(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
Mike,
I never said Cost is the best basis for assessing progress. You might have a job progressing on time but not within budget (money budget, man-hours budget …), or progressing not on time but within budget (the most common). You have to consider both in your assessment of progress.
US Government came out with the idea of Earned Value just to find out it is flawed, and because they respond very slow took them decades to realize it. Their need is different from that of the Contractor because of budgeting reasons, for many government Contractors, especially defense contractors that a project or a program for developing new weapons needs more funds it means more business, it is good to them, bad for the Government and the tax payer.
When you are assessing progress on a given job you can progress through critical activities with a combination of non-critical activities that might or might not increase your probabilities of meeting a target date. Neither Earned Value not the new boy in town Earned Schedule tells you about how your potential capacity of success changed, probability theory can give you a better insight. The same applies for the meeting of budget which depends on a combination of time factors and other non-time related cost risks.
Traditional Job Costing preferably using intuitive metrics such as unit cost based on physical units along with probability theory is a better deal.
But don’t worry; most probably US Government will create 49 new Earned Schedule metrics alone with new acronyms only Tutankhamen will remember. Not only will add confusion to the confusion but will solve nothing relevant.
EV is kind of confusing that some users even call BCWP to earned man-hours as if money, morons. "Bah! Humbug!".
Might BCWP bless you.
Best regards,
Rafael
Hi Rafael
"All CPM schedules have a Critical Path that at any given time determines the project completion and activities outside the critical path do not."
Really?
I would put in a caveat "All properly prepared CPM schedules ........."
I don't agree that cost is the best basis for assessing progress unless you have teased out only the time related costs.
Then you might as well do it properly and use resource hours.
Best regards
Mike Testro
All CPM schedules have a Critical Path that at any given time determines the project completion and activities outside the critical path do not. But as a project is progressed activities durations vary and this might shift critical path activities. The deterministic approach to evaluate schedule performance fails to forecast this variability and in most cases, if not always, will be optimistic and we will know at the end, when it is already too late.
From my readings about EVM “Earned Value Method” and ES “Earned Schedule” I could not even see their recognition of Early and Late cumulative S Curves, as if there is no such thing as Early and Late S Curves. Although some claim ES is the answer, it is not; even if eventually Early and Late dates are included in the procedure.
One of the basic problems is to determine how to weight the relative values of critical activities versus non critical activities as well as for cost values as pointed out before by Deveaux. I believe there is no such thing as an exact answer but that the probabilistic method is the scientific way to deal with the issue.
There is a tendency to avoid probabilistic methods because they are based on assumptions about the probabilistic distributions for cost and performance (activity durations) as if the deterministic assumptions are flawless. A well thought probability distribution even when subjective is better than a well thought fixed deterministic assumption.
By comparing the cumulative distribution of success for cost (budget versus actual), as well as comparing the cumulative distribution of success for project end (budget versus actual) a better evaluation can be made.
Applying statistical values to a CPM schedule can be done easily with the help of the computer, the output is performed at a click of the mouse and for the surprise of many I believe easier to understand than the confusing terms of EVM. Anyone can understand what a 10% probability of finishing a job on time means.
Perhaps something good comes out of it. Maybe Contractors will be allowed to schedule for early completion in the hope to finish on time instead of being required to submit a baseline schedule that uses all contract time.
Anyone with a basic knowledge on statistical methods for scheduling knows that if a job scheduled for 100 days have a success probability of 50% to finish as scheduled, the same job scheduled for 80 days might have a success probability of 40% to finish on day 80 as scheduled but perhaps an 80% success probability of success to finish by day 100. This because of how activities in the deterministic schedule not in the critical path can become critical and drive further the job duration while activities with lower duration not in the critical path will have no impact.
Mike, quite true.
The trouble with all methods of tracking project progress is that, almost invariably, we're not just dealing with making widgets. In some activities, we're designing fidgets. In others, we're hiring midgets, pointing digits, dating Bridgets, and testing gobbledegooks. Even if each activity sometimes appears "cookie-cutter", they almost never are: if you're building a highway 100 miles long and you've completed 90 miles, what percent complete are you? What if the last ten miles are through the Florida Everglades?
Earned value is a VERY blunt-instrument approach to weighting activity-driven milestones by some aspect of the activity. The most common is total costs, but labour costs and labour hours are also commonly used. (Even if using total costs, there is very good reason for also isolating just the labour hour component and using EV tracking on that metric, too.)
Of course, since you mention SPI, I will also point out that, while CPI tracking has value, SPI is usually not only distorted, but can actually, in the short term, reward the contractor for ignoring critical path activities and doing work out of sequence. For schedule tracking purposes, EV metrics:
The distortion can be lessened by:
The flaws in EVMS are sufficiently serious that the US DoD (which created and mandated the technique in the 1960s) has recognized the inadequacy of earned value tracking and is currently looking into completely overhauling the methodology for US defense contractors.
Under any circumstances, every worthwhile text on EV will emphasize that SPI tracking should always be complemented by progressing of the critical path.
Fraternally in project management,
Steve the Bajan
Hi Fellow Planners
To misquote one of your contributors " I thought I knew everything about the subject (EV) but now I've forgotten"
What if I don't have costs associated with the activities, I'm just a simple planner, not a cost engineer so how would I know what the costs are
If I want to know the SPI, surely this is nothing more than Earned(Achieved) vs Planned (900widgets/1000widgets = 0.9)
Thanks
MikeW
Email: PPadmin@planningplanet.com
Replies