5.1 Product complexity affects project management techniques

Project complexity can be defined in many ways. In a classical sense, it is related to the quantity of code developed or the number of people assigned to the project. Complexity is further increased through the use of multiple, diverse technologies that are necessary in creating complex modern systems today.

Contemporary work addressing software development has examined the notion of complexity from a number of different angles. The following is presented by Brooks [1] and corroborated by McConnell [2]:

• The effort expended (man months) on a project is nonlinear with respect to code size (Figure 5.1);

Figure 5.1: Effort expended on a typical software project is nonlinear as the size of the project increases.

• The actual programming rate decreases as code size increases, exceeding the time allocated to completing the project and resulting in a greater amount of code being generated (Figure 5.2);

Figure 5.2: As software increases in size, the actual rate at which the software solution is developed is slower than predicted.

• The actual debug rate is significantly slower than what is generally estimated (Figure 5.3).

Figure 5.3: The software is debugged at a far slower rate than is typically predicted, which is forecast as an S -shaped curve. Note also that with the passage of time, the amount of software developed tends to increase beyond the amount predicted (the maximum value of the predicted debugging rate).

Armed with this information, what should the goal of development be?

• To...