Task completion calculation

Hello, back with another question: how does GP calculate overall project completion from its tasks? Or task completion from its sub-tasks?
This figure appears in the pdf report and it would be good to know what it means. Thanks in advance and keep the excellent work you’re doing.

For each leaf task its completion is duration * percentage and this accounts for “completed days” of that task. If the duration of a leaf task is 10 days and percentage is 30% then its “completed days” virtual property is 3 days.

For each summary task the sum of all durations of child tasks is “planned days” and the sum of completed days of the child tasks is completed days of summary task. Completion percentage is completed days / planned days.

If you have a summary task with two child tasks, one with duration=10 days and completion percentage=50%, and another with duration=5 days and completion percentage=60% then planned days of the summary task is 15 and completed days is 100.5 + 50.6 = 5+3 = 8. And its completion percentage is 8/15 = 53%

The same calculations are applied recursively up to the project level.

Hi,
From the explanation, the % completion seems to be based on duration completed and planned days, rather than “work” completed.

When I enter value to % complete column, I input from the % of “work done” perspective. Is there anyway I can generate total % work completion from GanttProject? This will give a better view of the “real” work completion for the project.

TQ.

Feel free to replace “completed days” with “work done” in the text above. I used the term “completed days” just because it is graphically shown as a black progress bar which, naturally, covers a few days from the task total duration. Otherwise, completion percentage is just a percentage of some total value.

Hi,
Renaming label can’t quite help. Assuming summary tasks:
Task 1, 10 days, 50% completed
Task 2, 5 days, 60% completed

The 50% I entered is intend to reflect 50% of functions in task 1 that has been completed, which i may done that using less or more than 50% of the planned days.

So I would like to get (50+60)/200*100=55% work completed, instead of 53% based on current Gantt project calculation. Because % of work completed can reflect more correctly of work progress aspect. For example, Gantt project gives in report as total % completion as 25% but using my own work completion is 9%, so I was unable to get a correct view of work progress. The current % completion is fine except I’m looking at the other aspect of progress.

TQ.

Hi, not sure if I understand you correctly, but GP logic seems OK to me. I use the “work completed” in the progress. If I assign say 10 days for a task, and finish it in 3, then I would report it 100% after 3 days and GP takes it to the upper level as 100%. However, if you have another subtask, that has not even started yet (0%), then what is the completion of the project after these 3 days? 50 or 0? At least, GP indicates that you are doing something, getting somewhere, but it can not predict the success of the project as a whole.

Hi Sir,

Below are 2 same project files except, test1’s Project > mod1’s tasks_1 & task_2 are not linked) and test2’s are linked.

I was expecting same 37% for both files, but may I know why test1 gives 35% instead? If task_1 & task_2 are linked, only then it becomes 37% (technically is 37.5% based on your shared formula).

test1-35.gan (4.6 KB)

test2-37.gan (4.7 KB)

Appreciate your advice because some of my real project which have 100+ tasks and subtasks are linked and some are not, concern in case calculation have big differences which we not aware.

Thank you.

Hello,
thanks for your questions. Indeed, completion percentage in GP 2.8.X is not working correct in some subtle details. This is fixed in the development version, and I posted my formula by looking at the code from the development version. Sorry that I didn’t look at the differences.

So, in GP 3.0 code both these projects produce the same result (37%). Since GP 3.0 will take some time to get to production, I will consider backporting the fix to GP 2.8.X

Hi Sir,

Thanks for the clarification, would appreciate the correction be made available in 2.8.x.

Is it possible to share the roadmap on how soon will this be made available?

Thank you.

We do not have any roadmaps, sorry. In an open-source fun project all estimates and predictions are wrong.

For the record, an issue in the tracker:

1 Like

Hi Sir,

Could you help explain why the attached is giving 37% completion, yet only got 2 tasks that achieve 50% only, the rest of the 6 tasks are 0% completion. If tasks 2 is shifted to start earlier, the overall % completion will reduce, although no change to the rest of the 6 tasks.

test3-37.gan (4.8 KB)

I notice this in my real gantt chart because the % completion is higher than what I think it should be.

Thank you.

That’s because GP 2.8.X takes into account summary task lengths (which is a bug).

Hi Sir,

Meaning the cause is same as what you logged?https://github.com/bardsoftware/ganttproject/issues/1737

Any workaround for both case that we from users side can do, while waiting for new 2.8.x release? at the moment feeling doubtful if to submit what Ganttproject has calculated for project progress reporting…

Thank you.

One good option is to start using beta version of GanttProject 3.0.
It includes all bugfixes from 2.8.X, file format in GP 3.0 is compatible with GP 2.8.X (well, actually just the same), and the major changes were in GP Cloud integration, while all existing features have not been changed, so it must be pretty stable.

You can download packages for all supported platforms from this Dropbox folder: https://www.dropbox.com/sh/ijzr7s7ozxqexls/AADlRUGq667XQjEK8j4StEp2a?dl=0

Packages for Windows and macOS come with bundled Java Runtime, Linux and ZIP require Java Runtime with Java FX, e.g. Liberica (bellsoft-java11-full package) or Zulu