# Cracking Unit Diagnostics

*/Opportunities/Cracking_Unit_Diagnostics*

## Opportunity Overview

**Wedge**: The initial beachhead targets mid-sized, independent refineries operating aging cracking units where historical sensor data is abundant but in-house analytical headcount is zero. Winning this niche proves the model accuracy on the most failure-prone equipment without facing the rigid procurement barriers of supermajors. Expansion proceeds by moving from diagnostics to automated control parameter optimization, and laterally into hydrocracking and coking units within the same facility.
**Timing**: Recent advancements in multimodal time-series foundation models allow for the ingestion of sparse, noisy SCADA data alongside unstructured maintenance logs without requiring site-specific, manually tuned physics models.
**Why This I C P**: Refinery Reliability Engineers bear the direct KPI for uptime and face severe penalties for unplanned outages costing upwards of $1M per day, making them highly motivated buyers with decentralized operational budgets.
**Size Of Prize**: There are roughly 700 active oil refineries globally containing approximately 1,500 total cracking units. Refineries currently spend roughly $250,000 annually per unit on third-party diagnostic consulting, specialized monitoring software, and manual inspection labor, yielding a total addressable market of $375M.
**Gap Narrative**: Refinery reliability engineers monitor fluid catalytic cracking units using thousands of raw sensor feeds but lack automated correlation to predict coking, catalyst degradation, or mechanical failure before a shutdown. Current solutions require manual data extraction and specialized metallurgical analysts who take weeks to diagnose anomalies. This opportunity delivers real-time, automated root-cause diagnostics that directly correlate multiphysics sensor data to specific unit degradation modes.
**Defensibility**: The system compounds value through a proprietary failure-mode dataset where every anomaly detected and confirmed by physical inspection trains the model to recognize edge-case degradation signatures across the global fleet. As the platform integrates deeply into the plant distributed control system for continuous monitoring, switching costs become prohibitively high due to the operational risk of replacing a proven early-warning system.
**Why This Thesis**: A Service-as-Software approach bypasses the need for the refinery to hire specialized data scientists or integrate complex dashboarding tools. The product directly delivers actionable diagnostic reports and maintenance schedules, replacing the existing workflow of hiring third-party process engineering consultants.

## Opportunity Linked Thesis

**Thesis**: [Software](/Theses/Software)

## Opportunity Linked I C P

**Icp**: [Petrochemical Refinery](/CompanyTypes/Petrochemical_Refinery)

## Opportunity Market Sizing

_Illustrative — target and order-of-magnitude estimate figures, not an achieved track record (this Thing is concept-stage)._

**S A M**: ~$150-200M (North American and European petrochemical complexes)
**S O M**: ~$10-25M
**T A M**: ~2,500 global petrochemical and refining facilities × ~$200k-250k/yr per site for advanced process diagnostics ≈ $500-625M
**Growth Rate**: ~8-12%/yr, driven by aging downstream infrastructure, skilled operator retirement, and the rising margin impact of unplanned unit outages
**Paid Comparable Spend**: ~$150k-300k/yr per site spent on third-party non-destructive testing contractors, manual turnaround inspections, and generic process historian software

## Opportunity Incumbents

- [AspenTech HYSYS](/Products/AspenTech_HYSYS) — Tool
- [Honeywell Forge](/Products/Honeywell_Forge) — Tool
- [KBC Petro-SIM](/Products/KBC_Petro-SIM) — Tool
- [Lummus Technology Consulting](/Products/Lummus_Technology_Consulting) — Service
- [In-House Excel Models](/Products/In-House_Excel_Models) — Spreadsheet
- [KBR Engineering Services](/Products/KBR_Engineering_Services) — Service
- [Emerson AMS Suite](/Products/Emerson_AMS_Suite) — Tool

## Opportunity Win Conditions

**Kill Thresholds**:
- Integration time > 14 days per cracking unit during pilot phase
- Weekly active users < 2 per facility after day 30
- False positive anomaly alert rate > 15%
- Paid conversion at $150k/yr < 25% for completed 60-day pilots
**Leading Metrics**:
- Time-to-first-prediction (hours from historian connection to first yield output)
- Weekly active users (specifically process engineers and unit operators)
- Alert acceptance rate (% of diagnostic warnings actioned in the UI)
- Data pipeline uptime (% of hours with uninterrupted sensor ingestion)
**What Proves Right**: Plant engineers connect real-time historian data and configure their cracking unit models within 48 hours without vendor services. Process engineers log in daily to review catalyst coking rates and yield predictions instead of relying on batch Excel reports. Facilities sign annual contracts at $150k per year after successfully predicting at least one impending temperature excursion during a 60-day pilot.
**What Proves Wrong**: Operators ignore the diagnostics dashboard, deferring to legacy Aspen models or manual contractor reports for operational decisions. Data integration from on-premise historian systems requires more than three weeks of custom engineering per site. Facilities abandon the pilot because the diagnostic model fails to accurately account for physical feed composition changes.

## Opportunity Build Profile

**Hardest Part**: Mapping inconsistent symptom-based field observations to exact component failures across legacy equipment models without hallucinating incorrect or dangerous repair steps.
**Min Viable Scope**: Build exclusively for commercial HVAC rooftop units from three major brands, taking a field symptom and outputting the top likely failed parts with testing sequences. Deliberately leave out live IoT telemetry ingestion, part ordering, and quoting.
**Cold Start Problem**: The system lacks historical ground-truth data linking symptoms to actual successful repairs. Break this by scraping public PDF manuals for a narrow equipment category and partnering with one service company to bulk-ingest their closed service tickets.
**Time To First Value**: Immediate upon first diagnostic query. The gating step is indexing the specific OEM equipment manual into the system.
**Data Moat Available**: true
**Technical Difficulty**: High

## Neighborhood

### Where the gap lives

- [Chemical refineries](/Customers/Chemical_refineries) — latent gap · Customers

### Incumbent in

- [Aspen HYSYS](/Products/Aspen_HYSYS) — incumbent in · Products
- [Emerson AMS](/Products/Emerson_AMS) — incumbent in · Products
- [Lummus Technology Consulting](/Products/Lummus_Technology_Consulting) — incumbent in · Products
- [KBC Petro-SIM](/Products/KBC_Petro-SIM) — incumbent in · Products
- [KBR Engineering Services](/Products/KBR_Engineering_Services) — incumbent in · Products
- [Honeywell Forge](/Products/Honeywell_Forge) — incumbent in · Products
- [In-House Excel Models](/Products/In-House_Excel_Models) — incumbent in · Products

### Applies thesis

- [Petrochemical Refinery](/CompanyTypes/Petrochemical_Refinery) — applies thesis · CompanyTypes

### Embodies

- [Software](/Theses/Software) — embodies · Theses

### Similar Opportunities

- [Alarm Flood Triage](/Opportunities/Alarm_Flood_Triage) — similar · Opportunities
- [Acoustic Burner Diagnostics](/Opportunities/Acoustic_Burner_Diagnostics) — similar · Opportunities
- [Refinery Dynamic Deadband Adjustment](/Opportunities/Refinery_Dynamic_Deadband_Adjustment) — similar · Opportunities
- [Pump Diagnostics Monitor](/Opportunities/Pump_Diagnostics_Monitor) — similar · Opportunities
- [Predictive Audit Scoring for Refineries](/Opportunities/Predictive_Audit_Scoring_for_Refineries) — similar · Opportunities
- [AI Predictive Maintenance](/Opportunities/AI_Predictive_Maintenance) — similar · Opportunities
- [Maintenance Diagnostics](/CompanyTypes/Industrial_Facility_Operator/Opportunities/Maintenance_Diagnostics) — similar · Opportunities
- [Catalyst Replacement Scheduling for Agrochemicals](/Opportunities/Catalyst_Replacement_Scheduling_for_Agrochemicals) — similar · Opportunities
- [Generative Shift Handovers for Refineries](/Opportunities/Generative_Shift_Handovers_for_Refineries) — similar · Opportunities
- [Predictive Equipment Diagnostics](/Opportunities/Predictive_Equipment_Diagnostics) — similar · Opportunities
- [Refiner Energy Optimizer](/Opportunities/Refiner_Energy_Optimizer) — similar · Opportunities
- [Machine Telemetry Engine](/Opportunities/Machine_Telemetry_Engine) — similar · Opportunities
- [Feedstock Blending Optimizer](/Opportunities/Feedstock_Blending_Optimizer) — similar · Opportunities
- [Crusher Diagnostics API](/Opportunities/Crusher_Diagnostics_API) — similar · Opportunities
- [Refiner Tuning Agent](/Opportunities/Refiner_Tuning_Agent) — similar · Opportunities
- [SCADA Alert Triage Automation](/Opportunities/SCADA_Alert_Triage_Automation) — similar · Opportunities
- [SCADA Fault Diagnostics](/Opportunities/SCADA_Fault_Diagnostics) — similar · Opportunities
- [AI Turnaround Planner](/Opportunities/AI_Turnaround_Planner) — similar · Opportunities
- [Algorithmic Refiner Energy Optimization](/Opportunities/Algorithmic_Refiner_Energy_Optimization) — similar · Opportunities
- [Distillation Yield Engine](/Opportunities/Distillation_Yield_Engine) — similar · Opportunities