plan op• Data kinds: none → design (an op determined by its arguments alone — it takes no image or data input)
• Call: import fullseye as fs; fs.ledger.doe_box_behnken(factors=3, center=3) -> 'dict' (to call the implementation directly, import doe; doe.doe_box_behnken(factors=3, center=3) -> 'dict'; from the registry, opsmultivar.get("doe_box_behnken"))
Box-Behnken design (Box & Behnken 1960) for 3, 4 or 5 factors -> design.
Three levels per factor and no run at a corner of the cube: for every pair of factors, a 2^2 factorial (±1, ±1) with all other factors at 0, plus center centre runs — 12, 24 and 40 edge runs for k = 3, 4, 5 (the pair construction of Box & Behnken's table for these k). Useful when the extreme corners are unsafe or cannot be set. k = 4 is rotatable. Larger k use incomplete-block constructions that are not reproduced here and are refused.
• design_of_experiments family guide
• Sample-data catalog (download URLs / licences) — 2-D uses skimage.data (BSD/public domain) plus synthetic images; 3-D lists download URLs for real data sources (Stanford, PDS, …).
• Operator provenance and references — the sources of the research/methods this op family came from.
• The canonical algorithm (author, year) and its uses are named in the family usage guide above.
• doe_plan_tour — py -3.11 examples/doe_plan_tour.py
design as input)doe_decode · rsm_fit · doe_main_effects · sens_morris · sens_sobol
plan)doe_lhs · doe_orthogonal_array · doe_fractional_factorial · doe_central_composite
*Provenance: doe.py — MULTIVAR operator registry. This per-op note is generated by tools/opdocs.py md (do not hand-edit).*
© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.