If an activity has +10 days of total float, which means if we delay this activity by 10 days it becomes critical path. Just to simulate, if we make this activity's duration increased by 11 days. Then this activity would drive the project end date and would be on longest path. This scenario is understood. Based on this, the situation should also true with negative float. That is, if an activity has -30 days of negative float, should this does mean that this activity is pushing out the end date by 30 days ?
In my schedule, I have like -120 days negative float but still the activity is not on critical path. I am struggling to interpret this situation. Please share your thoughts.
P.S : In the schedule, we have imposed an constraint on the finish milestone to map the contractual position. Since the project is delayed, there is a negative float.
I wont say that it is not possible I have no reason to believe that Mr Harris is incorrect but that was back in 2011 after 6 years I think that if it was truly a problem I believe that it would be fixed by now.
NO difference from any other software tools still performs the forward and backward pass and then calculates the float.
Zoltan, I think we agree.
Tom you are wrong yes if you improve the activity's early finish by 25 days it will bring back the activity on track. It seems P6 DOES supportive this perspective.
NO difference from any other software tools still performs the forward and backward pass and then calcualtes the float.
Sharief,
“… if an activity is showing -25 days of negative float then improving the activity's early finish by 25 days should bring back the activity on track. It seems P6 is not supportive from this perspective.”
As Robert highlighted, "On track" is a vague phrase implying conformance to baseline dates. Yes, accelerating the TF=-25 activity by 25 days will make EF=LF and TF=0. This will lead to recovery of the completion milestone to its constraint date if and only if 1) the activity was in fact on the driving path to the constrained completion milestone; 2) the activity's total float is not controlled by some other late constraint; and 3) there were no parallel paths within 25 days of driving. If these conditions don’t hold true, then all the acceleration does is remove the activity from the list of concurrent delays. P6 easily determines the first two conditions (with LP or MFP analysis), and (in the absence of a Drag metric in P6) you can answer the third using trial and error or perhaps the 3rd party tool I mentioned earlier.
With multiple calendars, you can’t assure any of those conditions using total float alone.
Mike,
“As far as I am aware P6 is the only software that generates negative float as a routine.”
Perhaps you are not familiar with a little bit of software called Microsoft Project. MSP routinely generates negative float whenever a deadline or other late constraint is violated.
p6 generates float be it positive or negative by comparing the differences between the early dates and the late dates after the forward and backward passes. NO different form any other scheduling tool.
As far as I am aware P6 is the only software that generates negative float as a routine.
If it happens in Asta it flags up a default until you remove the constraint.
Best regards
Mike Testro
The problem with this kind of constraint is that it makes it difficult to highlight the true critical path and even if you define this correctly, the amount of negative float in the schedule will likely confuse many of your stakeholders.
I understand why you've set things up this way but bringing back an activity by it's negative float does not bring it back on track. The only way to do this would be to bring it back to it's baseline dates. For example an activity might have started with 60d float and is now showing as having -25d. To achieve your original dates you'd need to recover 85 days on this activity, not 25d. And while it is helpful to be able to see how much time you need to recover in order to achieve your baseline dates, how useful is this now? You are best focusing on your critical path until this regresses behind the next most critical path (by monitoring the critical drag), instead of being overwhelmed by a schedule with 30/50/70% of activities showing with negative float. You can always use "Variance - BL Project Finish Date" in this way instead to see the slippage, whilst keeping your float values sensible.
Personally at this point I would keep your planned and contract dates disconnected.
Thank you very much everyone for the very interesting responses.
Basically, my idea of constraining finish milestone to reflect the contractual completion is to know the amount of time a delayed activity to be improved/recovered in order to meet the original contractual completion date. For instance, if an activity is showing -25 days of negative float then improving the activity's early finish by 25 days should bring back the activity on track. Unfortunately, it seems P6 is not supportive from this prespective.
Mike,
As advised removing the constraint will allow the free flow of logic and will also convert the negative float to positive. But this Total float is calculated with respect to the new completion date. The then calculated total float will indicate how much an activity could be delayed without delaying the new project completion date. But not against the previous project completion date. In order to achieve this, I was trying to impose a constraint and try to get the information. Morevover I have not used mandatory constraint, the free flow of logic is still permissible that to only finish milestone of the . Based on the above discussions it seems, this scenario couldn't mapped.
It would be very helpful to me, for any adivse on the work around from CPM prespective.
This debate just demonstrates what nonsense P6 creates with its use of negative float as well as its use of longest path.
That late dates can be earlier is nonsense, a requirement to get negative float values, in any case a separate name and field shall be used for the calculation that is not required to show criticality if god math is used to calculate constrained floats.
As soon as there are constraints such as date constraints or resource constraints P6 longest path breaks apart and the results are unreliable/flawed.
The longest path is broken when activities are no longer driven by relationships; that is, when activity dates are driven by constraints or resource leveling. Longest path calculation includes interproject relationships. Therefore, activities designated as on the longest path may change depending on whether you schedule a project alone or with its related projects. If a project has interproject relationships and you schedule it alone, the interproject relationships are treated as scheduling constraints.
Sharief,
The answers to both your questions comes down to activity calendars. The Total Float of each activity (i.e. LF-EF) is computed according to its own calendar. If the driving logical path to project completion (i.e. the Longest Path) runs through activities with different calendars, then total float MAY not be the same. (Consider the simple example of a 7-day activity finishing on Friday driving a 5-day activity starting Monday. The 7-day activity sees 2 days of float even though it is on the driving path.)
If you have introduced seasonal or daily weather calendars, shift calendars, curing calendars, etc. then you may see wide variations in total float along the "Critical Path" and any other logical paths. This is why the Longest Path alogorithm exists in P6.
If an activity has negative float in P6, its Late Finish is less than its Early Finish - plain and simple. Most instances are the result of an unsatisfied late constraint on some later activity. To properly understand the meaning of "-120 days negative float," you must understand a) what is controlling the Early Finish of the activity (trivial in a well-constructed schedule); b) what is controlling the Late Finish of the activity; c) what worktime constraints (from the activity calendar) are applied in computing TF=LF-EF, in hours; and d) what factor (from the activity calendar) is used to convert the measured TF from hours to days.
While I have used negative float as a simple indicator that one or more contractual constraints may be violated, Total Float in general is not reliable for defining the Critical Path and Near-Critical Paths in complex schedules with multiple calendars and/or contractual milestones. You would be well-advised to use Longest Path, Multple Float Path (with free float option), and perhaps a 3rd-Party tool like Schedule Analyzer and its "Longest Path Value" metric, for such purposes.
This debate just demonstrates what nonsense P6 creates with its use of negative float.
The question you should be asking is "What is the completion date?"
So take off all constraints and let the logic flow to see the true critical path.
Best regards
Mike Testro
Robert/Zoltan,
I am using Longest Path as Critical Path. You are right, there are activities with greater than -120 days which are not appearing on the critical path. That looks strange to me as well. To be specific our project completion date is 31-Dec-17 and when I updated the progress the forecast completion date is 21-Mar-18. P6 shows that the overall negative float is -67. Neverthless, it's understood that P6 is showing -67 days based on working days, instead of -80 days of calender days.
I am trying to understand two things here :
1. Why activities with more negative float are not appearing on longest path
2. Even on the longest path, different activites have different total float. Few have -130 days, -79 days, -117 days etc. But as reach to last activity the total negative float appears to be uniform which is -67 days, which ties back to overall total float.
Appreciate very much for your clarification.
-120 days is a random number and does not tell the whole story. Is this calendar days or working days ?
You said that you have in imposed contractual completion date. I assume that since it is a contractual day and most contract stipulate the contractual completion date as X number of days afetr award or NTP that the imposed finish date is on a 7 day contineous calendar. So if the end date or last activity is -120 (calendar days) I would take it that all of the other activities are -86 days (working days).
This might be where your confusion comes in in understanding the negative float.
Do as Robert Bell suggested and defined your critical path as the Longest Path.
You can also filter for LONGEST Path AND Critical
Another thing you could check is the bar settings in the Gantt chart. Your critical bars' filters and formatting could have been changed.
Hi Sharief,
Do you have activities with even greater negative float than -120 days? If so, and you have defined your critical path as the Longest Path, then these activities will not show as critical.
Perhaps alternatively you have defined critical as anything with less than, say, -150 days float because of a position you were at before?