JuMP and HiGHS join forces to improve open energy modeling
announcements, open-energy-modeling ·Author: Miles Lubin, Carleton Coffrin, Oscar Dowson, Julian Hall, and Changhyun Kwon
The JuMP Steering Committee is pleased to announce that we, through NumFOCUS, have received a large grant from the Breakthrough Energy Foundation to improve the performance of JuMP and HiGHS on open energy models.
In our recent post, Steering Committee changes, we announced that Juan Pablo was stepping and Julian Hall (the lead developer of HiGHS) was joining the committee: this grant was the reason!
What this grant will be used for
Our main goal for this grant is to collate a suite of benchmark LP and MIP model instances from JuMP-based open energy models such as GenX.jl, TulipaEnergyModel.jl, Sienna, and SpineOpt.jl, and then use the benchmarks to guide improvements in JuMP and HiGHS.
Among other things, we will be parallelizing the HiGHS MIP solver and improving the debugging experience of JuMP by adding generic IIS support.
Project partners
HiGHS is world-leading open-source software for mathematical optimization, created by Julian Hall and Ivet Galabova from solvers written by University of Edinburgh PhD students. Development began in 2018, and a MIP solver written by Leona Gottwald was added in 2022.
JuMP is an open-source domain-specific algebraic modelling language for mathematical optimization embedded in Julia. JuMP development began in 2013. JuMP was recognized by the 2015 COIN-OR Cup, the 2016 INFORMS Computing Society Prize, and the 2021 Beale–Orchard-Hays Prize.
The agreement runs for one year, and the people being funded are:
- Dr. Julian Hall, the lead developer of the HiGHS solver
- Dr. Ivet Galabova, the HiGHS Integration and Development manager
- Dr. Joaquim Dias Garcia, a core contributor to JuMP
- Dr. Oscar Dowson, a core contributor to JuMP, and also a member of the Steering Committee.
Next steps
If you are an open energy modeller who uses JuMP or HiGHS and you want to stay
in touch with our progress or provide us with feedback and examples, write to
jump-highs-energy-models@googlegroups.com. We’d love to hear how you are using
JuMP or HiGHS to solve problems related to open energy modelling.
The first part of our work will focus on creating the benchmark examples in the jump-dev/open-energy-modeling-benchmarks repository. Please open an issue if you have ideas or suggestions.
We will also be posting monthly updates to this blog under the #open-energy-modeling tag.