Case studies/user logos

[odow]

I've been talking to a lot of JuMP users recently. I think we could do a lot better with advertising the people that are happily using JuMP. Making a list of people that I've talked to recently

Academia

• DTU Denmark (Thomas Stidsen, et al, https://orbit.dtu.dk/en/persons/jalal-kazempour)
• University of Setif Algeria (Bouras Ikram)
• Ecole des Ponts
• Mines Paris
• Nantes
• President of ROADEF (https://homepages.laas.fr/sungueve/)
• TUDelft (https://www.tudelft.nl/en/eemcs/the-faculty/departments/software-technology/algorithmics/projects/nextgenopt)
• UW Madison

Government

• LANL
• NREL (Sienna, Reopt)
• ANL
• SINTEF
• CSIRO
• NZ EA (https://www.emi.ea.govt.nz/Wholesale/Tools/JADE)

Industry

• AtOptima
• Renault (https://github.com/LouisBouvier/InventoryRoutingLNS.jl, https://arxiv.org/abs/2209.00412)

Energy system models

• GenX https://github.com/GenXProject/GenX
• Sienna https://www.nrel.gov/analysis/sienna.html
• AnyMOD.jl https://github.com/leonardgoeke/AnyMOD.jl
• SpinneOpt.jl https://github.com/spine-tools/SpineOpt.jl
• @sstroemer's stuff: https://live.juliacon.org/talk/C7DFSF (Is this the same as NextGenOpt?)
• E4ST https://pretalx.com/juliacon2023/talk/LUVDCY/
• https://www.tudelft.nl/en/eemcs/the-faculty/departments/software-technology/algorithmics/projects/nextgenopt

[sstroemer]

My small input for that:

• NextGenOpt is not ours, as far as I know it's lead by Mathijs de Weerdt (TU Delft), but I don't know them (my affiliation with TU Delft is with the TPM faculty).
• "Our" model will soon (TM) be open-source under "IESopt", affiliation for that would be "AIT Austrian Institute of Technology", which could be similar to SINTEF, LANL, ... in the government section (we are > 50% government owned).
• I know a few "outdated" models (CapacityExpansion.jl, NemoMod.jl, POMATO)

[odow]

as far as I know it's lead by Mathijs de Weerdt

Yeah I've been talking to some people involved. @abelsiqueira is also part of the team.

[sstroemer]

Sorry, one more that I remembered: PowerModelsACDC.jl from KU Leuven

[joaquimg]

I like the logos idea.
Quotes from these people and more could be nice as well.

[odow]

JuMP (or Julia) was used by at least three teams in https://gocompetition.energy.gov
ARPA-e Benchmark
TIM-GO
quasiGrad

[abelsiqueira]

@datejada, @clizbe, you can add the model (and TNO's name?) here, if you're up for it.

[matbesancon]

In academia: Zuse Institute Berlin, TU Berlin

[datejada]

In Energy System Models: TNO and the eScience Center are developing TulipaEnergyModel.jl https://github.com/TulipaEnergy/TulipaEnergyModel.jl

[matbesancon]

@stelmo @exaexa can we mention HH Düsseldorf and U. Luxembourg?

[stelmo]

Yes! At the Heinrich Heine University in Düsseldorf (me) and University of Luxembourg (@exaexa) we are happily using JuMP to build constraint-based reconstruction and analysis software (e.g. https://github.com/LCSB-BioCore/COBREXA.jl).

[stelmo]

In fact, @exaexa recently built an entire framework to make LP/QP modeling simpler to connect to JuMP and all of its solvers: https://discourse.julialang.org/t/ann-constrainttrees-jl-tidy-constrained-systems-for-jump/105563

[exaexa]

Hi all, confirming that we use a lot of JuMP too :D (thx @matbesancon @stelmo for ping)

• In academia: LCSB at Uni.Lu. Please take logos from https://github.com/COBREXA/ConstraintTrees.jl/tree/master/docs/src/assets . Links: https://github.com/lcsb-biocore/COBREXA.jl/ and https://github.com/LCSB-BioCore/FBCModelTests.jl
• What would be the way to report the PerMedCoE EU H2020 project here? Technically that is the framework that enabled a major part of the above developments, and will continue using JuMP in European HPCs. (Just in case, the logo is in the same folder as above)

Thanks!

[clizbe]

Clarification: TulipaEnergyModel.jl is the model in the NextGenOpt project, involving TNO, eScience, TUDelft (Mathijs de Weerdt), and Utrecht University.

[odow]

@clizbe oops, sorry. Fixed here: #121

[clizbe]

Oh no! That's not what I meant. It's indeed TNO's model - but just clarifying that "the NextGenOpt model" that people were trying to remember is not a separate model, it's Tulipa.