EARNED AI // v2.0
01 What02 Engines03 Pages04 Why
Spec sheet ↓Launch demo →
ANSI/EIA-748 · DOE 413.3B · DCMA 14-PT

Earned value,
measured/audited/narrated

A project-controls intelligence platform that ingests Primavera P6 schedules, recomputes EVM from first principles, quantifies risk with real Monte Carlo, and writes the monthly report for you — with every formula, float, and variance open to inspection.

Built from scratchCPM engineMonte CarloDCMA 14-PointXER parserClaude-generated narratives
Open live demo ↗Read the spec sheet
Instrument · Project selector
BAC
$2.40M
CPI
0.94
SPI
0.87
% Complete
62.4
100%75%50%25%0%0%25%50%75%100%T0
PV (plan) EV (earned) AC (actual)TODAY · DATA DATE
01 — WHAT IT DOES

Oversight, in three registers.

Mirrors DOE PMSO workflow
from CD-0 to close-out
I.
Cost / EVM oversight

Validate the contractor's numbers.

  • CPI/SPI trended against baseline
  • Cost-at-completion breach alerts
  • EAC comparison across three methods
  • TCPI feasibility, flagged automatically
II.
Schedule surveillance

Prove the schedule is real.

  • DCMA 14-Point health assessment
  • Critical-path integrity checks
  • Float anomalies & constraint abuse
  • Earned Schedule: SPI(t), SV(t)
III.
Risk & reporting

From probability to prose.

  • Monte Carlo cost & schedule bands
  • AI anomaly detection with severity
  • Portfolio roll-up and drill-down
  • Narrative reports generated via Claude
02 — CORE ENGINES

Six engines, hand-rolled.

Tap a row to expand
capabilities
6 engines · 32 capabilities · 0 dependencies on commercial EVM libraries
03 — APPLICATION SURFACE

Nine screens, one story.

Scroll horizontally
or use arrows →
Swipe for all nine
BAC
$2.4M
EV
$1.8M
CPI
0.94
SPI
0.87
01 · Home · KPIs
Executive Dashboard
Portfolio KPIs, S-curves, cash flow, cost breakdown, risk score, performance gauges.
K-GSF
72%
Yucca
45%
Rail
28%
Terminal
91%
02 · Portfolio · Cards
Project Portfolio
Every project as an interactive card with health indicator, key metrics, and risk status.
Foundation
Structural
Electrical
HVAC
Finishing
03 · Schedule · Gantt
Schedule View
Timeline Gantt with critical-path highlighting and earned-schedule metrics.
86
QUALITY
✓
✓
✗
✓
✓
✓
✓
✗
✓
✓
✓
✓
✓
✓
04 · Quality · DCMA
Schedule Quality
14-Point pass/fail grid with remediation guidance and composite quality score.
05 · Financials · WBS
Financial Analytics
Variance waterfall, WBS-level analysis, cost breakdown, performance indices radar.
EAC
$2.6M
ETC
$820K
VAC
−$180K
06 · Forecasting · EAC
Forecasting
EAC/ETC projections, trend analysis, scenario comparison, completion date ranges.
P10
$2.1M
P50
$2.4M
P90
$2.9M
07 · Risk · Monte Carlo
Monte Carlo
Histogram distributions, confidence intervals, criticality index, sensitivity.
CPI below 0.9 — cost overrun risk
Schedule variance trending negative
WBS 3.2 resource underallocation
Phase 1 on-track — milestone met
08 · AI · Insights
AI Insights
Narrative insights with severity levels, recommendations, impact scoring, filters.
CSV
AI REPORT
09 · Reports · Claude
Reports & Export
AI-generated narrative reports, CSV data exports, one-click download + copy.
04 — WHY IT'S BUILT THIS WAY

Every line is defensible.

No black-box calcs
No vendored formulas

Hand-built EVM engine, not a library wrapper

ANSI/EIA-748

Every formula — PV, EV, AC, CV, SV, CPI, SPI, three EAC methods, ETC, VAC, TCPI, and Earned Schedule — is implemented from first principles. In oversight work, you need to know whether the contractor's numbers are correct, not just what a tool says.

CPM engine with cycle detection

Schedule analysis

Forward pass, backward pass, total float, critical-path identification, near-critical flagging — written from scratch. Cycle detection catches circular dependencies that silently corrupt commercial schedule networks.

Full DCMA 14-Point assessment

DOE 413.3B

All fourteen checks: missing logic, open ends, hard constraints, negative float, high float, duration, invalid dates, resource loading, missed tasks, CP %, CPLI, BEI, lead/lag, relationship mix — the exact deliverable a federal oversight analyst runs during an IPS health check.

Real Monte Carlo, not ±5% theater

Risk quantification

PERT and triangular distributions, configurable iterations, P10 through P90 bands, criticality index, sensitivity ranking. CD-2 and CD-3 approvals require probabilistic range estimates — this understands the statistics behind them.

AI anomaly detection with severity

Surveillance

A rules engine flags CPI/SPI degradation, cost-at-completion breaches, variance inflection points, and WBS anomalies — each with severity, confidence, and a specific recommendation — rather than waiting for the monthly CPR to reveal the problem.

P6 XER import, native

P6 / IPS

Custom parser for Oracle Primavera's tab-delimited XER export maps activities, relationships, resources, and calendars into a type-safe model. Real project controls work starts with real schedule data.

Portfolio aggregation + drill-down

PMSO

Aggregated BAC, EV, AC, CPI, SPI across the portfolio, with cross-project variance ranking and risk scoring — rolled up for executive reporting, drillable to WBS detail for root-cause analysis.

Claude writes the report, you keep the judgment

Reporting

One-click executive narrative synthesizes EVM metrics, risk indicators, schedule status, and WBS variances into a structured 500-800 word analysis — an analyst-in-the-loop accelerator, not a replacement.

DCMA 14-Point, implemented in full.

Composite quality score · 86 / 100 · 12 pass · 2 fail
01
Logic
✓PASS
02
Leads
✓PASS
03
Lags
✗FAIL
04
Relationship types
✓PASS
05
Hard constraints
✓PASS
06
High float
✓PASS
07
Negative float
✓PASS
08
High duration
✗FAIL
09
Invalid dates
✓PASS
10
Resources
✓PASS
11
Missed tasks
✓PASS
12
Critical path %
✓PASS
13
CPLI
✓PASS
14
BEI
✓PASS

Built by a project controls analyst for project controls analysts.

Designed & developed by Eric Mitchell · EarnedAI v2.0 · 2026
Launch the demo →
TYPESET · INSTRUMENT SERIF / JETBRAINS MONO
ENGINES · TS · NEXT · CLAUDE