Generated by the worldenergydata field-development playbook (epic #567). Recommendation is heuristic (Concept Select / FEL-1 fidelity), not a sanctioned design.
The engine's top recommendation (Tlp) matches the concept actually built (Tlp). The recommendation is produced on a blind probe — the as-built concept is stripped before scoring — so the agreement reflects the depth, basin and host-proximity signal, not the answer.
| Concept | Score | Trees | Rationale / flags |
|---|---|---|---|
| tlp ★ | 0.783 | dry | typical of gom development practice |
| semisub_fps | 0.782 | wet | typical of gom development practice |
| spar | 0.771 | dry | typical of gom development practice |
| subsea_to_shore | 0.688 | wet | — |
| fpso | 0.602 | wet | — atypical concept for gom |
★ = the concept actually selected for this field (where known).
Sources: SubseaIQ field catalog (~2014) + BSEE + curated vessel/ subsea catalogs. Schematics are deterministic, generated from the field concept.