# Predictive Turnaround Scheduling

*/Opportunities/Predictive_Turnaround_Scheduling*

## Opportunity Overview

**Wedge**: Target mid-sized chemical manufacturing plants in the US Gulf Coast as the initial beachhead. These facilities experience punishing downtime costs but lack the dedicated scheduling departments of super-major oil refineries, making them highly receptive to automated scheduling for short outages. Expand outward by graduating to tier-1 oil refineries with month-long megaprojects, then push horizontally into routine daily maintenance scheduling.
**Timing**: Advances in multimodal foundation models now permit automated extraction of resource constraints from unstructured shift logs, work orders, and piping diagrams. Combined with modern cloud-based heuristic solvers, the system computes probabilistic schedule adjustments in seconds, a task that previously required overnight batch processing.
**Why This I C P**: Turnaround managers at industrial facilities bear the direct cost of schedule overruns, which frequently exceed $1 million per day in lost production. Their performance is judged entirely on bringing the plant back online on time, creating immediate willingness to adopt tools that guarantee critical path visibility.
**Size Of Prize**: There are roughly 4,500 heavy industrial facilities like refineries and chemical plants in North America and Europe. Capturing an average software and service spend of $75,000 per facility annually yields a $337M addressable prize.
**Gap Narrative**: Industrial plant turnarounds operate on static schedules that fail the moment execution begins due to emergent work and field delays. Master schedulers lack a dynamic system that ingests daily shift progress to recalculate the critical path and shift resource constraints instantly. This leaves turnaround managers running massive outages off outdated Gantt charts, resulting in days of lost production.
**Defensibility**: Defensibility compounds through the accumulation of proprietary execution data. As the system processes thousands of work orders, it builds an empirical database of actual task durations versus planned durations for specific equipment types. This historical variance data trains the predictive model, creating a localized accuracy advantage that generic scheduling algorithms cannot match.
**Why This Thesis**: Service-as-Software fits this problem because updating schedules requires expert reasoning to resolve resource collisions and sequence logic. An agentic system acts as a digital master scheduler, automatically processing shift reports to adjust the schedule rather than requiring human planners to manually update complex enterprise software.

## Opportunity Linked Thesis

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

## Opportunity Linked I C P

**Icp**: [Oil Refinery](/CompanyTypes/Oil_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**: ~$100-150M North American and European tier-1 refineries
**S O M**: ~$10-25M
**T A M**: ~4,000 global downstream refining and chemical facilities × ~$150k/yr ≈ ~$600M
**Growth Rate**: ~7-11%/yr, driven by aging facility infrastructure demanding more frequent outage cycles and the retirement of veteran turnaround planners
**Paid Comparable Spend**: ~$300k-600k/yr per facility on legacy Primavera P6 licenses, external turnaround consultants, and dedicated master scheduler salaries

## Opportunity Incumbents

- [Oracle Primavera P6](/Products/Oracle_Primavera_P6) — Tool
- [Custom Excel Trackers](/Products/Custom_Excel_Trackers) — Spreadsheet
- [Prometheus Turnaround Management](/Products/Prometheus_Turnaround_Management) — Tool
- [SAP Maintenance Planner](/Products/SAP_Maintenance_Planner) — Tool
- [External STO Consultants](/Products/External_STO_Consultants) — Service
- [Microsoft Project Professional](/Products/Microsoft_Project_Professional) — Tool

## Opportunity Win Conditions

**Kill Thresholds**:
- Primavera P6 or SAP data integration takes > 14 days per facility
- Automated sequence acceptance rate < 60% during initial pilot
- Sales cycle for a paid POC exceeds 120 days
- D30 planner retention < 40% during active turnaround prep phase
**Leading Metrics**:
- Time-to-successful Primavera P6 schedule ingestion
- Automated sequence acceptance rate
- Manual override percentage per generated schedule
- Daily active usage by maintenance planners during T-minus 90 days
- Average time to resolve identified critical-path conflicts
**What Proves Right**: Master schedulers integrate the system alongside Primavera P6 and actively accept the automated critical-path adjustments during the T-minus 90 day planning window. Facilities convert $25k paid pilots into $150k annual contracts after verifying a reduction in schedule conflict resolution time. User cohorts of maintenance planners log in daily during the active outage execution phase to re-optimize shift schedules based on daily progress.
**What Proves Wrong**: Turnaround managers refuse to authorize cloud uploads of proprietary plant equipment data, blocking the initial data ingestion process. Veteran planners manually override more than half of the suggested scheduling sequences because the logic fails to account for undocumented physical space constraints like concurrent scaffolding limits. Sales cycles stall past six months as plant IT departments require on-premise deployments the architecture cannot support.

## Opportunity Build Profile

**Hardest Part**: Extracting and normalizing dependency logic from messy legacy Primavera P6 files and solving the constraint satisfaction problem for dynamic labor and parts availability.
**Min Viable Scope**: Deliver static schedule risk analysis for single-unit facility turnarounds to flag probable critical path delays based on baseline task times. Deliberately exclude live intra-day rescheduling, automated contractor dispatching, and direct materials procurement.
**Cold Start Problem**: Requires historical execution actuals to train duration risk predictions before generating accurate delay alerts. Break this by offering an initial ingestion tool that converts existing static schedules into baseline dependency health-checks.
**Time To First Value**: 2 to 4 weeks of onboarding gated by the export and data mapping of legacy facility schedule files.
**Data Moat Available**: true
**Technical Difficulty**: High

## Neighborhood

### Surfaced from

- [Captive Pulp and Paper Cogeneration Plants](/CompanyTypes/Captive_Pulp_and_Paper_Cogeneration_Plants) — surfaces · CompanyTypes

### Incumbent in

- [Custom Excel Tracker](/Products/Custom_Excel_Tracker) — incumbent in · Products
- [SAP Maintenance Planner](/Products/SAP_Maintenance_Planner) — incumbent in · Products
- [Oracle Primavera P6](/Products/Oracle_Primavera_P6) — incumbent in · Products
- [Prometheus Turnaround Management](/Products/Prometheus_Turnaround_Management) — incumbent in · Products
- [External STO Consultants](/Products/External_STO_Consultants) — incumbent in · Products
- [Microsoft Project Professional](/Products/Microsoft_Project_Professional) — incumbent in · Products

### Applies thesis

- [Oil Refinery](/CompanyTypes/Oil_Refinery) — applies thesis · CompanyTypes

### Embodies

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

### Similar Opportunities

- [AI Turnaround Planner](/Opportunities/AI_Turnaround_Planner) — similar · Opportunities
- [Turnaround Scheduling Agent](/Opportunities/Turnaround_Scheduling_Agent) — similar · Opportunities
- [Turnaround Logistics Coordinator](/Opportunities/Turnaround_Logistics_Coordinator) — similar · Opportunities
- [Ghost Capacity](/Opportunities/Ghost_Capacity) — similar · Opportunities
- [Resonance Labs](/Industries/Manufacturing/Opportunities/Resonance_Labs) — similar · Opportunities
- [Generative Site Scheduling](/Opportunities/Generative_Site_Scheduling) — similar · Opportunities
- [Predictive Shift Scheduling For Manufacturing](/Opportunities/Predictive_Shift_Scheduling_For_Manufacturing) — similar · Opportunities
- [Skill Matching for Manufacturing](/Opportunities/Skill_Matching_for_Manufacturing) — similar · Opportunities
- [Predictive Critical Path](/Opportunities/Predictive_Critical_Path) — similar · Opportunities
- [Shift Forge](/Opportunities/Shift_Forge) — similar · Opportunities
- [Schedule Recovery Service](/Opportunities/Schedule_Recovery_Service) — similar · Opportunities
- [Predictive Maintenance Agent](/Opportunities/Predictive_Maintenance_Agent) — similar · Opportunities
- [Dynamic Scheduling for Textile Mills](/Opportunities/Dynamic_Scheduling_for_Textile_Mills) — similar · Opportunities
- [Shipyard Scheduling Software](/Opportunities/Shipyard_Scheduling_Software) — similar · Opportunities
- [Autonomous Loop Tuning](/Opportunities/Autonomous_Loop_Tuning) — similar · Opportunities
- [Schedule Dependency Engine](/Opportunities/Schedule_Dependency_Engine) — similar · Opportunities
- [Headless Supply Orchestration](/Occupations/Production_Occupations/Opportunities/Headless_Supply_Orchestration) — similar · Opportunities
- [AI Maintenance Dispatch](/Industries/Manufacturing/Opportunities/AI_Maintenance_Dispatch) — similar · Opportunities
- [Haul-Out Scheduling](/CompanyTypes/Full-Service_Boatyards/Opportunities/Haul-Out_Scheduling) — similar · Opportunities
- [Changeover Orchestration Agent](/Opportunities/Changeover_Orchestration_Agent) — similar · Opportunities