Question

Two algorithmic methods of Software Cost Estimation are Walston & Felix and Bailey
& Basili. compare and contrast these two methods.

Answer 1

Hi,
Please find the answer below:

Both the methods are based on estimate variable size in LOC or FP. The estimate models take the mathematical equation in the form of

E = a + b* <size-variable> ^c

where a , b and c are constants.

size-estimate-variable is LOC or FP

E is effort in man-months.

Walston & Felix method:
This model is an example of single variable model.
The equation is calculated based upon variety of projects data that fitted based on the below equation

E = 5.2*KLOC^0.91

The standard error rate of using this equation to estimate depends on several factors.
For example, this model might not be a good estimator for projects that reuse the code.
Measuring LOC itself is error prone.

For example,
effort for (10 lines of source code ) is different from effort for (writing 10 lines of doc comments)

Bailey & Basili method:
This model is an example of met model. This model may achieve more accurate results which takes
past performance of projects within local environment.This model requires to compute baseline equation -> analyze factors that reflect local environment -> model the new project estimation

E = 5.5 + 0.73*KLOC^1.16

The major drawback of this approach is that you need deep understanding of the local environment and history of past projects for accurate estimation of new projects.

Hope this helps.