Financial Analysis — Cleveland Works¶
Mission: Bottom-up value and cost analysis for every Cleveland initiative. Each card defines the formula and variables — the math structure, not necessarily the final number. Numbers are tagged by source:
workshop-confirmed,estimated, orneeds-corporate. This is the audit trail: every dollar in the business package traces back to a card here.Methodology: Value per §3 of consolidation-plan.md (5-step: metric → current state → target → delta → sensitivity). Cost per §4 (effort-based: team composition × duration × rates).
Status: PILOT — template validation. Review before applying to other sites.
Last updated: 2026-04-16 (appendix preparation — no content changes, audit trail preserved as constructed)
How to Read This File¶
Initiative cards define the value formula and cost structure for each of Cleveland's 27 initiatives. They are grouped by parent site project (CLV-P01..P11).
Three card types: - Type A — Anchored: Formula defined, key variables have workshop-confirmed values. Math can be partially computed now. - Type B — Structured: Formula defined, but most variables need corporate data to populate. The math is built, the inputs are TBD. - Type C — Absorbed: No standalone formula. Value captured in parent project (enabler, seed, or subsumed scope).
Project roll-ups aggregate initiative cards into site project totals.
Corporate Inquiry Table (end of file) collects all needs-corporate variables into one table for IE to take to Cleveland-Cliffs.
CLV-P01: Ops-Maintenance Data Integration¶
CLV-01: Ops-Maintenance Data Integration¶
Card Type: A — Anchored Corporate Project: PRJ-01
Value Analysis¶
Value Types: Throughput gain + Efficiency gain Value Formula:
misattributed_delay_hours_per_month × production_value_per_hour × attribution_correction_rate
+ root_cause_resolution_speedup_hours × incidents_per_month × labor_cost_per_hour
+ repeat_failure_reduction_% × repeat_failure_annual_cost
| Variable | Value | Source | Status |
|---|---|---|---|
| misattributed_delay_hours_per_month | [TBD] | Cross-ref ops delay reports vs Tabware WOs | needs-corporate |
| production_value_per_hour | [TBD] | 1SP throughput × margin/ton | needs-corporate |
| attribution_correction_rate | 50-70% | Conservative — not all misattributions are recoverable | estimated |
| root_cause_resolution_speedup_hours | [TBD] | Current mean time to diagnose vs target | needs-corporate |
| incidents_per_month | [TBD] | Tabware work order volume | needs-corporate |
| labor_cost_per_hour | [TBD] | Blended maintenance tech rate | needs-corporate |
| repeat_failure_reduction_% | 15-25% | Industry benchmark for closed-loop maintenance | estimated |
| repeat_failure_annual_cost | [TBD] | Frequency × cost per event from delay reports | needs-corporate |
Workshop-Sourced Range: $2-5M/yr Confidence: Medium-High — validated by 7+ stakeholders across 5 transcripts Key Quote: "We need to be able to integrate them" — Jamie Betts. Dan Hartman: "We spent a lot of time going back afterwards and correcting delay categories."
Cost Analysis¶
| Component | Vooban Team | IE | Notes |
|---|---|---|---|
| Discovery & design | Solution architect, PM | [TBD] | Map delay categories → Tabware hierarchy |
| Data engineering | Data engineer | [TBD] | SQL delay DB + Tabware API integration |
| ML/AI development | ML engineer | [TBD] | Semantic matching layer (NLP) |
| Application/UX | Frontend dev | [TBD] | Unified dashboard |
| Infrastructure | Minimal | [TBD] | On-prem, existing data sources |
| Change management | — | [TBD] | Ops + maint alignment workshops. 20% of technical (union). |
CLV-P02: Predictive Maintenance Platform¶
CLV-12: PdM Platform (Multi-Asset)¶
Card Type: A — Anchored Corporate Project: PRJ-03
Value Analysis¶
Value Types: Cost avoidance + Throughput gain Value Formula:
(unplanned_failure_cost × failures_per_year × prevention_rate)
+ (rate_delay_minutes_per_failure × failures_per_year × prevention_rate × production_value_per_minute)
| Variable | Value | Source | Status |
|---|---|---|---|
| unplanned_failure_cost (bag house fan) | [TBD] | Repair cost + emergency labor | needs-corporate |
| failures_per_year (bag house) | [TBD] | Maintenance records | needs-corporate |
| rate_delay_per_failure (bag house) | 5-6 min/heat at 2 fans | Brian Thompson: "at 2 fans = rate delayed" | workshop-confirmed |
| heats_per_day | 28 | Stubna: 1SP target | workshop-confirmed |
| prevention_rate | 30-50% | Conservative H1 capture rate | estimated |
| production_value_per_minute | [TBD] | 1SP throughput × margin | needs-corporate |
Workshop-Sourced Range: $3-12M/yr (PoV through full expansion) Confidence: Medium — bag house and scrubbing data-rich, scaling is the question Key Quote: "It just works until it doesn't. And then, we all freak out." — Brian Thompson
Cost Analysis¶
| Component | Vooban Team | IE | Notes |
|---|---|---|---|
| Discovery & design | Solution architect, data scientist, PM | [TBD] | Asset selection, data audit |
| Data engineering | Data engineer | [TBD] | Pi historian + AssetWatch + cloud vibration consolidation |
| ML/AI development | ML engineer, data scientist | [TBD] | Anomaly detection models per asset |
| Application/UX | Frontend dev | [TBD] | Asset health dashboard, alert management |
| Infrastructure | Moderate | [TBD] | ML inference, data pipeline, possible edge compute |
| Change management | — | [TBD] | Operator/maint trust building. 20% of technical. |
Note: PdM PoV is a separate SOW (8 weeks, Cleveland, Chad-greenlit Mar 16). This card covers the full platform economics.
CLV-22: BOF Bag House Predictive Monitoring¶
Card Type: A — Anchored (PRIMARY PdM PoV target) Corporate Project: PRJ-03
Value Analysis¶
Value Types: Cost avoidance + Throughput gain + Risk mitigation (environmental) Value Formula:
(fan_failure_events_per_year × avg_downtime_hours × production_value_per_hour)
+ (rate_delay_heats_avoided × margin_per_heat)
+ (environmental_fine_risk_reduction)
| Variable | Value | Source | Status |
|---|---|---|---|
| fan_failure_events_per_year | [TBD] | Maintenance records | needs-corporate |
| avg_downtime_hours_per_failure | [TBD] | Historical repair times | needs-corporate |
| production_value_per_hour | [TBD] | 1SP throughput × margin | needs-corporate |
| rate_delay_heats_avoided | ~2-4 heats/day when at 2 fans | Workshop: lose 5-6 min/heat × 28 heats | workshop-confirmed |
| margin_per_heat | [TBD] | ~300 tons × margin/ton | needs-corporate |
| environmental_fine_risk | [TBD] | Compliance history + penalty structure | needs-corporate |
Workshop-Sourced Range: $2-5M/yr Confidence: Medium-High — data-rich, clear degradation patterns, strong champions
Cost Analysis¶
| Component | Vooban Team | IE | Notes |
|---|---|---|---|
| Discovery & design | Data scientist, PM | [TBD] | Included in PdM PoV scope |
| Data engineering | Data engineer | [TBD] | Pi historian extraction |
| ML/AI development | ML engineer | [TBD] | Fan degradation model, module cleaning optimization |
| Application/UX | Frontend dev | [TBD] | Real-time health dashboard |
| Infrastructure | Minimal | [TBD] | Pi historian already connected |
| Change management | — | [TBD] | Low — remote area, less operator disruption |
CLV-23: BOF Scrubbing System Predictive Monitoring¶
Card Type: A — Anchored (SECONDARY PdM PoV target) Corporate Project: PRJ-03
Value Analysis¶
Value Types: Throughput gain + Cost avoidance Value Formula:
vessel_unavailability_events_per_year × heats_lost_per_event × margin_per_heat × prevention_rate
+ emergency_repair_cost_per_event × events_per_year × prevention_rate
| Variable | Value | Source | Status |
|---|---|---|---|
| vessel_unavailability_events_per_year | [TBD] | Historical — when scrubbing forced vessel offline | needs-corporate |
| heats_lost_per_event | ~5 heats/day per furnace lost | John Messi: "a vessel is a goal for us" | workshop-confirmed |
| margin_per_heat | [TBD] | ~300 tons × margin/ton | needs-corporate |
| prevention_rate | 30-50% | Conservative H1 | estimated |
| emergency_repair_cost_per_event | [TBD] | Labor + materials for unplanned scrubbing intervention | needs-corporate |
| degradation_cycle_length | ~3 weeks | John Messi: predictable repair turn cycle | workshop-confirmed |
Workshop-Sourced Range: $2-5M/yr Confidence: Medium — predictable degradation ideal for modeling Key Quote: "Generates a shitload of data. Gets a bit load of attention." — John Messi
Cost Analysis¶
Included in CLV-12 PdM Platform scope. Marginal cost to add scrubbing model once bag house pipeline exists.
CLV-09: Spare Parts Inventory Intelligence¶
Card Type: B — Structured Corporate Project: PRJ-03
Value Analysis¶
Value Types: Inventory optimization + Cost avoidance Value Formula:
(inventory_value × carrying_cost_% × reduction_%)
+ (downtime_from_stockouts_hours × production_value_per_hour)
+ (failure_from_substitution_events × cost_per_event)
| Variable | Value | Source | Status |
|---|---|---|---|
| inventory_value (Cleveland) | [TBD] | Axiom ERP. MDT confirmed $104M. | needs-corporate |
| carrying_cost_% | 25% | Industry standard | estimated |
| reduction_% | 10-15% | Conservative (MDT has 10% duplicates) | estimated |
| downtime_from_stockouts_hours | [TBD] | Delay reports filtered by parts-related root cause | needs-corporate |
| production_value_per_hour | [TBD] | 1SP throughput × margin | needs-corporate |
| failure_from_substitution_events | [TBD] | Dan Hartman: 4340 vs stainless example | needs-corporate |
Workshop-Sourced Range: $1-4M/yr Confidence: Low-Medium
Cost Analysis¶
| Component | Vooban Team | IE | Notes |
|---|---|---|---|
| Discovery & design | Solution architect, PM | [TBD] | Inventory audit scope |
| Data engineering | Data engineer | [TBD] | Axiom + Tabware cross-reference |
| ML/AI development | Data scientist | [TBD] | Duplicate detection, min/max optimization |
| Application/UX | Frontend dev | [TBD] | Inventory dashboard, alert system |
| Infrastructure | Minimal | [TBD] | — |
| Change management | — | [TBD] | High — procurement policy change. 25%. |
CLV-P03: Maintenance Workflow & Copilot¶
CLV-07: Maintenance Co-Pilot (Technician Assist)¶
Card Type: A — Anchored Corporate Project: PRJ-06
Value Analysis¶
Value Types: Efficiency gain + Risk mitigation (knowledge capture) Value Formula:
(diagnosis_time_saved_per_repair × repairs_per_month × labor_cost_per_hour)
+ (repeat_failure_reduction_% × repeat_failure_annual_cost)
+ (documentation_compliance_improvement × regulatory_risk_value)
| Variable | Value | Source | Status |
|---|---|---|---|
| diagnosis_time_saved_per_repair | [TBD] | Dan: "half the time looking for the answer" — estimate 30-60 min/repair | needs-corporate |
| repairs_per_month | [TBD] | Tabware work order volume | needs-corporate |
| labor_cost_per_hour | [TBD] | Blended maintenance tech rate | needs-corporate |
| repeat_failure_reduction_% | 15-25% | From closed-loop documentation | estimated |
| repeat_failure_annual_cost | [TBD] | Cross-ref with CLV-01 repeat failure data | needs-corporate |
| documentation_compliance_improvement | [TBD] | Current compliance rate vs target | needs-corporate |
Workshop-Sourced Range: $0.5-2M/yr direct + data quality uplift (enabler for PdM and analytics) Confidence: Medium-High — explicitly requested by 5+ stakeholders Key Quotes: "If you had something where they can talk to their phone..." — Paul. "Where's my Ask Jeeves?" — Dan. "Design a solution that circumvents the need to fill tedious forms in Tabware" — Stubna.
Cost Analysis¶
| Component | Vooban Team | IE | Notes |
|---|---|---|---|
| Discovery & design | Solution architect, UX researcher, PM | [TBD] | Field shadowing critical — 1990 LISP lesson |
| Data engineering | Data engineer | [TBD] | Tabware API, knowledge base ingestion |
| ML/AI development | ML engineer | [TBD] | Voice-to-structured (LLM/STT), RAG over repair history |
| Application/UX | Frontend dev, mobile dev | [TBD] | Voice-first mobile app, offline-capable |
| Infrastructure | Moderate | [TBD] | STT/LLM inference, offline sync, connectivity gaps |
| Change management | — | [TBD] | Critical — trust + UX make or break. 25%. Union. |
CLV-08: Procurement Decision Tree / Auto-Approval¶
Card Type: A — Anchored Corporate Project: PRJ-06
Value Analysis¶
Value Types: Efficiency gain + Throughput gain (de-bottleneck) Value Formula:
(approval_cycle_reduction_days × orders_per_month × labor_cost_per_day)
+ (downtime_from_parts_delay_hours × production_value_per_hour × reduction_%)
+ (vendor_followup_calls × time_per_call × labor_rate × 12)
| Variable | Value | Source | Status |
|---|---|---|---|
| approval_cycle_reduction_days | ~38 days (from ~40 to ~2) | Paul: "two months for a $300 power supply" | workshop-confirmed |
| orders_per_month | [TBD] | Axiom purchasing data | needs-corporate |
| labor_cost_per_day | [TBD] | Admin/purchasing loaded rate | needs-corporate |
| downtime_from_parts_delay_hours | [TBD] | Delay reports × parts root cause | needs-corporate |
| production_value_per_hour | [TBD] | 1SP throughput × margin | needs-corporate |
| reduction_% | 50-70% | Auto-approve for known patterns | estimated |
| vendor_followup_calls | ~250/month | Dan: "Brandon manually calling ~250 vendors" | workshop-confirmed |
| time_per_call | 10-15 min | Estimated | estimated |
| labor_rate | [TBD] | Purchasing agent loaded rate | needs-corporate |
Workshop-Sourced Range: $1-3M/yr direct + de-bottleneck value Confidence: Medium-High — deepest pain point across every conversation Key Quotes: "I could only order nine gallons to keep it under 500" — Paul. Purchasing "unchecks the box" after 60 days — Dan.
Cost Analysis¶
| Component | Vooban Team | IE | Notes |
|---|---|---|---|
| Discovery & design | Solution architect, PM | [TBD] | Approval pattern analysis |
| Data engineering | Data engineer | [TBD] | Axiom integration |
| ML/AI development | Data scientist | [TBD] | Rules engine (Phase 1), fraud detection (Phase 2) |
| Application/UX | Frontend dev | [TBD] | Approval dashboard, status visibility |
| Infrastructure | Minimal | [TBD] | Rules engine, Axiom connector |
| Change management | — | [TBD] | HIGH — policy change, corporate buy-in needed. 25%. |
CLV-24: Caster Segment Lifecycle Tracking¶
Card Type: B — Structured Corporate Project: PRJ-06
Value Analysis¶
Value Types: Cost avoidance + Efficiency gain Value Formula:
emergency_segment_changes_per_year × (cost_per_emergency_change - cost_per_planned_change)
+ production_loss_hours_from_segment_failures × production_value_per_hour
| Variable | Value | Source | Status |
|---|---|---|---|
| emergency_segment_changes_per_year | [TBD] | Evan's records (currently Excel) | needs-corporate |
| cost_per_emergency_change | [TBD] | Emprotech invoices + labor | needs-corporate |
| cost_per_planned_change | [TBD] | Emprotech invoices + labor | needs-corporate |
| production_loss_hours_from_segment_failures | [TBD] | Caster downtime records | needs-corporate |
| production_value_per_hour | [TBD] | 1SP throughput × margin | needs-corporate |
Workshop-Sourced Range: $1-3M/yr Confidence: Medium Key Detail: Currently one contractor (Chris Callahan) manages on Excel with color coding. Springs breaking with no campaign life data.
Cost Analysis¶
| Component | Vooban Team | IE | Notes |
|---|---|---|---|
| Discovery & design | Solution architect | [TBD] | Digitize Evan's process |
| Data engineering | Data engineer | [TBD] | Excel → structured DB, Tabware linkage |
| ML/AI development | Minimal | [TBD] | Campaign life prediction (later phase) |
| Application/UX | Frontend dev | [TBD] | Segment tracking dashboard |
| Infrastructure | Minimal | [TBD] | — |
| Change management | — | [TBD] | Low — Evan already wants this. 15%. |
CLV-25: Critical Spares Identification & Digitization¶
Card Type: B — Structured Corporate Project: PRJ-06
Value Analysis¶
Value Types: Cost avoidance Value Formula:
stockout_events_per_year × avg_production_loss_per_stockout_hours × production_value_per_hour
| Variable | Value | Source | Status |
|---|---|---|---|
| critical_spares_count | ~500 | Brian Thompson: effort started Jan 2025 | workshop-confirmed |
| stockout_events_per_year | [TBD] | Maintenance records — unplanned downtime due to parts | needs-corporate |
| avg_production_loss_per_stockout_hours | [TBD] | Historical delay duration for parts-related events | needs-corporate |
| production_value_per_hour | [TBD] | 1SP throughput × margin | needs-corporate |
Workshop-Sourced Range: $1-3M/yr Confidence: Medium-High — effort already in progress, AI accelerates Key Detail: Richie has crane contactors on a department shelf — nobody else knows. No tracking after parts arrive at door 40.
Cost Analysis¶
| Component | Vooban Team | IE | Notes |
|---|---|---|---|
| Discovery & design | Data scientist, PM | [TBD] | Cross-ref Tabware hierarchy × failure history |
| Data engineering | Data engineer | [TBD] | Item master population, Tabware + Axiom |
| ML/AI development | Data scientist | [TBD] | Criticality scoring, gap identification |
| Application/UX | Frontend dev | [TBD] | Spares visibility dashboard |
| Infrastructure | Minimal | [TBD] | — |
| Change management | — | [TBD] | Low — builds on existing effort. 15%. |
CLV-P04: Process Risk & Cobble Prediction¶
CLV-04: Cobble Prediction & Prevention¶
Card Type: A — Anchored Corporate Project: PRJ-05
Value Analysis¶
Value Types: Cost avoidance + Throughput gain Value Formula:
cobbles_per_year × (equipment_damage_cost + downtime_hours × production_value_per_hour + scrap_tons × margin_per_ton) × prevention_rate
| Variable | Value | Source | Status |
|---|---|---|---|
| cobbles_per_year | [TBD] | HSM operational records | needs-corporate |
| equipment_damage_cost_per_cobble | [TBD] | Drive spindle replacement etc. | needs-corporate |
| downtime_hours_per_cobble | [TBD] | Historical delay data | needs-corporate |
| production_value_per_hour | [TBD] | HSM throughput × margin | needs-corporate |
| scrap_tons_per_cobble | [TBD] | Quality records | needs-corporate |
| margin_per_ton | [TBD] | Product mix average | needs-corporate |
| prevention_rate | 20% | Leadership target: "-20% cobble rate" | workshop-confirmed |
Workshop-Sourced Range: $3-10M/yr Confidence: Medium — need cobble frequency and cost-per-cobble Key Quote: "Machine learning will say, I want to ignore that head end... if I run this same product, and it alarmed once last time, and now this time it alarmed twice — there's something I want to learn from." — Dan Hartman
Cost Analysis¶
| Component | Vooban Team | IE | Notes |
|---|---|---|---|
| Discovery & design | Solution architect, data scientist, PM | [TBD] | L2 data audit, cobble event correlation |
| Data engineering | Data engineer | [TBD] | L2 + Pi historian + delay reports integration |
| ML/AI development | ML engineer, data scientist | [TBD] | Multi-variate risk model (temp, chemistry, mill setup) |
| Application/UX | Frontend dev | [TBD] | Operator risk score display, integrated with HMI |
| Infrastructure | Moderate | [TBD] | Real-time inference, L2 integration |
| Change management | — | [TBD] | Moderate — operator trust critical. 20%. |
CLV-11: Operator Decision Support (BF/HSM)¶
Card Type: B — Structured Corporate Project: PRJ-05
Value Analysis¶
Value Types: Cost avoidance + Efficiency gain + Risk mitigation Value Formula:
operator_error_events_per_year × avg_cost_per_event × reduction_%
+ training_time_saved_per_new_operator × new_operators_per_year × labor_rate
| Variable | Value | Source | Status |
|---|---|---|---|
| operator_error_events_per_year | [TBD] | Delay/incident reports tagged to operator error | needs-corporate |
| avg_cost_per_event | [TBD] | Equipment damage + downtime + scrap | needs-corporate |
| reduction_% | 15-30% | Context-aware guidance reduces error rate | estimated |
| training_time_saved_per_new_operator | [TBD] | Current onboarding months vs target | needs-corporate |
| new_operators_per_year | [TBD] | HR hiring/turnover data | needs-corporate |
| labor_rate | [TBD] | Loaded operator rate | needs-corporate |
Workshop-Sourced Range: $1-5M/yr Confidence: Medium Key Quote: "Give me those answers and I still might have a decision matrix... but at least I can focus on the critical part of that decision rather than finding and tracking all those variables." — Dan Hartman
Cost Analysis¶
| Component | Vooban Team | IE | Notes |
|---|---|---|---|
| Discovery & design | Solution architect, domain specialist, PM | [TBD] | Expert knowledge elicitation |
| Data engineering | Data engineer | [TBD] | L2 + Pi historian + alarm history |
| ML/AI development | ML engineer, data scientist | [TBD] | Context-aware alarming, decision models |
| Application/UX | Frontend dev | [TBD] | Operator-facing guidance display |
| Infrastructure | Moderate | [TBD] | Real-time inference at HMI level |
| Change management | — | [TBD] | High — safety critical, operator override policy. 25%. |
CLV-17: BOF Endpoint Prediction¶
Card Type: C — Absorbed Corporate Project: PRJ-05 Reason: Seed initiative — not field-validated at Cleveland. No stakeholder-confirmed metrics. Value Contribution: Absorbed into CLV-P04 roll-up. BOF endpoint prediction is one component of broader process risk reduction. $0.5-2M/yr estimated from industry benchmarks. Cost Contribution: One ML model within CLV-P04 scope.
CLV-18: Caster Breakout Prediction (ML)¶
Card Type: C — Absorbed Corporate Project: PRJ-05 Reason: Seed — not discussed with Cleveland stakeholders. Value Contribution: Absorbed into CLV-P04. $1-3M/yr industry estimate. Safety-critical application. Cost Contribution: ML model development within CLV-P04 scope.
CLV-20: BF Thermal State Prediction¶
Card Type: C — Absorbed Corporate Project: PRJ-05 Reason: Seed — BF C5/C6 have Emerson instrumentation but no field champion validated. Value Contribution: Absorbed into CLV-P04. $1-3M/yr estimated. Cost Contribution: ML model within CLV-P04 scope. Requires Emerson data access.
CLV-P05: Quality & Yield Traceability¶
CLV-06: 90-Day Slab Remelting Reduction¶
Card Type: B — Structured Corporate Project: PRJ-04
Value Analysis¶
Value Types: Cost avoidance + Throughput gain Value Formula:
slabs_remelted_per_month × tons_per_slab × (margin_lost + reprocessing_cost_per_ton)
× reduction_%
| Variable | Value | Source | Status |
|---|---|---|---|
| slabs_remelted_per_month | [TBD] | Janus WMS data | needs-corporate |
| tons_per_slab | ~25-30 tons | Standard slab weight | estimated |
| margin_lost_per_ton | [TBD] | Product margin vs scrap value | needs-corporate |
| reprocessing_cost_per_ton | [TBD] | Internal scrap processing + BOF energy | needs-corporate |
| reduction_% | 50% | Achievable with aging alerts + reallocation | estimated |
Workshop-Sourced Range: TBD — need remelting volume Confidence: Medium Key Detail: "We don't have enough pain in terms of recycling it" — easy remelting masks the real cost.
Cost Analysis¶
| Component | Vooban Team | IE | Notes |
|---|---|---|---|
| Discovery & design | Solution architect | [TBD] | Janus data audit |
| Data engineering | Data engineer | [TBD] | Janus + Axiom integration |
| ML/AI development | Minimal | [TBD] | Rules-based aging alerts, AI reallocation matching |
| Application/UX | Frontend dev | [TBD] | Slab aging dashboard |
| Infrastructure | Minimal | [TBD] | — |
| Change management | — | [TBD] | Low-moderate. 15%. |
CLV-14: Through-Process Traceability (Heat→Coil)¶
Card Type: C — Absorbed (Enabler) Corporate Project: PRJ-04 Reason: Foundation/enabler — no direct standalone value. Unlocks cross-process root cause analysis. Value Contribution: Enables $13-27M/yr in yield improvement across CLV-P04 and CLV-P05 scope. Value captured in parent projects. Cost Contribution: Significant data engineering effort — L2 integration across all process steps.
CLV-19: Surface Defect Detection (CNN on SIS)¶
Card Type: C — Absorbed Corporate Project: PRJ-04 Reason: Seed — not discussed with Cleveland stakeholders. Palmer named surface inspection but evidence is stronger at Middletown (Ametek 60% accuracy, 4+ lines). Value Contribution: $2-5M/yr industry estimate. Cleveland may have Global Gauges (2 HSM stands) as starting point. Cost Contribution: Classifier retraining — bounded ML. Entry at Middletown, scale to Cleveland.
CLV-P06: Production Scheduling & S&IOP¶
CLV-02: Cross-Stage Scheduling / S&IOP¶
Card Type: A — Anchored Corporate Project: PRJ-02
Value Analysis¶
Value Types: Throughput gain + Efficiency gain Value Formula:
(additional_heats_per_day_from_optimization × margin_per_heat × 365)
+ (changeover_reduction_% × changeover_hours_per_year × production_value_per_hour)
+ (missed_shipdate_reduction_% × penalty_cost_per_missed_shipment × shipments_per_year)
| Variable | Value | Source | Status |
|---|---|---|---|
| additional_heats_per_day | [TBD] | Andrew: "1% improvement = tens of millions" | needs-corporate |
| margin_per_heat | [TBD] | ~300 tons × margin/ton | needs-corporate |
| changeover_reduction_% | 30-50% | Target from CLV-03 | estimated |
| changeover_hours_per_year | [TBD] | HSM delay data — roll change bucket | needs-corporate |
| production_value_per_hour | [TBD] | 1SP throughput × margin | needs-corporate |
| missed_shipdate_events_per_year | [TBD] | Commercial/logistics | needs-corporate |
| penalty_cost_per_missed_shipment | [TBD] | Customer contracts | needs-corporate |
Workshop-Sourced Range: $10-30M/yr (Andrew: "1% improvement = tens of millions") Confidence: Low-Medium — massive scope, validated as real pain by Stubna Key Quote: "The commercial side has a certain set of priorities... none of the constraints talk to each other."
Cost Analysis¶
| Component | Vooban Team | IE | Notes |
|---|---|---|---|
| Discovery & design | Solution architect, domain specialist, PM | [TBD] | Cross-functional process mapping |
| Data engineering | Data engineer (senior) | [TBD] | Axiom + L2 + Tabware + scheduling integration |
| ML/AI development | ML engineer, optimization specialist | [TBD] | Constraint optimization, demand forecasting |
| Application/UX | Frontend dev | [TBD] | Planning dashboard, re-scheduling interface |
| Infrastructure | Significant | [TBD] | Optimization engine, real-time data feeds |
| Change management | — | [TBD] | Very high — cross-functional, cultural. 25%+. |
CLV-03: Roll Change Sequencing Optimization¶
Card Type: C — Absorbed Corporate Project: PRJ-02 Reason: Subsumed into CLV-02 scope. Dan Hartman confirmed roll change optimization is part of broader scheduling story. Value Contribution: $5-15M/yr estimated — captured in CLV-02 changeover reduction component. Cost Contribution: One optimization model within CLV-P06 scope.
CLV-15: Dynamic Pricing by Capacity Consumption¶
Card Type: C — Absorbed Corporate Project: PRJ-02 Reason: Strategic H3 initiative. Touches commercial policy. Value is real ($5-20M/yr) but requires scheduling foundation (CLV-02) first. Value Contribution: Captured in CLV-P06 total. Independent sizing deferred to H3. Cost Contribution: Requires capacity-adjusted profitability model — separate development phase.
CLV-26: Raw Materials Cost Optimization¶
Card Type: C — Absorbed Corporate Project: PRJ-02 Reason: Early-stage, identified by John Messi. Needs raw materials spend data to scope. Value Contribution: TBD — "that's where all the money is... we're a raw materials company." Largest cost center but no formula possible yet. Cost Contribution: Burden/alloy optimization model — cross-functional scope.
CLV-P07: Intra-Plant Logistics¶
CLV-10: Intra-Plant Coil Movement Optimization¶
Card Type: B — Structured Corporate Project: PRJ-07
Value Analysis¶
Value Types: Efficiency gain + Throughput gain Value Formula:
(current_moves_per_coil - target_moves_per_coil) × coils_per_month × crane_cost_per_move
+ crane_hours_freed × alternative_use_value_per_hour
| Variable | Value | Source | Status |
|---|---|---|---|
| current_moves_per_coil | 3-4 | Workshop: "moving coils way too much... triple time" | workshop-confirmed |
| target_moves_per_coil | 1-2 | Target | estimated |
| coils_per_month | [TBD] | Shipping records | needs-corporate |
| crane_cost_per_move | [TBD] | Crane operating cost + labor | needs-corporate |
| crane_hours_freed | [TBD] | Derived from move reduction | needs-corporate |
Workshop-Sourced Range: TBD — need movement data Confidence: Low Key Quote: "We're moving coils from 116 door to 45 door, then a week later to 30... triple time."
Cost Analysis¶
| Component | Vooban Team | IE | Notes |
|---|---|---|---|
| Discovery & design | Solution architect, PM | [TBD] | Movement data analysis |
| Data engineering | Data engineer | [TBD] | SQL coil tracking DB |
| ML/AI development | Data scientist | [TBD] | Storage allocation optimization |
| Application/UX | Frontend dev | [TBD] | Warehouse-style placement dashboard |
| Infrastructure | Minimal | [TBD] | — |
| Change management | — | [TBD] | Moderate — crane operator workflow change. 20%. |
CLV-16: Rail Car Inventory Visibility¶
Card Type: C — Absorbed Corporate Project: PRJ-07 Reason: Depends on external railroad cooperation (NS, CSX). Low confidence. No champion beyond shipping team. Value Contribution: Part of broader logistics optimization. Value captured in CLV-P07 roll-up. Cost Contribution: API integration with railroad systems (if accessible).
CLV-P08: Caster Chemistry Optimization¶
CLV-05: Caster Chemistry Transition Optimization¶
Card Type: B — Structured Corporate Project: PRJ-08
Value Analysis¶
Value Types: Cost avoidance (scrap reduction) Value Formula:
chemistry_transitions_per_day × scrap_tons_per_transition × margin_per_ton × reduction_%
+ in_between_grade_recovery_tons_per_year × margin_per_ton
| Variable | Value | Source | Status |
|---|---|---|---|
| chemistry_transitions_per_day | [TBD] | Caster scheduling data | needs-corporate |
| scrap_tons_per_transition | [TBD] | Quality records — transition zone yield loss | needs-corporate |
| margin_per_ton | [TBD] | Product mix average | needs-corporate |
| reduction_% | 30-50% | Optimized cut points | estimated |
| in_between_grade_recovery_tons | [TBD] | Currently scrapped, could be allocated to secondary grades | needs-corporate |
Workshop-Sourced Range: $2-8M/yr Confidence: Medium — Prime Metals model failure is a risk signal Key Quote: "We spent millions on [Prime Metals]... It's supposed to give us a cut point... We don't even use it."
Cost Analysis¶
| Component | Vooban Team | IE | Notes |
|---|---|---|---|
| Discovery & design | Solution architect, domain specialist | [TBD] | Prime Metals forensics first — why did it fail? |
| Data engineering | Data engineer | [TBD] | L2 caster + Axiom orders integration |
| ML/AI development | ML engineer | [TBD] | Cut point optimization, sequence optimizer |
| Application/UX | Frontend dev | [TBD] | Caster operator guidance |
| Infrastructure | Moderate | [TBD] | Real-time inference at caster |
| Change management | — | [TBD] | Moderate — Prime Metals failure created skepticism. 20%. |
CLV-P09: Utility Management¶
CLV-21: Utility Management System¶
Card Type: A — Anchored Corporate Project: New (not in PRJ-01..PRJ-11)
Value Analysis¶
Value Types: Direct cost reduction + Efficiency gain Value Formula:
compressed_air_leak_savings
+ water_consumption_reduction_% × water_annual_cost
+ nitrogen_optimization_value
+ manual_charting_hours_saved × labor_rate × 12
| Variable | Value | Source | Status |
|---|---|---|---|
| compressed_air_savings | $7M/yr | Engineering study — 3-figure air leaks found | workshop-confirmed |
| water_consumption_reduction_% | ~19% | Paul's projection from metering deployment | workshop-confirmed |
| water_annual_cost | [TBD] | Utility billing | needs-corporate |
| nitrogen_optimization_value | [TBD] | Pressure drop impact on 1SP | needs-corporate |
| manual_charting_hours_saved | [TBD] | Dan: descale system = 18 variables weekly | needs-corporate |
| labor_rate | [TBD] | Maintenance tech rate | needs-corporate |
Workshop-Sourced Range: $7-15M/yr ($7M compressed air alone, rest additive) Confidence: Medium — compressed air backed by engineering study, others need scoping Key Detail: Paul Aaron Dash has been building toward this for 20+ years. He is the last utility engineer.
Cost Analysis¶
| Component | Vooban Team | IE | Notes |
|---|---|---|---|
| Discovery & design | Solution architect, PM | [TBD] | Metering infrastructure audit |
| Data engineering | Data engineer | [TBD] | Wonderware + Emerson meters + manual data |
| ML/AI development | Data scientist | [TBD] | Anomaly detection, load optimization |
| Application/UX | Frontend dev | [TBD] | Utility monitoring dashboard |
| Infrastructure | Moderate | [TBD] | Metering gaps may need hardware |
| Change management | — | [TBD] | Low — Paul is the champion and user. 15%. |
CLV-P10: Environmental Compliance¶
CLV-13: Environmental Compliance Automation¶
Card Type: B — Structured Corporate Project: New
Value Analysis¶
Value Types: Efficiency gain + Risk mitigation Value Formula:
manual_tracking_hours_per_year × labor_rate
+ compliance_gap_fine_risk × probability_reduction
| Variable | Value | Source | Status |
|---|---|---|---|
| manual_tracking_hours_per_year | [TBD] | SDS tracking, refrigerant reporting, EPA filings | needs-corporate |
| labor_rate | [TBD] | Environmental/compliance staff rate | needs-corporate |
| compliance_gap_fine_risk | [TBD] | Historical fines + penalty schedule | needs-corporate |
| probability_reduction | 50-80% | Automated tracking eliminates human error gaps | estimated |
Workshop-Sourced Range: $0.5-1M/yr Confidence: Low — need to scope Key Detail: Andrew Mullen mentioned it affects bonuses. Burns Harbor has exceedances.
Cost Analysis¶
| Component | Vooban Team | IE | Notes |
|---|---|---|---|
| Discovery & design | Solution architect | [TBD] | Compliance requirement mapping |
| Data engineering | Data engineer | [TBD] | Purchase records → SDS linkage |
| ML/AI development | Minimal | [TBD] | Rules-based automation |
| Application/UX | Frontend dev | [TBD] | Compliance dashboard + reporting |
| Infrastructure | Minimal | [TBD] | — |
| Change management | — | [TBD] | Low. 15%. |
CLV-P11: Enterprise Data Platform Strategy¶
CLV-27: Enterprise Data Platform (Cloud Strategy)¶
Card Type: C — Absorbed (Enabler) Corporate Project: Enabler (not a standalone project) Reason: Enabler — the progressive data foundation emerges from doing the work, not as a separate initiative. CEO wants it, Databricks/Snowflake/Fabric under evaluation, but we recommend bottom-up. Value Contribution: Unlocks $50-100M+ optimization across all projects (H3). No standalone financial card — value is captured in the projects it enables. Cost Contribution: Platform licensing + architecture decisions made as H1/H2 projects mature. Not costed separately.
Project Roll-Ups¶
| Project | Initiatives | Anchored (A) | Structured (B) | Absorbed (C) | Workshop Range | Confidence |
|---|---|---|---|---|---|---|
| CLV-P01 Ops-Maint Integration | CLV-01 | 1 | 0 | 0 | $2-5M/yr | Med-High |
| CLV-P02 PdM Platform | CLV-12, CLV-22, CLV-23, CLV-09 | 3 | 1 | 0 | $3-12M/yr | Medium |
| CLV-P03 Maint Workflow & Copilot | CLV-07, CLV-08, CLV-24, CLV-25 | 2 | 2 | 0 | $3-11M/yr | Med-High |
| CLV-P04 Process Risk & Cobble | CLV-04, CLV-11, CLV-17, CLV-18, CLV-20 | 1 | 1 | 3 | $6-23M/yr | Medium |
| CLV-P05 Quality & Yield | CLV-06, CLV-14, CLV-19 | 0 | 1 | 2 | $2-5M + enabler | Low-Med |
| CLV-P06 Scheduling & S&IOP | CLV-02, CLV-03, CLV-15, CLV-26 | 1 | 0 | 3 | $20-65M/yr | Low-Med |
| CLV-P07 Logistics | CLV-10, CLV-16 | 0 | 1 | 1 | TBD | Low |
| CLV-P08 Caster Chemistry | CLV-05 | 0 | 1 | 0 | $2-8M/yr | Medium |
| CLV-P09 Utility Management | CLV-21 | 1 | 0 | 0 | $7-15M/yr | Medium |
| CLV-P10 Environmental | CLV-13 | 0 | 1 | 0 | $0.5-1M/yr | Low |
| CLV-P11 Data Platform | CLV-27 | 0 | 0 | 1 | Enabler | Low |
| TOTAL | 27 | 9 | 8 | 10 | $46-145M/yr |
Card type distribution: 9 Anchored (33%), 8 Structured (30%), 10 Absorbed (37%). The 17 cards with formulas (A+B) cover the bulk of the value — the 10 Absorbed initiatives contribute within parent projects.
Corporate Inquiry Table — Cleveland Works¶
Purpose: All variables tagged
needs-corporatein one table. Send to IE for Cleveland-Cliffs data request.
Production & Throughput¶
| # | Variable | Needed For | Question to Ask CLF | Priority |
|---|---|---|---|---|
| 1 | production_value_per_hour (1SP) | CLV-01, CLV-04, CLV-08, CLV-10, CLV-12, CLV-22, CLV-23, CLV-24, CLV-25 | What is the 1SP production value per hour? (tons/hour × margin/ton, or revenue per operating hour) | Critical — used across 9 cards |
| 2 | margin_per_heat | CLV-02, CLV-22, CLV-23 | What is the average margin per heat at 1SP? (~300 tons × $/ton) | Critical |
| 3 | margin_per_ton | CLV-04, CLV-05, CLV-06 | What is the average product margin per ton at Cleveland? (by product mix if possible) | Critical |
Maintenance & Reliability¶
| # | Variable | Needed For | Question to Ask CLF | Priority |
|---|---|---|---|---|
| 4 | misattributed_delay_hours_per_month | CLV-01 | How many delay hours per month are disputed between ops and maintenance? (Or: what % of total delay hours have no matching work order?) | High |
| 5 | repeat_failure_annual_cost | CLV-01, CLV-07 | What is the annual cost of repeat equipment failures? (frequency × avg cost per event, or total maintenance rework spend) | High |
| 6 | repairs_per_month | CLV-07 | How many maintenance work orders are created per month in Tabware? (all types) | High |
| 7 | fan_failure_events_per_year (bag house) | CLV-22 | How many unplanned bag house fan failures occurred in the last 12 months? Duration of each? | High |
| 8 | vessel_unavailability_events_per_year | CLV-23 | How many times was a BOF vessel taken offline due to scrubbing system issues in the last 12 months? | High |
| 9 | emergency_segment_changes_per_year | CLV-24 | How many unplanned/emergency caster segment changes per year? Cost of each (Emprotech + labor + production loss)? | Medium |
| 10 | stockout_events_per_year (critical spares) | CLV-25 | How many production delay events per year are attributable to parts unavailability? | Medium |
Procurement & Inventory¶
| # | Variable | Needed For | Question to Ask CLF | Priority |
|---|---|---|---|---|
| 11 | orders_per_month | CLV-08 | How many purchase orders/requisitions are created per month at Cleveland? (from Axiom) | High |
| 12 | inventory_value (Cleveland) | CLV-09 | What is the total spare parts inventory value at Cleveland? (Middletown confirmed $104M) | High |
| 13 | unique_part_numbers | CLV-09 | How many unique part numbers in Tabware/Axiom at Cleveland? | Medium |
| 14 | annual_spare_parts_spend | CLV-09 | What is the annual spare parts procurement spend at Cleveland? | Medium |
Labor & Cost Rates¶
| # | Variable | Needed For | Question to Ask CLF | Priority |
|---|---|---|---|---|
| 15 | labor_cost_per_hour (maintenance tech) | CLV-01, CLV-07, CLV-11 | What is the loaded hourly rate for a maintenance technician? (wages + benefits + overhead) | Critical — used across 3+ cards |
| 16 | labor_cost_per_hour (purchasing/admin) | CLV-08 | What is the loaded hourly rate for a purchasing agent/admin? | Medium |
| 17 | new_operators_per_year | CLV-11 | How many new operators are hired/transferred into roles per year at Cleveland? | Low |
Process & Quality¶
| # | Variable | Needed For | Question to Ask CLF | Priority |
|---|---|---|---|---|
| 18 | cobbles_per_year (HSM) | CLV-04 | How many cobble events occurred at the HSM in the last 12 months? Average downtime per cobble? | High |
| 19 | equipment_damage_cost_per_cobble | CLV-04 | What is the average repair cost per cobble? (drive spindles, work rolls, etc.) | High |
| 20 | chemistry_transitions_per_day | CLV-05 | How many chemistry transitions does the caster make per day on average? | Medium |
| 21 | scrap_tons_per_transition | CLV-05 | How many tons of scrap/downgrade are generated per caster transition on average? | Medium |
| 22 | slabs_remelted_per_month | CLV-06 | How many slabs (or tons) are remelted due to >90 day aging per month? | Medium |
Logistics & Shipping¶
| # | Variable | Needed For | Question to Ask CLF | Priority |
|---|---|---|---|---|
| 23 | coils_per_month | CLV-10 | How many coils are shipped per month from Cleveland? | Medium |
| 24 | crane_cost_per_move | CLV-10 | What is the estimated cost per crane move? (operating cost + labor time) | Low |
Utilities¶
| # | Variable | Needed For | Question to Ask CLF | Priority |
|---|---|---|---|---|
| 25 | water_annual_cost | CLV-21 | What is Cleveland's annual water/utility cost? | Medium |
| 26 | nitrogen_optimization_value | CLV-21 | What is the production impact when nitrogen pressure drops affect 1SP? (frequency × downtime) | Medium |
Compliance¶
| # | Variable | Needed For | Question to Ask CLF | Priority |
|---|---|---|---|---|
| 27 | compliance_gap_fine_risk | CLV-13 | What is Cleveland's environmental fine history? What are the penalty amounts for SDS/refrigerant non-compliance? | Low |
| 28 | manual_tracking_hours_per_year | CLV-13 | How many person-hours per year are spent on manual environmental compliance tracking? | Low |
Summary: 28 variables needed. 3 Critical (used across many cards), 10 High, 10 Medium, 5 Low priority.