Software Project Estimation Questions and Answers

Answer: C

The principle is the W5HH principle.

Barry Boehm's W5HH principle organises a project plan around seven questions: Why is the system being developed? What will be done? When will it be done? Who is responsible? Where are they located? How will the job be done technically and managerially? How much of each resource is needed? These cover objectives, milestones and schedules, responsibilities, approaches and resources.

'W3HH' and 'WHO' are not the names of Boehm's principle.

Answer: D

The technique is analogous estimation.

Analogous (top-down) estimation bases the effort, cost or duration of a new project on actual figures from similar past projects, adjusted for known differences. It is quick and useful early on, when detail is scarce.

The Delphi technique pools expert opinion through anonymous rounds until a consensus forms. COCOMO is a parametric model that computes effort from size (KLOC) and cost drivers. Its constants were fitted to historical data, but it is a formula, not a comparison with similar projects. Function point analysis measures software size from its functions.

Answer: A

The function point value is 612.06.

Unadjusted function points, then the value adjustment, with 4 factors at 0, 4 at 3 and the other 6 at 4:

\[ \begin{aligned} UFP &= 30(4) + 60(5) + 23(4) + 8(10) + 2(7) \\ &= 120 + 300 + 92 + 80 + 14 = 606 \\ \sum F &= 4(3) + 6(4) = 36 \\ VAF &= 0.65 + 0.01 \times 36 = 1.01 \\ FP &= 606 \times 1.01 = 612.06 \end{aligned} \]

606 is the unadjusted count, 1.01 the adjustment factor and 0.65 its constant, so none of them is the function point value.

Answer: B

The critical path is the sequence of tasks that determines the project's finish date.

In network scheduling (CPM or PERT), the critical path is the longest chain of dependent activities from start to finish. Its activities have zero float, so any delay on them delays the whole project, and the length of this path is the project duration.

The path's length gives the minimum project duration, but it is defined as the longest chain of tasks, not as the shortest time, which is why option A is the trap. It has nothing to do with the likelihood of failure or with budgets.

Answer: B

The key's answer is Poor requirements management.

Surveys of failed projects, such as the Standish Group reports behind figures like these, put incomplete, unclear and changing requirements at the top of the causes of cancellation and of systems that do not do what users expected. A project built on badly managed requirements delivers the wrong features however well it is coded and tested.

Change management and quality control matter, but they act on a system whose requirements are already set; the question bank this comes from treats requirements as the root cause, not all three equally.

Answer: B

The metric that does not depend on the programming language is the function count.

Function points are worked out from the functions the system delivers to the user, such as inputs, outputs, inquiries and files, so the result is the same whatever language the program is written in. Lines of code (A) and the number of tokens (C) depend directly on the language, because one statement in one language can take many lines in another. Hence (D) is wrong.

Answer: A

Function points are calculated as the unadjusted function point count multiplied by the complexity adjustment factor, \(FP = UFP \times CAF\). The UFP is the weighted count of inputs, outputs, inquiries, files and interfaces, and the CAF adjusts it for fourteen general system characteristics.

The other products are not part of the method: "FAC" is not the adjustment factor, and multiplying UFP by cost or by productivity would give a monetary or effort-related figure rather than a function point count.

Answer: B

COCOMO stands for COnstructive COst MOdel, the software cost estimation model proposed by Barry Boehm in 1981. It predicts effort in person-months, and from it development time and staffing, from the estimated size of the product in thousands of delivered source instructions, with the multipliers depending on whether the project is organic, semi-detached or embedded. The other three expansions, Composite Cost Model, Constructive Composite Model and Comprehensive Construction Model, are invented names that do not belong to any estimation technique, so only (B) is right.

Answer: B

Lines of code and function points are both measures of the size of a software product, so the answer is (B). LOC counts the physical or logical source lines, and FP counts the functionality delivered to the user in terms of inputs, outputs, inquiries, files and interfaces, adjusted for complexity. Both feed estimation models such as COCOMO or FP-based productivity figures to give effort and cost. Length of code is only what LOC itself counts, and functionality is what FP counts, so each of those names just one of the two measures. Since neither alone covers both, (A) and (C) fall short, and (D) is wrong because size is a correct answer.

Answer: D

The correct matching is the Watson-Felix (Walston-Felix) model with cost estimation (ii), the Quick-Fix model with maintenance (iv), the Putnam resource allocation model with project planning (iii), and the Logarithmic-Poisson model with failure intensity (i). This is code ii iv iii i, option (D).

Options (B) and (C) disagree on the first pair or the second pair: the Quick-Fix model is a maintenance model and not a cost estimation or a failure intensity model, and the Watson-Felix model estimates cost, not failure intensity. Option (A) agrees on the first pair but gives the Quick-Fix model the failure intensity.