Prioritized action plan

camber.actionplan fuses three layers into one operator punch list: what's wrong (FDD findings), what it costs (camber.fault_economics), and what to do (camber.aso). Each actionable finding becomes an ActionItem with its estimated annual dollar impact and its advisory recommendation, ranked worst-dollars-first (severity breaks ties).

flowchart LR
  findings["FDD findings"] --> bap["build_action_plan"]
  econ["fault_economics ($/yr)"] --> bap
  aso["aso (what to do)"] --> bap
  bap --> item["ActionItem (equip, rule, cost, recommendation)"]
  item --> rank["rank worst-dollars-first (severity ties)"]
  rank --> plan["prioritized action plan"]
  plan --> out["action_plan_html / audit report section"]

Three layers fuse per finding, then rank into one prioritized punch list.

from camber.actionplan import build_action_plan, action_plan_html
from camber.fault_economics import EnergyPrice, EquipmentLoad

plan = build_action_plan(
    findings,
    loads={"AHU-1": EquipmentLoad(heating_capacity_kbtuh=200)},
    price=EnergyPrice(electricity_per_kwh=0.15),
)
for a in plan:
    print(
        a.severity,
        a.equip,
        a.rule,
        f"${a.annual_cost_usd:,.0f}",
        "→",
        a.recommendation.title if a.recommendation else "",
    )

Each ActionItem: equip, rule, severity, annual_cost_usd, costed (False when sizing is missing — the dollar figure is then omitted, never fabricated), recommendation (a camber.aso.Recommendation or None), and the source finding. action_plan_rows(plan) gives JSON/table rows; action_plan_html(plan) a self-contained table. Flags: loads, price, params (cost models), aso_params (recommendation targets), min_severity, costed_only.

In the audit report

AuditReport.to_html(recommend=True, loads=…, price=…) appends a Recommended actions section (the same ranked plan) beneath the findings; AuditReport.action_plan(...) returns the items directly. Everything stays advisory and read-only toward the BAS.

In config-driven runs

A declarative run can emit the action plan too: set "recommend": true (and optionally "price": {"electricity_per_kwh": 0.15}) in the config's report block, and the HTML report (out_html) includes the ranked Recommended actions section.