Steering Committee · May 2026
Agentic AI Pilot — Merchant ASU Fleet
Phase 2 Proposal · $1.8M · 6 months · 1 plant · Read-only · Kill switch at month 6
"Helping customers advance their business performance and reach their full potential."
Airgas
An Air Liquide Company
Steering Pack · May 2026
Executive Pre-Read · For Steering Committee

Approve $1.8M to buy an option on $16–23M/yr — with a kill switch.

The 35-plant merchant ASU fleet leaks $2.5–4.0M per plant per year across four measurable cost pools. The leaks sit across silos — APC cannot close them. Phase 1 is already delivered. Phase 2 is a six-month, single-plant, Finance-gated pilot of the coordination layer. Human-in-the-loop, read-only, kill switch at month 6.

Ask$1.8M ceiling Duration6 months Plant1 ASU GateMonth 6 · four criteria Decision byEnd of Q2 2026
Pilot budget
$1.8M
ceiling · no overruns
At stake
$16–23M/yr
fleet run-rate
Per-plant leak
$2.5–4.0M
across 4 pools
Phase 1 delivered
18 agents
+ 13-agent RCA
Twin score
8.68/10
vs. Aspen HYSYS
Payback (base)
~4 mo
on 1-plant pilot

00 Decision Required

This proposal originates from the LMB E2E Value Initiative. By moving beyond reactive, single-silo optimization, agentic AI activates our largest latent asset — the Airgas operating dataset — and converts it into a coordinated decision layer across plant, logistics, reliability, and energy. The ask is narrow, time-bound, and reversible.

Exhibit 1 · Decision at a Glance
Ask
Approve $1.8M for a 6-month agentic AI pilot on 1 ASU.
Cost
$1.8M ceiling. No overruns.
Sponsor
To be named · cross-silo authority over IT, Ops, Finance
Duration
6 months. July / August 2026 kickoff.
Kill switch
$450K/plant Finance-validated annualized savings by month 6, or program ends.
Downside
$1.8M maximum. No fleet commitment implied.
Upside
$16–23M/year run-rate at fleet scale · 12–18 month payback.
Decision by
End of Q2 2026.
The downside is capped at $1.8M. The planning base returns four-month payback on the single-plant pilot. The hard question is not whether we can afford to try — it is whether we can afford to let a competitor demonstrate coordinated fleet operations before we do.

01 The Economics

Each plant in the 35-plant merchant fleet leaks $2.5–4.0M per year across four measurable pools. None are hypothetical. Scenario ranges are re-anchored at a day-30 baseline lock with Finance before any value is booked. The program is NPV-positive over three years even at the 50% capture sensitivity. At the planning base, the pilot pays back in roughly four months on $450K/plant annualized at the pilot ASU. Full fleet payback runs 12–18 months including Phase 3 buildout.

Exhibit 2 · Three scenarios (annualized, fleet-wide at steady state)
Downside · 30% capture

Structural floor

$4.7M / yr

3-year NPV positive. No fleet rollout assumed. Pilot justifies itself on its own math.

Planning base · 50% capture

What we plan against

$7.9M / yr

4-month payback on pilot · 12–18 month payback on fleet rollout. Justifies Phase 3 authorization.

Target · 100% capture

Strategic north star

$16–23M / yr

Full addressable across four margin pools. Requires Phase 3 pass + >80% operator adoption.

Exhibit 3 · Per-plant value drivers (per year)
Value driver Type Low ($K) High ($K) Notes
Demand forecasting — fewer emergency spot purchasesHARD150200Medical O₂, fab ramps, refinery ramp-ends
Boil-off reduction via tank & load optimizationHARD120160Measurable via SCADA historian
Energy time-of-use arbitrage (power ≈ 60% of ASU cost)HARD100150Finance model tab: TOU-Arb
Unplanned downtime avoidance (predictive maintenance)AVOIDANCE80120Engineer + Finance co-signed
Route & load optimization (outbound logistics)HARD5090TMS feed · Grade A / B only
Operator productivity (excluded from headline)SOFT3050Upside narrative only · not booked
Conservative band used in business case $450$650 10% haircut applied to row totals

Hard savings are Finance-defensible (energy, yield, throughput). Avoidance savings are engineer-plus-Finance co-signed. Soft savings are excluded from the kill-switch gate and never blended in the headline figure.

02 What We Have Already Built

Phase 1 is done. It delivered a high-fidelity digital twin of the ASU — a simulator that models production, tanks, trucks, and customers the way an Aspen model does, but faster and with live decision layers attached. On top of it sits an 18-agent coordination layer. Scored against an Aspen HYSYS reference across ten weighted criteria, the twin scored 8.68/10. The weakest dimension was validation coverage at 4.0/10 — the reason the pilot exists.

→ What is built
  • Full-physics ASU digital twin: MAC, PHX, HP/LP distillation, argon, condenser-reboiler, storage, loading
  • 9 LLM coordination agents: per-plant sensor / process / economist; fleet-wide Hub, Demand Forecaster, Fleet Route Optimizer, Financial Controller, CFO, Ops Director
  • Three coordination modules: MPC plant optimizer, GCN graph-based logistics optimizer, SwarmCore orchestration layer
  • 13-agent RCA subsystem — Finance-grade incident report in minutes
  • Delivered in-house at zero incremental project cost
× What Phase 1 is NOT
  • Not a live deliberation loop against real plant signals
  • Not a closed-loop optimizer — agents are advisory, not actuating
  • Validation coverage against a specific live plant: 4.0/10 (weakest dimension)
  • Not production maturity — prototype-to-pilot transition only
  • No booked savings. Simulation evidence only until Finance validates live-plant data
Simulation is not booked savings. A team that tells you differently is selling you something. The pilot exists to convert simulation-grade evidence into live-plant evidence under Finance validation.

03 The Kill Switch

Before the upside — how this program ends if it does not work. The program is designed to kill itself cleanly. That is the feature, not a caveat. Four gate criteria at month 6; all must clear simultaneously.

GATE 01
$450K/plant Finance-validated

Annualized savings against day-30 locked baseline. Grade A (invoice / meter) & Grade B (engineer + Finance co-signed) only. Co-signed by the Steering Committee.

GATE 02
Zero safety incidents

No safety incidents attributable to agent recommendations across the full pilot duration. Non-negotiable. Program ends immediately on breach.

GATE 03
≥ 70% operator acceptance

Acceptance rate measured weekly from month 2 onward. Expect 40–50% at month 3, 70%+ at month 6. Ramp curve, not a flat line.

GATE 04
Documented site sequencing

Per-plant business cases and Phase 3 sequencing plan submitted to Steering by end of month 6. Pass triggers separate Phase 3 authorization.

Failure on any one criterion triggers immediate program end. Only Grade A and Grade B savings count toward the gate. Grade C (modeled) is upside narrative only.

04 The Pilot Shape

$1.8M · 6 months · 1 plant · operator-in-control · read-only · on-prem, behind the firewall. Every recommendation carries three things: a reason in one sentence, a confidence as a number, and a dollar impact. The system writes no setpoints. The operator approves, defers, or rejects.

Parameter Design
Pilot ceiling$1.8M · Grade A + Grade B savings only count toward the gate
Duration6 months · month-6 kill-switch review
Plant1 ASU · selected by operations feasibility and instrument density
ModeRead-only · zero setpoints written · recommendations surfaced to operator queue
ArchitectureOn-prem · behind the firewall · engineering-enforced (not policy-enforced)
Per-plant budget$450K · Finance-validated savings target annualized
Operator roleIn control · every recommendation requires explicit approval
Gate reviewMonth 6 · four criteria · all must clear
Go-liveJuly / August 2026 · requires Q2 2026 decision

05 Risks & Mitigations

Five material risks. Four are contained inside Phase 2 engineering scope. One — data governance — requires executive air cover from the sponsor.

Operator trust / low acceptance
MEDIUM
Transparent reasoning on every recommendation. Weekly acceptance review. 70% gate at month 6 catches this before scale.
Savings do not materialize at live plant
MEDIUM
Day-30 baseline lock with Finance. Grade A/B only. Kill switch ends program if $450K/plant is not hit by month 6.
Safety incident attributable to agent
LOW
Read-only architecture. No setpoints written. Engineering-enforced firewall. Zero-incident gate is non-negotiable.
Data governance & IT friction across silos
HIGH
Requires executive sponsor with cross-silo authority over IT, Ops, Finance. This is the one risk engineering cannot mitigate alone.
Competitor ships coordinated fleet first
MEDIUM
12–18 month window. Waiting one year costs the moat. Q2 2026 decision enables July kickoff on calendar.

06 The Competitive Arc

AI in industrial operations is already happening. Chemicals are ~50% deployed in production. Refining claims 30%+ productivity gains. Industrial gas is the sector that has not moved. Zero coordinated fleet deployments are publicly disclosed across Linde, Air Products, Air Liquide, or Nippon Sanso.

Four competitors occupy three quadrants — single-plant, single-agent AI. The upper-right quadrant — coordinated fleet operations — is empty. Not because it lacks value. Because it is hard: it requires cross-system data access, governance of operator trust, and a safety architecture that supports read-only agent recommendations at fleet scale. Hard = moat.

Chemicals
~50%

of manufacturers have deployed AI in production. Dow, BASF, SABIC run production-grade.

Refining · O&G
30%+

productivity gains reported. ExxonMobil, Shell, Chevron autonomous optimization.

Industrial gas
0

coordinated fleet deployments disclosed. Announcements, not shipped fleets.

07 Two Futures

The decision is which branch we are on by Q3 2026. Both are live paths; neither is free.

Branch A · Approve Q2 2026

Lead and set the benchmark

July kickoff on calendar. Month-6 gate data lands Q4 2026. If the gate clears, Phase 3 authorization request enters Q1 2027 on a de-risked footing. We are the reference case the industry studies.

Branch B · Wait 12 months

Pay the catch-up premium

A competitor announces coordinated fleet operations in 2027. We restart Phase 2 against a moving benchmark, without the moat window, and with a data-governance sponsor who now has to explain the delay. Same pilot, higher hurdle, less optionality.

08 The Ask

Three decisions, one sponsor, one date. All three must be taken together — partial approval stalls the program on the calendar we need.

01 · Approve
$1.8M pilot budget

Phase 2 against the already-approved agentic-AI envelope. Ceiling, not a forecast. By end of Q2 2026.

02 · Appoint
1 executive sponsor

Cross-silo authority over IT, Ops, Finance. The one risk engineering cannot mitigate alone.

03 · Fund
Q2 2026 release

Funding release before Q2 close — enables July/August kickoff with 6-month delivery on calendar.

$1.8M is the option premium on a $16–23M/yr run-rate. The kill switch makes the downside definite. This is not a bet on AI. It is a bet on the data we already own.
Next · Kickoff Jul 2026 · Gate Nov 2026 · Phase 3 authorization Q1 2027
Open full memo in Google Docs

09 Appendix A–F

A · Validation and Methodology. Digital twin scored 8.68/10 vs. Aspen HYSYS across ten weighted criteria; weakest dimension validation coverage at 4.0/10. Leakage estimates sourced from internal LMB E2E baseline diagnostic, cross-checked against ROI Workbook Baseline tab (per-plant revenue $64.2M, energy cost $15.7M, uptime 96.5%, 306.6 unplanned-downtime hrs/yr). Grading framework: Grade A invoice/meter · Grade B engineer + Finance co-sign · Grade C modeled (narrative only, never in the gate). Day-30 baseline locks with Finance after the sensor audit; synthetic-control plus pre/post method; exit write-down schedule signed at the same point.

B · The Agentic Stack. MPC optimizer recommends load/swing/product-split; GCN solves the logistics graph on every event; SwarmCore reconciles MPC/GCN conflicts with a written reason; DemandForecaster spots anomalies hours before dispatch; HubAgent runs hospital-first triage; CFOAgent prices every recommendation; OpsDirector ranks the operator queue by urgency × $impact; PlantProcessAgent and PlantSensorAgent run per-plant validation. Twelve curated injection scenarios exercise the stack end-to-end across production, storage, loading, transit, delivery, and billing.

C · Infrastructure and Cybersecurity (per Pilot Plant IT Briefing, May 2026). Per-plant edge inference rig $103.3K – $142.8K (low = current May 2026 estimate, high = supply-chain hardened ceiling absorbing memory shortages, GPU spot pricing, tariff and lead-time premiums). Compute subtotal $32.3K – $41.5K: Supermicro 2U dual-EPYC GPU server with 256 GB DDR5 ECC RDIMM and dual 25 GbE OCP NIC; NVIDIA RTX 6000 Ada 48 GB primary + hot spare (sized to run Gemma 4-31B at FP8 with KV-cache headroom and the 18-agent swarm concurrently, RTO < 15 min on GPU failover); 2× NVMe U.2 7.68 TB Gen4 (1 DWPD) for model weights, telemetry buffer, and 30-day historian hot tier. Network/power/enclosure: Fortinet FortiGate 100F (Purdue L3/L3.5 segmentation, OT→IT one-way ACLs, syslog forward to Group SOC); 10/25 GbE managed top-of-rack switch; structured cabling, fiber and patch panels; 6 kVA online double-conversion rack UPS; smart 208 V/30 A metered PDU; rack airflow and environmental monitoring; 24U half-rack cabinet; one-time 208 V/30 A plant-side electrical install. Services: Grafana + Prometheus monitoring stack; 40-hr field-engineer install and commissioning; Year-1 Supermicro/Dell ProSupport for chassis and GPU RMA; 15% hardware contingency. Open-weight models on Group-approved vendor list (Qwen 3.6 · Gemma 4) — no public-internet egress, ever. Alternatives considered (H100 PCIe/SXM5, MI300X, Groq LPU, Cerebras CS-3, consumer RTX 5090, refurb A100) all rejected on cost, cooling, ecosystem risk, or production-grade defensibility. Cyber posture: Purdue model L3/L3.5 enforced at the Fortinet firewall; one-way OPC UA tap from the historian into an industrial DMZ via a hardware data diode (electrically one-way, not software-filtered); zero outbound OT→IT or OT→internet during the pilot; all inference on-prem; tamper-evident hash-chained audit log of every agent recommendation and every operator response; external STRIDE threat model and penetration test at Phase 1 exit and re-run at Phase 3 exit.

C-bis · System-level software stack (per IT Briefing). All system-level software is OSS or free-from-vendor; license-bearing items isolated as managed subscriptions. Base OS / GPU / runtime: Ubuntu 22.04 LTS (CIS-hardened, AppArmor, Ubuntu Pro livepatch); systemd; NVIDIA driver 550+ · CUDA 12.4+ · cuDNN · NCCL mirrored from internal artifact registry; nvidia-smi + DCGM for GPU health; Docker Engine + Compose + Traefik for container runtime and TLS termination. Inference runtimes: vLLM (PagedAttention, continuous batching) as primary OpenAI-compatible serving with FP8 / AWQ paths; SGLang for structured generation and agent deliberation; NVIDIA TensorRT-LLM and Text Embeddings Inference (TEI) for low-latency production paths and bge-m3 reranker on the RAG pipeline. Observability: Prometheus (15s scrape, 30-day local retention, long-term forward to Group), node_exporter / DCGM exporter / snmp_exporter, Grafana operator+SRE dashboards, Alertmanager → Group on-call, Loki + Promtail to Group SIEM via syslog. Identity / secrets / transport: SAML/OIDC federated with Airgas AD; HashiCorp Vault (quarterly token rotation) + WireGuard for MFA-gated out-of-band access; TLS 1.3 + mTLS, certs from Group internal PKI via cert-manager. Backup / DR: Restic (encrypted, off-plant target), pg_dump / pgBackRest for PITR (5-min RPO), rclone mirror to Group archival. Year-1 managed subscriptions (recurring OpEx, per plant): $11.5K – $26K — Ubuntu Pro ($0.5–1.5K), Docker Business ($1–4.5K), NVIDIA AIE ($4.5–9K), Vault Enterprise ($2–5K, replaceable with OpenBao at zero cost), and a $3.5–6K misc/buffer for observability add-ons, certs, GitHub seats, and ad-hoc tooling.

D · Team and Build. 11 FTE blended (in-house + loaned + contractor) over 6 months at Philadelphia / Delaware Valley market rates (Q1 2026), fully loaded — $1.25M total. Lead ML/AI Engineer (1.0), Senior ML Engineer (2.0), Mid-level ML Engineer (1.0), MLOps/Platform Engineer (1.0), Senior SCADA/Controls Engineer (loaned 0.5), Data Engineer (1.0), OT Cyber Security Engineer (contractor 0.5), Plant Operations SME (loaned 0.5), Finance Analyst (loaned 0.5 — baseline lock + monthly savings validation), Safety/Process Engineer (loaned 0.5 — HAZOP, MOC, safety case), Evaluation/Red-team Engineer (1.0 — golden-set curation, regression harness, agent evals), Change Management Lead (1.0), Project Manager (0.5), specialized contractor buffer ($100K) and recruiting/onboarding/equipment ($40K). Salary sources: Salary.com, Built In, ZipRecruiter, ERI, Glassdoor (Philadelphia MSA, Q1 2026); contractor rates reflect $150–$300/hr senior AI/OT specialist market. Build-vs-buy: AspenTech GDOT is single-agent by design (no plant/logistics coordination layer); Seeq is analytics overlay not coordination; hyperscale agentic frameworks (Azure / AWS / GCP) require cloud egress structurally incompatible with Purdue L3/3.5 and the no-cloud-egress requirement Airgas customers and OT cyber reviewers will expect; vertical "agents for industrial" startups immature on safety-assurance, Finance-grade evidence, multi-agent coordination, or OT-native deployment. Build path secures the proprietary coordination layer, eliminates cloud egress, and improves unit economics at fleet scale. ADVANCE alignment memo on week-1 workstream — scope is complementary, non-duplicative with Group Digital & AI.

E · If the Gate Passes: Phase 3. Separate business case for 18-month fleet deployment across the 35-plant merchant fleet. First wave 5–6 plants selected on highest measured leakage and best data fidelity. Pilot team becomes the operational backbone — no incremental hiring in the Phase 3 staffing model. Independent gate criteria and independent Finance co-sign. If Phase 3 clears, the pilot template becomes the candidate blueprint for Group-wide deployment across the Air Liquide ASU footprint.

F · Risk Deep Dive. F1 Operator trust is the single biggest risk — observation-only scope, weekly co-design, dedicated change-management lead, no-blame policy signed by plant management and union, 70% acceptance gate measured weekly. F2 Data quality addressed by a 30-day sensor audit + calibration sprint before any agent activation. F3 Build-vs-buy defended on proprietary coordination, cloud-egress incompatibility, and unit economics at fleet scale. F4 Safety — zero control authority enforced architecturally, full assurance case per VDE-AR-E 2842-61 and ISA/IEC 61511, global kill switch drilled weekly under 30 seconds. F5 Full v2.1 risk register tracks nine named risks, each with a named owner and a triggering week on the project plan.

Full appendix tables, hardware BOM line-by-line, salary source list, and risk-register owners available in the editable Google Doc.

~/ cloudflare pages deploy ready · all assets relative