Say I have a process taking 19 h. Some improvement measures allow to reduce the required time to 17 h (to reach the same level of output).
Does it make sense to say that the process is now (1-17/19)*100 ~= 10% more efficient?
If not, what kind of quantification of the process improvement would make sense?
Processes can be improved in a variety of ways:
and probably tons of other dimensions, too. If your process is repeatably two hours faster, rather than a statistical outlier, then you could certainly say your new-and-improved process is 10.5% faster than before, but it may or may not be more 10.5% more efficient.
Whether or not your process is actually more efficient depends on how you choose to interpret efficiency and what you want to get out of an optimized process. Consider these examples.
These aren't trick questions. The answer is just as likely to be yes as it is no, because for most organizations "efficiency" is a buzzword that's really a proxy for something else. In my own experience, it's most often a proxy for value/time or value/cost, but an engine manufacturer might think an improvement in the rate of energy capture is what the process ought to be optimized for.
Unlike some of the hard sciences, efficiency is a less objective term in project management. Therefore, your mileage will vary.
I would call that a 10% savings of time. But time is only one factor of efficiency. To look at the efficiency savings you need to capture the ratio of output/input of the before and output/input of the after and compare the delta.
In Lean, there is a Process Cycle Efficiency metric calculated as a ratio between cycle time and value-added time. From the 17h how much of it was valuable effort? For example: 5h (value-added time) / 17h (cycle time) = 30% (process efficiency).
