Page 1 of 1

Pedigree uncertainty

Posted: 30 Apr 2013 16:28
by schivlg
A few questions about the use of pedigree uncertainty:

1) If the uncertainty type is kept as "No distribution", will the pedigree uncertainty be taken into account during a monte carlo simulation?
2) If a normal or lognormal distribution is selected will the pedigree uncertainty be used?
3) Once a pedigree uncertainty has been entered for a flow is there any way to remove it entirely?

I'd like to try and make use of pedigree information to add uncertainty but am not sure if it's actually making a difference in the results right now.

Greg

Re: Pedigree uncertainty

Posted: 25 May 2014 22:03
by mptsang
Hi,

I was curious about similar thoughts and wondering if there was a clear answer to any of the above questions?

Best,

Michael

Re: Pedigree uncertainty

Posted: 26 May 2014 14:08
by aciroth
Hi Michael and all,
the pedigree uncertainty will not be automatically taken into account in the Monte Carlo simulation, only the uncertainty that is specified. And indeed, it is not possible so far to completely remove the pedigree information if it has been specified once (maybe not optimal but since it is not directly connected to the uncertainty that is used, this is maybe tolerable).
Hth,
Andreas

Re: Pedigree uncertainty

Posted: 09 Sep 2014 11:51
by mudeser
1) If the uncertainty type is kept as "No distribution", will the pedigree uncertainty be taken into account during a monte carlo simulation?
2) If a normal or lognormal distribution is selected will the pedigree uncertainty be used?

Re: Pedigree uncertainty

Posted: 10 Sep 2014 10:48
by aciroth
Hi, in both cases: no; the pedigree uncertainty at present is purely informative and not taken up in the Monte Carlo simulation. But I see that it could make sense to include it.

Re: Pedigree uncertainty

Posted: 13 Apr 2017 15:48
by ChrissyBrand
Hi,
is there a written source where I can find the explaination that the pedigree uncertainty is not taken into account for the monte carlo simulation? I need a source for this information for my bachelor thesis...
Greetings, Christina

Re: Pedigree uncertainty

Posted: 13 Apr 2017 15:59
by aciroth
Hi Christina,
actually that is my own personal judgement that the original ecoinvent pedigree uncertainty is based on expert judgement without further specification of background sources and thus not too reliable. But as you may have seen, in recent openLCA 1.6.2 we have added the possibility to include the uncertainty from a pedigree-like data quality system in the process uncertainty or not, see screenhot below: If you check "system defines uncertainty", then the pedigree uncertainty is considered in the uncertainty.
Best wishes,
Andreas

- and now you even have a "written source" ;)

Re: Pedigree uncertainty

Posted: 08 Jul 2017 04:34
by smnipa
Hi Andrew,

I am using open LCA 1.4 version. To do Monte Carlo simulation I included all uncertainty information in pedigree uncertainty (pedigree matrix) in all of the process card and I used 1000 iterations. My process chain is very big, its life cycle assessment of cotton apparel from cradle to gate. Problem is I am not getting simulation result. Please kindly advise me what wrong in my Monte Carlo simulation and what I did wrong? I included picture here. Please advise me the about the solution.