Leaderboard - Challenge 2 - Trial Event 2

Teams

The following teams participated in Trial 2.

Team Name Lead Affiliation Team Lead Members        
ARPA-e Benchmark Los Alamos National Laboratory (LANL) Carleton Coffrin          
CasePower Case Western Reserve University Vira Chankong Hanieh Agharazi Kenneth A Loparo Kamlesh Mathur Pitjaya Tangtatswas Fan Zhang
GERS USA GERS USA Juan Manuel Gers Camilo Acosta David Alvarez Garzon Caceres Sergio Rivera Diego Rodriguez
GMI-GO Georgia Institute of Technology Andy Sun Santanu Dey Amin Gholami Kaizhao Sun Shixuan Zhang  
GO-SNIP Lehigh University Frank Curtis Daniel Molzahn Andreas Waechter Ermin Wei Elizabeth Wong  
Gordian Knot Virginia Tech Vassilis Kekatos Mohammadhafez Bazrafshan Sarthak Gupta      
GOT-BSI-OPF Global Optimal Technology, Inc. Gilburt Chiang Hsiao-Dong Chiang Bin Wang Simon Wyatt    
GravityX individual Hassan Hijazi          
magos Lawrence Livermore National Laboratory (LLNL) Ignacio Aravena-Solis Nai-Yuan Chiang Quentin Lété Shmuel Oren    
Monday Mornings Lawrence Berkeley National Laboratory (LBNL) Miguel Heleno Alexandre Moreira da Silva Alan Valenzuela Meza      
NU_Columbia_Artelys Artelys Richard Waltz Daniel Bienstock Jorge Nocedal      
Pearl Street Technologies Pearl Street Technologies Marko Jereminov David Bromberg Larry Pileggi Athanasios Terzakis Hui Zheng  
SUGAR-CMU Carnegie Mellon University Aayushya Agarwal Timothy McNamara Amritanshu Pandey      
Tartan Buffs University of Colorado Boulder Kyri Baker Constance Crozier Javad Mohammadi      

Rankings

Scoring according to the Scoring document, summarized as

For each scenario s, a prior point solution is constructed by keeping all variables fixed to their values in the prior operating point in the base case and the contingencies, projecting first the base case and then the contingencies to ensure feasibility of all hard constraints (details in the formulation document). Let MSpps denote the market surplus of the prior point solution for scenario s. This value is computed independently from the solutions provided by competitors.

For any entrant, the solution to scenario s is evaluated and a market surplus MStotals is assigned. If no solution is returned, or the solution is incorrectly formatted, or the solution is determined to be infeasible, or the evaluated market surplus is less than MSpps, then the assigned market surplus is MSpps.

The score MSgains, representing the gain in market surplus relative to the prior point, is computed as

MSgains = MStotals - MSpps

The score over a given set S of scenarios is

MSgain = ∑iS MSgains

Ensemble is the set of best scores for a Division. The rankings were updated June 7, 2021, after reevaluating scenarios impacted by platform issues.

MSgain for all 68 Trial 2 scenarios
Division 1   Division 2   Division 3   Division 4
ensemble 3,837,022,102      ensemble 3,845,656,920      ensemble 3,842,904,210      ensemble 3,934,049,141
Gordian Knot (VaTech) 2,470,036,976   GMI-GO (GaTech) 2,826,182,551   Gordian Knot (VaTech) 2,518,991,804   GMI-GO (GaTech) 2,825,894,212
Tartan Buffs (Colo.) 2,377,089,334   Gordian Knot (VaTech) 2,780,865,623   GMI-GO (GaTech) 2,330,501,443   Gordian Knot (VaTech) 2,781,867,601
GMI-GO (GaTech) 2,330,291,160   Tartan Buffs (Colo.) 2,339,925,937   ARPA-e Benchmark 2,020,368,479   ARPA-e Benchmark 2,156,629,064
ARPA-e Benchmark 2,014,906,062   ARPA-e Benchmark 2,156,388,467   GravityX 1,487,324,488   GravityX 1,809,300,649
Pearl Street Technologies 1,391,721,333   Pearl Street Technologies 1,393,651,045   Pearl Street Technologies 1,391,721,333   Pearl Street Technologies 1,393,651,045
GOT-BSI-OPF 1,388,966,716   GOT-BSI-OPF 1,390,690,032   GOT-BSI-OPF 1,390,241,597   GOT-BSI-OPF 1,389,622,503
SUGAR-CMU 776,628,587   SUGAR-CMU 1,274,934,074   Tartan Buffs (Colo.) 1,381,855,891   Tartan Buffs (Colo.) 1,382,807,021
GravityX 583,639,778   GO-SNIP (Lehigh) 920,172,686   GO-SNIP (Lehigh) 494,559,040   SUGAR-CMU 1,274,934,074
GO-SNIP (Lehigh) 494,553,988   GravityX 492,620,283   SUGAR-CMU 291,216,524   GO-SNIP (Lehigh) 920,159,704
magos (LLNL) 55,581,023   GERS USA 1,853,807   NU_Columbia_Artelys 281,439,834   NU_Columbia_Artelys 276,489,919
GERS USA 1,853,807   NU_Columbia_Artelys 1,484,889   GERS USA 1,853,807   GERS USA 1,853,807
NU_Columbia_Artelys 1,484,889   magos (LLNL) 2.49E-04   magos (LLNL) 6.34E-05   magos (LLNL) 1.58E-04
CasePower 0   CasePower 0   CasePower 0   CasePower 0
Monday Mornings (LBL) 0   Monday Mornings (LBL) 0   Monday Mornings (LBL) 0   Monday Mornings (LBL) 0
 

Additional information is available.