Function Point Measurement: When Should a Data Group Be Taken Into Account?
Function points are generally used at certain key stages of a project: during the scoping or end of design phase to estimate workloads, and at the end of the project to review the project and produce KPIs.
In both cases, the entire measurement is taken into account, and the question of whether an elementary process (EP) or a Data Group (DG) should be taken into account is governed solely by IFPUG rules.
However, it may be useful to evaluate subsets of the project (stage, batch, release, sprint in agile methodology, or during a V-model cycle, for example, to report progress in terms of % of PF completed vs. expected). In this case, only the basic processes (BP) and Data Groups (DG) completed during the period in question should be taken into account.
While it is generally easy to identify the basic processes carried out during the period in question, it is more difficult to identify the data groups concerned.
When should a data group be considered for inclusion in the measurement: when creating tables? When performing read/write functions, or for all queries using the GD?
We propose counting the data group with the first elementary process referencing it and thus counting the Function Points of the DG in the subset considered (batch, sprint, etc.).
