FIELD BY FIELD · PROPOSED CARD vs THE API

What exists, and what does not

Every figure on the proposed card, checked against the AssetGroup DTO the grid already binds to. Seven are re-dressing. Eight are new work, and they are not evenly sized.

Already served
7no API change
New, cheap
4aggregates and rollups
New, expensive
4new sources or history
Blocked on time
1the 90-day trend

Already in the DTO7 — presentation only

These are what Jaya means. The card can be rebuilt around them with no backend change at all.

units · modelscount, modelCount
health 12%fleetHealthPct
At risk 8,644 · 82% of fleetatRisk / count
Under contract 0.5% · 51 units · 10,549 gaponContract / count
Replace cost $143.1MtotalReplacementCost
the “Critical” badgeclinicalImportance
the model listtopEquipment[]

Not in the API8 — new work

Four are ordinary additions to the group query. Four need something that does not exist yet.

4 vendors · avg age 8.2 yrTwo aggregates on the group query. Cheap.
↓4 pts 90dA health-history table. Nothing stores yesterday’s fleetHealthPct, so this needs a snapshot and a daily job — and shows nothing until 90 days of it exist.
$38.2M due <12 moReplacement cost filtered by end-of-life / useful-life date.
347 open work orders · 62 past SLAA work-order join, and a definition of “past SLA”. No sign of either near AssetGroup.
Why units are at risk — 4 causes with countsAt risk decomposed by cause. Today atRisk is a single number; this needs past-end-of-support, beyond-useful-life, unsupported-OS and open-recall counted apart.
per-model units, at-risk %, end of supporttopEquipment carries the name and manufacturer only.
“207 other models”modelCount minus the four shown. Cheap, but needs the rollup’s own figures.
the footer alertA derived rule — past support and no contract, grouped by manufacturer.
the one that is not a build item

↓4 pts 90d is not expensive to write — it is expensive to have. Nothing in the schema records what fleet health was yesterday, so the work is a snapshot table plus a daily job, and then ninety days of waiting before the card can show a real number. Anything promised on it should be dated from when the snapshot starts, not from when the code merges.

the one that may have no source

347 open work orders · 62 past SLA appears nowhere near AssetGroup, and “past SLA” is not defined anywhere in the card. Before this is scheduled, someone has to confirm the product holds work-order data at all, and what an SLA breach counts as. It is the single largest unknown in the document.