(1) Explain your problem, don't simply post "This isn't working". What were you doing when you faced the problem? What have you tried to resolve - did you look for a solution using "Search" ? Has it happened just once or several times?
(2) It's also good to get feedback when a solution is found, return to the original post to explain how it was resolved so that more people can also use the results.
Printer-friendly version
Send to friend
A good model shall be able to consider predictable situations that might impact activity duration.
The following are extracts of the Enhanced Scheduling presentation in my prior posting that showcase how better models can be attained.
To pretend activity production rates are always linear with resource quantity is wrong. Frequently when crew size is increased other non-labor resources change with a significant impact on the productio rate. It does not means non-linear crew production rates cannot be determined and modeled, you should if your software can handle the challenge.
More interesting is when the production rates are seasonal. This is not difficult to model if using teams and shifts in CPM model.Not all scenarios can be predicted, not all software can create such models, in such cases manual adjustments might be required. When manual adjustments are executed they shall be submitted to all stakeholders to keep the schedule transparent.
I mostly used physical complete or units complete type in my project updates since the remaining duration of the works does not only related with the quantity of the work remained. By using those types, you set your remaining duration in accordance with the current conditions of the site like weather, work force, material supply etc. and you are able to see the completion % of an activity. If you have resource loaded schedule, both of them can be used by updating the actual of the resources and remaining duration.
a
Resources - Resource Constrained Scheduling Capabilities of PM Software
At home when project schedules are subject to Owner scrutiny any changes in schedule logic or date constraints must be resubmitted for approval. On the other hand the temporal constraints added by software resource leveling are accepted as is. When Monte Carlo is required the changes in activity durations in many instances of the run create changes in the temporal constraints, this can only be done with automatic leveling as you do not manually make the changes for every instance. Because of this many of our scheduling specifications mandate automatic resource leveling.
Because of the limited resource modeling capabilities that go far beyond the capabilities of CPM of the 60’s I will take it as a fact that "most of you" do not use automatic leveling. In many cases if using P6 "best result out of 3 rules“ falls short of the modeling needs. The following presentation illustrates some examples of real life scenarios most CPM software cannot dream to model, therefore "most of you" must use manual methods to make up for the lack of functionality.
As shown in the presentation the capability for modeling Volume of Work is lacking in most software. Except in pure duration activities, Volume of Work represents true progress measure, when used with resource/crew production rates it will correctly calculate remaining duration, no need for manual updating methods that are prone to error.
When software lacks functionality even periodic updating will require much manual calculations.
I do not agree with specifications as strict as ours that go as far as rejecting payment applications if submitted update is not based on previously approved model. Even when my preference is to use automatic resource leveling I believe the contractor shall be allowed to make the necessary changes by manual methods if this is his choice and not forced to use automatic leveling. The same for when the software of his choice lacks some of the advanced functionalities mentioned in the presentation and manual adjustments are required to make up for the lack of functionality.
With regard to statistical methods I do not agree with mandating Monte Carlo as it is not the only valid method to analyze non deterministic schedules. The 3 schedules approach accepts the necessary manual adjustments and do not require special software, a methodology that can be performed with any scheduling software.
Rafael,
Thanks for the pointers.
Resource loading is one aspect that helpe but automatic leveling (as most of us would agree) is something that I am not a big fan of.
Second example masks vulnerabilities of 1 day activities for sure.
Thanks for your opinions and comments beyond the scope of original discussion.
Another example of undesired artificial split happens when you need the activity to be scheduled to start early enough to be finished the same day, if not possible then it shall be scheduled to start and finish on the following day. Concrete pouring is an everyday example, if the predecessors are finished you do not start a 6 hour concrete pour to happen few hours one day and few hours next day.
A poor model will allow Pour SOG to happen on separate days as shown for SOG 3B.
Sam,
I believe your approach is the correct one, activities duration shall be whatever it is necessary in order to create a good model.
Artificial split of activities can yield undesired results under resource leveling as it might break desired contiguous execution into non-contiguous segments.
Say you artificially split Activity 1 into Activity 1A and Activity 1B, all activities resource loaded with 1 ea Resource A with availability of 1 ea.
Before leveling.
After leveling, Activity 1A and Activity 1B are one week apart instead of contiguous.
I have tried to provide granularity by utilizing steps in certain cases. I may be moving away from my original quest here but In general, I am okay with greater timelines equal to 2 to 3 reporting period. What really gets me is one day duration when the hand off must happen from one department to another or from one GC to another
no duratrion should be more than 20 working days if it is break it down to smaller peices stages or phases
Unfortuantely the duration is spread is quite wide from 1d to 50d excluding long lead items which are longer than that.
Hi Sam
If your tasks are no longer than 10 days then you need only 3 stages:
0% Not Started - 50% Started - 100% Finished.
Little errors will balance out and are rectified next week.
Best regards
Mike Testro
It sounds like Physical % Complete will make more sense then instead of remaining duration although bit tedious.
Would % complete change if Remaining duration is reduced or increased based on the status input?
Rafael has it correct if using P6 use remaining duration. If you use duration % complete it can cuase your remaining durations to be in fractions of a day. i.e 5.5 days
Also you always want an accurate forecaste of the number of days to complete the remaining work NOT the total number of days to complete but the remaining work days not calendar days but estimated work effort in terms of days.
P6 does not speaks the language of Construction Contractors as we use Volume of Work to define activity progress.
The following reference might be of help with regard to P6:
A) DURATION % COMPLETE
Enter a percent complete value in the Activity % Complete column or directly into the Duration % Complete column. Primavera P6 will automatically calculate the Remaining Duration for the activity. Or enter a Remaining Duration value and let P6 calculate the percentage.
This percent complete type is best used for more level of effort type activities where measurable deliverables are not expected and the percent complete represents elapsed time more than deliverables completed.
When Physical % Complete is used, you must adjust the Remaining Duration as this is not calculated by Primavera P6. You must enter a Remaining Duration manually, or set an Expected Finish date on the activity.
If you don’t adjust the Remaining Duration, P6 will add the Actual Duration to the Remaining Duration. This will cause unnecessary variance in the project and will likely alter your critical path.
The Physical % Complete progress type should be used when dealing with discrete work that equates to one or more measurable delivery items.
The Units % Complete type is used when resources are assigned to an activity and actual units worked will be tracked.
As with Physical % Complete, Units % Complete doesn’t calculate the remaining duration for activities, so you will need to use either the Remaining Duration field or an Expected Finish date to manually control the calculated end date for your activity.
I find it superior approach to use volume of work and software to calculate remaining duration considering the relationship between volume of work, resource production rates and availability, but if not an option I find it easier and less error prone to ask for remaining duration and let the software calculate the %. I believe using % duration can be misleading in case activity duration increases/decreases as if an activity with Original duration of 32 days after 20 days still has a remaining duration of 18 days, figuring out the correct % can be tricky, on the other hand the software will get it right if you enter remaining duration.
If I were a P6 user I would favor using Remaining Duration
I worked for many decades in the construction industry as Project Manager; usually we reported progress based on Volume of Work irrespective of the software used. Progress based on elapsed duration was limited to activities where no matter how many resources or what production rates their duration was fixed, for example concrete curing and delivery activities.
Email: PPadmin@planningplanet.com
Replies