# Distillation Yield Engine

*/Opportunities/Distillation_Yield_Engine*

## Opportunity Overview

**Wedge**: Target continuous ethanol and biofuel distillation plants first. Feedstocks fluctuate but the target molecule is highly standardized, offering immediate baseline comparisons and fast proof-of-value by measuring daily output volume against historical averages. Expansion moves outward to multi-component specialty chemical separations, and eventually to upstream reactor optimization.
**Timing**: Time-series transformers and physics-informed neural networks now execute on standard edge hardware using existing plant historian data. Previous iterations required custom-built, rigid thermodynamic models updated manually by chemical engineering consultants over several years.
**Why This I C P**: Mid-sized specialty chemical and biofuel plants lack the massive in-house process control teams of mega-refineries. They operate on tight margins and utilize standard column designs that allow for repeatable software deployment across different operator sites.
**Size Of Prize**: There are approximately 3,500 mid-sized chemical and biofuel distillation facilities in the US and Europe. At an average annual software subscription of $150,000 per facility representing a fraction of recovered product value, the total addressable prize is $525M.
**Gap Narrative**: Mid-sized chemical plants rely on static Advanced Process Control models that degrade as distillation column internals foul. Operators manually buffer temperature and reflux setpoints to avoid off-spec batches, sacrificing two to four percent of total potential yield. The market lacks a solution that continuously adapts thermodynamic models to real-time sensor data without requiring multi-month engineering integrations.
**Defensibility**: Defensibility compounds through workflow lock-in and plant-specific historical data accumulation. As the software ingests successive thermal dynamics and physical fouling cycles for a specific column, its predictions achieve extreme local accuracy, creating a severe switching cost for any competing model that must begin baseline data collection from zero.
**Why This Thesis**: A pure software deployment matches the plant requirement to keep legacy Distributed Control Systems intact while overlaying optimization. Ingesting standard historian data streams to deliver continuous setpoint recommendations replaces bespoke consulting engagements with a scalable, low-friction installation.

## Opportunity Linked Thesis

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

## Opportunity Linked I C P

**Icp**: [Commercial Distillery](/CompanyTypes/Commercial_Distillery)

## Opportunity Market Sizing

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

**S A M**: ~$150M-250M (focusing on ~10,000 mid-to-large scale commercial and craft distilleries in North America and Europe)
**S O M**: ~$10M-15M (targeting 500-750 early adopter distilleries over the next 3 years)
**T A M**: ~35,000 global commercial distilleries × ~$20,000/yr platform subscription ≈ ~$700M
**Growth Rate**: ~12-18%/yr, driven by rising raw material costs and increasing competition in the premium craft spirits category forcing margin optimization
**Paid Comparable Spend**: ~$15,000-40,000/yr on generic SCADA licenses, manual lab sample testing labor, and lost revenue from suboptimal spirit cuts

## Opportunity Incumbents

- [Aspen Plus Optimizer](/Products/Aspen_Plus_Optimizer) — Tool
- [Honeywell Forge APC](/Products/Honeywell_Forge_APC) — Tool
- [Custom Spreadsheet Trackers](/Products/Custom_Spreadsheet_Trackers) — Spreadsheet
- [In-House Python Scripts](/Products/In-House_Python_Scripts) — DIY
- [Seeq Advanced Analytics](/Products/Seeq_Advanced_Analytics) — Tool
- [Aveva Process Simulation](/Products/Aveva_Process_Simulation) — Tool

## Opportunity Win Conditions

**Kill Thresholds**:
- Average integration and deployment time exceeds 30 days
- Operator acceptance rate of cut recommendations falls below 60 percent after 45 days
- Measured yield improvement remains below 1.5 percent across the first 5 installations
- Hardware retrofit costs required to capture data exceed 5000 USD per site
**Leading Metrics**:
- Time from sensor connection to first valid yield prediction
- Percentage of automated cut recommendations accepted by the operator
- Daily reduction in manual lab sample volume per shift
- Absolute percentage increase in usable spirit volume per batch
**What Proves Right**: Distilleries integrate the engine with their existing control systems within two weeks and accept the cut-point recommendations on daily production runs. Cohorts retain at over 85 percent after six months because the increased usable alcohol yield strictly covers the software cost. Operators pay the 20000 USD annual subscription without extended pilots after validating a 3 percent or greater improvement in total spirit yield.
**What Proves Wrong**: Head distillers override the system cut recommendations on more than half of the batches due to unquantified flavor profile concerns. Integration with legacy sensors requires more than 50 hours of custom engineering per site. The engine fails to reduce manual lab sampling frequency by at least 30 percent, eliminating the primary labor cost offset.

## Opportunity Build Profile

**Hardest Part**: Building a predictive model that respects non-linear thermodynamic constraints in real-time to prevent physical plant destabilization or safety incidents.
**Min Viable Scope**: Provide a read-only setpoint recommender for single-column continuous distillation. Exclude closed-loop control, multi-column cracking, and predictive maintenance.
**Cold Start Problem**: Industrial operators refuse to deploy unproven control models on live columns. Break this by ingesting historical OSIsoft PI data from a single design partner to prove theoretical yield lift entirely offline.
**Time To First Value**: 4 weeks to validate offline yield models, gated by the extraction and mapping of historical historian sensor data.
**Data Moat Available**: true
**Technical Difficulty**: High

## Neighborhood

### Where the gap lives

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

### Incumbent in

- [Seeq Advanced Analytics](/Products/Seeq_Advanced_Analytics) — incumbent in · Products
- [Honeywell Forge APC](/Products/Honeywell_Forge_APC) — incumbent in · Products
- [In-House Python Scripts](/Products/In-House_Python_Scripts) — incumbent in · Products
- [Aspen Plus Optimizer](/Products/Aspen_Plus_Optimizer) — incumbent in · Products
- [Aveva Process Simulation](/Products/Aveva_Process_Simulation) — incumbent in · Products
- [Custom Spreadsheet Trackers](/Products/Custom_Spreadsheet_Trackers) — incumbent in · Products

### Applies thesis

- [Commercial Distillery](/CompanyTypes/Commercial_Distillery) — applies thesis · CompanyTypes

### Embodies

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

### Similar Opportunities

- [Feedstock Blending Optimizer](/Opportunities/Feedstock_Blending_Optimizer) — similar · Opportunities
- [Refiner Tuning Agent](/Opportunities/Refiner_Tuning_Agent) — similar · Opportunities
- [Autonomous Loop Tuning](/Opportunities/Autonomous_Loop_Tuning) — similar · Opportunities
- [Refinery Dynamic Deadband Adjustment](/Opportunities/Refinery_Dynamic_Deadband_Adjustment) — similar · Opportunities
- [Algorithmic Refiner Energy Optimization](/Opportunities/Algorithmic_Refiner_Energy_Optimization) — similar · Opportunities
- [AI Yield Optimization for Oleochemical Plants](/Opportunities/AI_Yield_Optimization_for_Oleochemical_Plants) — similar · Opportunities
- [Yield Setpoint API](/Opportunities/Yield_Setpoint_API) — similar · Opportunities
- [AI Scale-Up for Chemical Producers](/Opportunities/AI_Scale-Up_for_Chemical_Producers) — similar · Opportunities
- [Refiner Energy Optimizer](/Opportunities/Refiner_Energy_Optimizer) — similar · Opportunities
- [Thermal Energy Telemetry for Mills](/Opportunities/Thermal_Energy_Telemetry_for_Mills) — similar · Opportunities
- [Combustion Optimization API](/Opportunities/Combustion_Optimization_API) — similar · Opportunities
- [Peak Shedding for Chemical Plants](/Opportunities/Peak_Shedding_for_Chemical_Plants) — similar · Opportunities
- [Cracking Unit Diagnostics](/Opportunities/Cracking_Unit_Diagnostics) — similar · Opportunities
- [Digester Process Agent](/Opportunities/Digester_Process_Agent) — similar · Opportunities
- [Steam Load Balancing](/Opportunities/Steam_Load_Balancing) — similar · Opportunities
- [Bleach Dosing Engine](/CompanyTypes/BCTMP_Mills/Opportunities/Bleach_Dosing_Engine) — similar · Opportunities
- [Freeness Control Agent](/Opportunities/Freeness_Control_Agent) — similar · Opportunities
- [Dynamic Recipe Tuning for Biomanufacturing](/Opportunities/Dynamic_Recipe_Tuning_for_Biomanufacturing) — similar · Opportunities
- [Alarm Flood Triage](/Opportunities/Alarm_Flood_Triage) — similar · Opportunities
- [AI Polymer Batch Optimization](/Opportunities/AI_Polymer_Batch_Optimization) — similar · Opportunities