myUmberto - Forum

Umberto LCA+ => Data (ecoinvent v3, GaBi) => Topic started by: pbeilschmidt on December 11, 2015, 09:59

Title: What's new in ecoinvent v3.2?
Post by: pbeilschmidt on December 11, 2015, 09:59
I am currently previewing (beta testing) the implementation of ecoinvent v3.2 LCI data in Umberto NXT LCA.

ecoinvent has published the datasets on Nov 30, 2015 and ifu Hamburg (just like other LCA software makers) now has the task to include the datasets in their software, test them in their LCA models and ship the new software package with ecoinvent v3.2 database to their customers.

An official shipping date for Umberto NXT with the new database has not yet been announced. Christmas holiday season is just not the best time of year to deliver a new product to the market...

Here are a few impressions regarding the new database. This is a subjective selection only. Those of you interested in a comprehensive summary of the changes and new items in ecoinvent v3.2 should read the official Report of Changes ( provided on the ecoinvent website. [Moreno Ruiz E., Lévová T., Bourgault G., Wernet G., (2015). Documentation of changes implemented in ecoinvent database 3.2. Ecoinvent Centre, Zürich, Switzerland]

It should also be noted that the system model that has been previosly recommended as the default for allocation has been renamed in ecoinvent v3.2 to system model 'Allocation at the point of substitution' (abbreviated as "APOS"). The APOS system model is explained on the ecoinvent website ( "The APOS system model was previously called "Allocation, ecoinvent default". The name was changed with the release of ecoinvent version 3.2."

It was probably not a light-hearted decision by the ecoinvent experts to return to advising the use of the cut-off approach: "For users new to LCI databases, ecoinvent recommends using the cut-off system model when starting to work with ecoinvent version 3. This model can be easier to understand for first time users. The cut-off system model applies the same modelling choices as ecoinvent version 1 and 2. This system model is therefore also recommended for experienced users looking for continuity with their previous work."

As for the impact assessment methods (LCIA) it is worth mentioning that the implementation of the IPCC2013 method in the ecoinvent v3.2 release has been modified, and that its application will most likely yield results different to the ones of the the IPCC2013 method provided in ecoinvent v3.1. A more detailed description of this issue can be found in this post (,1673.0.html).

This page on the ecoinvent website ( is also a good starting point to learn what's new in ecoinvent v3.2
Title: Re: What's new in ecoinvent v3.2?
Post by: uniquename on January 20, 2016, 11:14
Let's have a look at the new feature "market groups". They will make working with unit datasets a lot easier.

Unit datasets can be used to understand and examine where the impact comes from and as a starting point if you want to create your own datasets by adapting the existing ecoinvent processes. However, in ecoinvent v3.1 there were many unit datasets with MANY flows with the same name, e.g. electricity. One market in ecoinvent can never overlap another market. So a global dataset [GLO] automatically would include many inputs.

Now with ecoinvent v3.2, enter the market groups.
Market groups are a new type of datasets. They consist of markets that are oftentimes used together. As ecoinvent puts it in their "Documentation of changes implemented in ecoinvent database 3.2" (

"In  Market  Groups,  all  markets  contained within  their  region  are
combined and scaled based on their relative production volumes. Therefore, they are a consumption
mix for their region."

Market groups are available for materials that more or less every transforming or treatment activity needs and which are available from market datasets with a variety of geographies. These materials include amongst others electricity, water, diesel and other fuels.
The market groups are available for regions that are self-explanatory like [GLo], [US] or [CN] (China).
Please have a look at the attached picture that shows an excerpt from the unit dataset market group for electricity, medium voltage [ENTSO-E].
ENTSO-E describes the "European Network of Transmission System Operators for Electricity", on their website they list their member countries and companies:

Summing it up: market groups make working with unit datasets a lot more comfortable than before. One market group dataset instead of several identical materials is very useful.