`MultiLCA` gives zero scores for single demand/multiple impact categories

[pjamesjoyce]

I understand the main use case for MultiLCA is multiple demands, multiple impact methods, but I get an odd behaviour when trying to run it for a single demand and multiple impact assessment methods. It runs with no errors or warnings, but gives a result of 0.0 every time.

For example, where EF31 is a list of the EF3.1 impact assessment methods:

EF31 = [('EF v3.1', 'acidification', 'accumulated exceedance (AE)'),
 ('EF v3.1', 'climate change', 'global warming potential (GWP100)'),
etc...

and demands consists of one item:

demands = {'isopropyl acetate production': {13507: 1}}

The following runs as a MultiLCA

method_config = {
    'impact_categories': EF31,
}

assert bc.method_config.MethodConfig(**method_config)

data_objs = bd.get_multilca_data_objs(functional_units=demands, method_config=method_config)

mlca = bc.MultiLCA(
    demands=demands,
    method_config=method_config,
    data_objs=data_objs
)
mlca.lci()
mlca.lcia()

but the results are all zero:

to_df = []
for (i, a), s in mlca.scores.items():
    to_df.append(dict(
        impact = i[1],
        activity = a,
        score = s
    ))
df = pd.DataFrame(to_df)
df

	impact	activity	score
0	acidification	isopropyl acetate production	0
1	climate change	isopropyl acetate production	0
2	ecotoxicity: freshwater	isopropyl acetate production	0
3	energy resources: non-renewable	isopropyl acetate production	0
4	eutrophication: freshwater	isopropyl acetate production	0
5	eutrophication: marine	isopropyl acetate production	0
6	eutrophication: terrestrial	isopropyl acetate production	0
7	human toxicity: carcinogenic	isopropyl acetate production	0
8	human toxicity: non-carcinogenic	isopropyl acetate production	0
9	ionising radiation: human health	isopropyl acetate production	0
10	land use	isopropyl acetate production	0
11	material resources: metals/minerals	isopropyl acetate production	0
12	ozone depletion	isopropyl acetate production	0
13	particulate matter formation	isopropyl acetate production	0
14	photochemical oxidant formation: human health	isopropyl acetate production	0
15	water use	isopropyl acetate production	0

If you add a second item to the demands e.g.:

demands = {'isopropyl acetate production': {13507: 1},
 'potassium carbonate production, from potassium hydroxide': {7843: 1}}

and then run exactly the same code, the scores appear for the original item:

	impact	activity	score
0	acidification	isopropyl acetate production	0.0180329
1	acidification	potassium carbonate production, from potassium hydroxide	0.0160523
2	climate change	isopropyl acetate production	3.68333
3	climate change	potassium carbonate production, from potassium hydroxide	3.07375

etc....

Is this expected behaviour? and if so, what's best practice for running multiple impact categories for one demand?

Thanks

[cmutel]

We have explicit tests for a single functional unit, so I'm not sure what is happening here. Can you check the same setup (same functional unit and first impact category) in the normal LCA? Otherwise, I would need more info on the way your data is setup to be able to debug anything...

[pjamesjoyce]

Sorry @cmutel - ignore this - whatever was causing it for me must've been been fixed between dev17 and dev21 - teach me to always update to the latest and greatest before posting an issue!