Problem Based Estimations

Lines of code and function points were define as basic measures from which productivity metrics can be computed.  The LOC and the FP data are used in 2 ways during software project estimation that are (1)   as an estimation variable which is used   to size each element of the software and (2) as baseline metrics collected from previous projects and used in conjunction with estimation variables to develop effort and cost projections.

The LOC and FP estimation are distinct estimation methods since both have a number of characteristics in common like the project planner starts with a   bounded statement of software scope and from this statement attempts to decompose software into problem functions which can each be estimated individually. The LOC or FP the estimation variable is then estimated for each function.  Otherwise, the planner may choose another component for sizing like as objects or classes, changes, or business processes impacted. Baseline productivity metrics LOC or pm and FP or pm are then applied to the appropriate estimation variable and cost or effort for the function is derived.  The Function estimates are combined to give an overall estimate for the entire project.

It is very important   to note that there   is often substantial scatter in productivity metrics for a company which is making the use of a single baseline productivity metric suspect.  In normally LOC or pm and FP or pm averages should be computed through project domain.  Which are projects should be grouped through team size, complexity, application area, and other applicable parameters.  The Local domain averages should then be computed.  It should first be allocated to a domain when a new project is estimated and then   the appropriate domain average for productivity should be used in generating the estimate.

LOC and FP estimation methods differ in the stages of detail needed for decomposition and the goal of the partitioning. The LOC is used as the estimation variable; decomposition is absolutely essential and is often taken to considerable levels of detail. This is greater and a degree of partitioning the more likely which reasonably accurate estimates of LOC can be established.

For FP estimates decomposition works variously.   After than focusing on function, each of the information domain characteristics like data files, inquiries, inputs, outputs, and external interfaces and the 14 complexity adjustment values are estimated. Resultant estimates are used to derive a FP value which can be tied to past data and used to estimate and generate.

Regardless of the estimation variable which is use the project planner start through estimating a range of values for each function or   information domain value which is using historical data or when all else fail intuition, the planner estimates an optimistic most likely and pessimistic size value for each function or count for each information domain value.  With an implicit indication of the degree of uncertainty is provided.

When the range of values is specified.

A 3 points or expected value is then computer.  The expected value for the estimation variable size EV can be computed like a weighted average of the optimistic (sopt), most likely(S m)   and pessimistic (ppess) estimates for example,

EV = (sopt+4sm+spess)/6 ----------(5.1)

Follows a beta probability distribution and Gives heaviest credence to the most likely estimate and.

We will assume that there is a very small probability where the actual size result will fall outside the optimistic or pessimistic values. By using  standard  statistical  techniques  we  can  compute  the  deviation  of  the estimates. Thus, it should be noted that a deviation based on uncertain (estimated) data must be used judiciously.

LOC or FP productivity data are applied once expected value for the estimation variable has been determined historical.  Are the estimates correct?  The only reasonable answer to this question is that we cannot be sure.  With any estimation method no matter how sophisticated it is must be cross checked with other approach. Even then common sense and experience must prevail.