hi,
from reading all the replies, i think the discussion is genrally about the measurement of performance in EPC Contracts and to see if all tasks are being completed on schedule. the perimeter for the measurement of performance is percentage complete.
i am working on the claim involing extension of time. as i have to mimic the actual schedules made by the subcontractor, the calendar variations and use of different softwares really creates a porblem e.g. CS Project schedules the activities for the same day as its predecessor finishes (I havent cleared out the concept Why it happens). so the durations of these tasks vary as calendars and start, finish dates vary.
for the calculation of performance, i have two options:
1-look for a time slice when a large number of activities was complete. or
2-evaluate the percentage complete of a particular activity and use it.
amazingly, the resluts appear quiet close to what the actual performance was. SO i would say that percentage complete does help.
Safdar H. Hashmi

Your reflection is absolutely valid. That is why my percent complete figures regress from week to week, when not only activities planned have not been completed, but remanent durations extended time over time.

Unless you finish what you planned, the measurement will make you look bad. That’s what is is for. isn’t it?

It depends on what you are looking to calculate. The usual percent complete measure addresses how much effort has been expended vs. how much effort was planned. One of my friends puts it this way, "THat’s why when you are 90% coplete, you still have another 90% to go!" She means that expending effort is not the same as producing results.

The Microsoft MS Project development team noticed this a few years back when they were updating the s/w. The team leader decided to have the project team use ’Project 2000’ as the tool for developing ’Project 2002’. After only a few short weeks team members wanted off the project. They learned a number of lessons which you can read about at http://halmacomber.com/jammin/2002_08_25_archive.html#80714453. One of the more important lessons is monitoring completion. Tracking expended effort vs. planned effort they concluded was useless. They track percent of tasks completed as planned to be completed.

In the construction world the approach is know as plan percent complete (PPC). The team tracks on a daily basis did they finish the work they planned to be finished on a daily basis. 99.5% complete is considered incomplete. You either got it done or not. You got no credit for completing work that was not planned for that time.

This method brings attention to two key aspects of projects that Eli Goldratt speaks about in the Theory of Constraints: dependence and variability. The start of my work depends on the completion of your work. If I am to start without delay, then you must complete when you promise. The PPC measure places an emphasis on completing as promised rather than the effort to perform.

Hal
http://weblog.halmacomber.com

Hi Guest
I accepted your experiance and I think you shall not insist to be Weight factor based on cost,
what is your idea sombody else ?

Dear morteza,
You are right. but for coordination & normalization between items ( such as EPC items) cost factor ( with all of them, such as material, manpower , ... ) is better.

M.Rashidi

dear, I advise to useing manhours base if you have it give the best weight factor in project, cost base may give the wrong progress dring exceution project, because you are seeing everything from view of cost not physical.

[dulplicate post]

Tony,

I work with a number of EPC contractors. There is a shift underway away from using these weighted performance measures. Even Microsoft has noted the limited value of percent complete data. See: Microsoft Eats the Dog Food. The conclusion contactors are reaching is we must measure the completion of work according to plan rather than the progress of work usually in hours applied versus planned. The first is a measure of reliability. The second is often construed as a predictor of performance.

The measure for plan reliability percent of plan complete (PPC) is easy to use. Calculate the percent of tasks completed as planned in a time period. For instance, 2 tasks planned for completion each day during the week. Lets say the two tasks were missed on Monday, but all other tasks were completed on the days planned. The performance would be 80%. Performance can be calculated by major performer (specialty contractor) and aggregated without dealing with averages of averages.

The principal reason firms are moving to this measure is task starts (and finshes) on projects usually depend on the completion of other tasks. The PPC measure addresses the fundamental variable in improving the flow of work on the project.

Hope this helps.