I agree that considering attributes of float is an interesting idea for new metrics to measure project status, but wonder if "degradation" is the best way to look at it. The word conjures up only a unidirectional image for me, I'm afraid.
Over the course of the project, the float on any path can go up or down from that originally projected when the baseline schedule was established, as a function of whether the tasks take more/less time than originally planned.
Thus, describing a situation where float increases would require use of a negative "degradation" statistic, something that feels a little counterintuitive. Two sides to every story, though--leave us not neglect to pay due homage to "negative float"!
Conventionally, positive quantities or indices > 1 are used to indicate desirable situations. (Although, depending on how much your float increased, I guess I'd have to wonder if that was actually a "good" thing, and start looking upon your duration estimation/resource planning skills with a jaundiced eye if your float kept increasing.)
That said, what about a float metric that either indexes the current float to the float in the baseline plan (<1 is bad, indicating task slippage or emergence of resource constraints), or one which measures the rate of change in the float [the change in float from time(1) to time(2), divided by the time difference between time(1) and time(2)]?
In the case of the latter situation from the paragraph immediately above, a positive rate could be taken as a "good" indication, as long as things didn't get TOO good, as noted previously!
Taking it way out in the ether, I guess one could also then regress these empirics across a variety of like projects that went well/according to plan, define the functional relationships in two or three dimensions, calculate their derivatives, and use them to project float performance on future projects.
I don't surf a lot of PM trade rags or professional journals, so I don't know if these ideas have yet been surfaced in refereed print before. If they strike someone's fancy as being novel, and spark an interest in publishing a paper, please leave a note on this thread for contact "off-thread" (so to speak) to explore mutual interest. Alternately, probably now would be a good ex-post-facto time to reconsider eng-tips' intellectual property rights provisions!
