given computational task, as a function of the input data. The notion of constructive complexity is simple but general. The measure is simple enough to be described briefly in the classroom and the textbook. As an example, we use the shortest path problem. The objective is to educate operations research students and practitioners to write better spreadsheets, and to evaluate the quality of different spreadsheet designs. Constructive complexity could be used as a tool to achieve these objectives. We do not claim that it resolves all problems of spreadsheet design, but it is one step forward in quantifying spreadsheet design, and is intended to generate additional debate.