Function Point Analysis FPA: A Guide to Software Estimation

Examine the problems with lines of code as a widely used software metric, including undefined definitions, language dependence, and the difficulty of estimating size in analysis and design.

IFPUG argues against using lines of code for sizing, citing platform, language, and organization dependence, and noting some fourth generation languages do not rely on LOC.

Function points quantify software size and complexity by the functions delivered to the user based on logical design and functional specifications, independent of language, enabling early estimation for business applications.

Function points overcome lines of code problems by remaining language and platform independent, estimable early in analysis and design, and reflecting the system's external view for non-technical users.

Function points measure productivity and project size, and estimate staff, effort, schedule, and cost for outsourcing. As an industry standard, language- and platform-independent, they guide information systems decisions and baselines.

Count function points early and often to measure user-requested functionality independent of technology, ensuring a universal, consistent measure among projects and organizations, establishing baselines and enabling monitoring and control.

Identify who should count function points, establish independent, trained counters, and follow the steps to compute unadjusted points, apply 14 complexity factors, and derive adjusted points, man-days, and cost.

Illustrates function point counting process, guiding you to identify count type and boundary, count data and transaction functions, and compute unadjusted and adjusted function point counts using 14 complexity factors.

Identify the function point count type for a project, define counting scope and the application boundary, and map inputs, outputs, inquiries, and logical files to external systems.

Identify the unadjusted function point count by analyzing data functions (internal logical files and external interface files) and transactional functions (external inputs, external outputs, external inquiries) within the application boundary.

Explore the five function point components, external input, external output, external inquiry, internal logical file, and external interface file, and their roles in updating and retrieving data.

Explore how unadjusted function point count uses ret, fdr, and det to value data and transactional functions, considering external input, external output, external inquiries, and ilf/aif structures.

Identify data functions as the internal logical files and external interface files, including AIFs, and learn to assign functional complexity and unadjusted function points.

Count transactional functions in function point analysis by classifying external inputs, outputs, and inquiries; use FDR and dates to determine AI, EU, and EC complexity.

Compute the unadjusted function point count by filling the five components with low, average, or high complexity, multiplying by their weights, and summing to the total.

Compute the value adjustment factor (vaf) in function point analysis by assessing 14 general system characteristics on a 0–5 scale to adjust the unadjusted function point count.

Learn how to compute adjusted function point counts by applying the value adjustment factor to unadjusted FP for development, enhancement, and application projects, including baselines and conversion functionality.

This lecture presents a development project function point analysis, classifying data functions and transactional functions to derive an unadjusted function point count of 115.

Compute unadjusted function point counts for the development project by combining application contribution and conversion functionality, then apply a value adjustment factor to obtain the adjusted total.

Learn to compute enhancement project function points using the formula (ADT + CAGR + DL) times VF, accounting for added, modified, and deleted functions and the value adjustment factor.

Explains the computation of the function point count for application after enhancement projects, including unadjusted points before enhancement, added, deleted, or changed functionality, and the value adjustment factor.

Convert an adjusted function point count of 124 FP into seven man months using a productive factor of 18 FP per month, and estimate a labor cost of dollars 36,400.

Walk through a function point analysis example using the FPM method, calculating inputs, outputs, files, inquiries, and interfaces to derive unadjusted and adjusted function points and estimate effort and cost.

Size software projects accurately with function point analysis, enabling estimating models, productivity-based cost estimates, scope creep management, universal technology-independent sizing, and clear stakeholder communication.

Harness function point analysis as a uniform, technology-independent sizing metric. Measure productivity as function points per work month and quality as defects per function point to guide improvement.