Public

Deterministic multi-site ranking/optimization under constraints (Phase 3)

Rank candidate sites under hard constraints + signed weighted objectives (+maximize/-minimize). Three scoring modes: relative (min-max within batch, default), absolute (fixed 0-100, cross-run-stable), percentile (vs the viable-site population — 'better than X% of viable sites', cross-run + cross-region comparable). Returns the DCHubEnvelope with _entity=ranked_sites: rank, objective_score, per-field normalized{}, normalization_basis.

post/api/v1/rank-sites

Request

  • Base URL: https://dchub.cloud
  • URL: https://dchub.cloud/api/v1/rank-sites
  • Auth: none declared

Request body

candidatesobject[] required

Pre-enriched candidate objects {id?, lat?, lng?, <metric fields>}; carry site_evaluation_handoff through

constraintsobject

Hard filters {field: {min?, max?}}, fail-closed on a missing field

objectivesobject required

{field: signedWeight} — +weight maximizes, -weight minimizes

absoluteboolean

Fixed 0-100 scale (cross-run-stable)

percentileboolean

Percentile vs the viable-site population (takes precedence over absolute)

top_kinteger

Response

Ranked sites (_entity=ranked_sites)

_entity'facility' | 'market' | 'grid' | 'fiber' | 'gas' | 'deal' | 'site' | 'news' | 'energy' | 'incentives' | 'risk' | 'index' | 'pipeline' | 'infrastructure' | 'export' | 'changes' | 'alert' | 'meta' | 'semantic_search' | 'error' | 'record' required

Payload type discriminator — branch on this before parsing.

okboolean

Success flag; false + _entity='error' on failure.

_sourcestring
_citestring

Attribution string (CC-BY-4.0).

next_sessionobject

Context-aware next-step hints — the agent's state-machine menu for what to call next.

Example response

{
  "_source": "DC Hub — dchub.cloud"
}

Changes

No recorded changes to this operation across all 1 revision of this API.