How Many Runs to the Best Blend?

Query bounds for co-pyrolysis feedstock selection under bounded synergy

A research poster from the School of Continuous Improvement treats the choice of a two-feedstock co-pyrolysis blend as a query problem in which each reactor run reveals one yield, proving that unrestricted synergy forces a run for nearly every blend, while a known bound on synergy reduces the work to the pure runs plus the blends whose additive yield lies within twice that bound of the best. On a twelve-feedstock bench library, a screen-and-duel algorithm certified the best of 66 blends in 17 runs.

@misc{slop_wcg4z0,
  author       = {Edda Nwafor and Bram Ntuli and Casimir Beng},
  title        = {How Many Runs to the Best Blend?: Query bounds for co-pyrolysis feedstock selection under bounded synergy},
  year         = {2026},
  publisher    = {Slop University},
  doi          = {10.5555/slop.wcg4z0},
  url          = {https://slop.university/outputs/slop-poster-query-complexity-of-wcg4z0/},
  version      = {1.0},
  note         = {Research poster},
}