negative results

Post Reply
arunsriks
Posts: 2
Joined: 18 Sep 2017 21:21

negative results

Post by arunsriks » 24 Sep 2017 21:34

Hi,

I am getting negative values as results in all of the impact categories when I carry out the impact assessment for my system and it happens for all of the impact assessment methods. Can negative values be obtained as a final result? If no, can anyone suggest me why this is happening?

Regards,
Arun

aciroth
Posts: 750
Joined: 09 May 2010 23:28

Re: negative results

Post by aciroth » 25 Sep 2017 10:00

Hi Arun,
welcome - this can happen but it is unusual, it means that the product you are assessing is overall good for the environment (reduces emissions, produces resources). You will have negative results from
- system expansion modeling
- recycling and waste treatment modeling possibly
- omission / no consideration of flows that have impact in the impact assessment (so, the amount of your impacts is low and negative because the typically much higher "normal" impacts are not considered)
- incorrect specifcation of impact assessment methods
and maybe more.
Maybe you can post an example here?
Thank you,
Andreas

Sandra
Posts: 3
Joined: 13 Dec 2017 09:59

Re: negative results

Post by Sandra » 13 Dec 2017 11:28

Hi,
I am facing a similar problem. Most of my results are negative, although this can't be possible. I already checked the points you mentioned, Andreas: I am not using system expansion (but economic allocation with ecoinvent apos), I set waste treatment to 0, omission is possible but not at that extend, I tried different impact assessment methods ... and results still stay negative.
In my process "sugar production" (FU = 1t) the impact of "sugarbeet production" as an input has opposing positive/negative values than the process sugarbeet production itself (FU 1t), this does not make sense to me at all.
I made a screenshot from the result screen.

I would highly appreciate any help and ideas.
Regards, Sandra
Sugarproduction.jpg
Sugarproduction.jpg (130.61 KiB) Viewed 15359 times
sugarbeet.jpg
sugarbeet.jpg (114.2 KiB) Viewed 15359 times
InputsSugarProduction.jpg
InputsSugarProduction.jpg (274.18 KiB) Viewed 15359 times

jrsfinn
Posts: 8
Joined: 04 Dec 2017 01:15

Re: negative results

Post by jrsfinn » 15 Dec 2017 05:51

I was also curious about negative impact contributions.

I have attached a screenshot of 'Land Use' Top 5 Contributions in hopes to clarify my thinking:
Both the market for waste paper and the treatment of waste PET are negative contributors in this case. Does this mean that waste paper market (as some of that market could be recycling) positively contributes to land use as it can be re-used and decrease the land use for paper raw materials (ie trees)? And perhaps the treatment of waste PET positively impacts the land use as it will not be piled into a landfill

Please advise if I'm along the right track or not - would appreciate any feedback. Thank you!
Attachments
Neg.PNG
Neg.PNG (47.2 KiB) Viewed 15337 times

Sandra
Posts: 3
Joined: 13 Dec 2017 09:59

Re: negative results

Post by Sandra » 15 Dec 2017 16:14

Hi there,
regarding the negative land use values: I would interprete negative land use values as changing land that is used for industry/agriculture to natural state. I don't see that this is given in the waste treatment or market process but it could be an issue of interpretation.

Regarding the problem I posted a few days ago, I found out that I could fix the problem by setting a cut-off at the creation of the product system. Therefore I assume, the negative values origin from loops in the pre-chain where outputs (also by-products!) of some processes in the system are used as inputs for other processes and the size of the amounts needed excel the amounts produced. I did not expect this as I thought openlca calculates with a matrix inversion similar to the Leontief Inverse. I would be happy to get any further explanation, advice or help as the cut-off i had to set was rather large (1%).

Thanks, Sandra

aciroth
Posts: 750
Joined: 09 May 2010 23:28

Re: negative results

Post by aciroth » 16 Dec 2017 12:48

Hi Sandra,
yes indeed openLCA uses a matrix-calculation approach which is able to handle loops, so this should not be the reason.
Negative impacts are caused by avoided emission modeling (in system expansion), which can also implicitly contained in aggregated processes. If you have "messy" system processes in your product system (meant to be fully aggregated, with negative flows, but also with several products on input and output side; self-created, or also from databases like ProBas or also GaBi), you need to carefully check the connections of these processes.
Best wishes,
Andreas

Sandra
Posts: 3
Joined: 13 Dec 2017 09:59

Re: negative results

Post by Sandra » 19 Dec 2017 19:32

Hi Andreas,

unfortunately I can't see how to fix the problem with this approach. I am using ecoinvent (unit-process version) and my self created processes are all unit processes, too. However, there are direct emissions in some processes that are negative (e.g. to respect the soil as methane sink in agricultural processes).

When screening the analysis results, the large negative impacts are obviously rooted in the use of by-products in some processes e.g. sugar beet pulp used as feed. In the inventory results, the amount of sugar beets needed is negative although in the process itself the amount of sugar beet pulp, that was set as an input is positive.

It seems to me, the software changes the positive amount needed into a negative demand of the main product. However, I did not check any "avoided product" box or likewise (economic allocation is saved in all processes as well as for calculation).

I'd be very happy of any further idea or hint that could solve the issue.
Thanks a lot
Sandra

aciroth
Posts: 750
Joined: 09 May 2010 23:28

Re: negative results

Post by aciroth » 20 Dec 2017 12:30

Hm, it is indeed maybe difficult to discuss this over the forum
negative impacts are obviously rooted in the use of by-products in some processes
If a byproduct is used in another process without allocation, and the byproduct-process has some negative impacts, then this can overrule other results.
Let me know if we should discuss directly.
Best wishes

Post Reply

Who is online

Users browsing this forum: No registered users and 25 guests