It is a very important topic that you touch upon here, and one which does not necessarily require in depth statistical analyses of Life Cycle Inventories. It is as much a question for philosophy of science. Should we make the perfect the enemy of the good?
We will never reach a situation with no uncertainty. All models are representations, valid for a certain sample - with known and unknown uncertainties. By assuming that there at some point will be no uncertainty, means that one has not understood the concept of modelling. Waiting for this to happen is like waiting for Godot.
On case you want a scientific paper addressing this point, I would recommend a paper I wrote some years back on this topic, titled:
"Obligatory inclusion of uncertainty avoids systematic underestimation
of Danish pork water use and incentivizes provision of specific
inventory data."
Hi Eric
It is a very important topic that you touch upon here, and one which does not necessarily require in depth statistical analyses of Life Cycle Inventories. It is as much a question for philosophy of science. Should we make the perfect the enemy of the good?
We will never reach a situation with no uncertainty. All models are representations, valid for a certain sample - with known and unknown uncertainties. By assuming that there at some point will be no uncertainty, means that one has not understood the concept of modelling. Waiting for this to happen is like waiting for Godot.
On case you want a scientific paper addressing this point, I would recommend a paper I wrote some years back on this topic, titled:
"Obligatory inclusion of uncertainty avoids systematic underestimation of Danish pork water use and incentivizes provision of specific inventory data."