What challenges did Georgia-Pacific face before AI implementation?

Reduce paper tears and breaks in converting lines Predict equipment failure 60–90 days ahead to reduce unplanned downtime Consolidate data and site-specific knowledge from distributed experts

What was the impact of the AI solution?

For one converting line, eliminated 40% of parent-roll tears Increased profits by millions of dollars for one production line At one OSB facility, reduced waste by 30% and increased annual profits by millions of dollars Can predict equipment failure 60–90 days in advance to help reduce unplanned downtime Rapidly scaling the approach across similar facilities

Rank#937of 3,449

Source

Evidence: Medium50/100

Georgia-Pacific predicts converting line failure and optimizes production with AWS

Use case typePredictive maintenanceUpdated Jun 13, 2026

Georgia-Pacific, owned by Koch Industries, produces paper and tissue parent rolls at manufacturing facilities across North America. The company needed to reduce tears and breaks in converting lines, cut unplanned downtime, and predict equipment failure 60–90 days in advance. To address the challenge, Georgia-Pacific built an AWS-based advanced analytics solution centered on Amazon S3, Amazon EMR, and Amazon SageMaker. Real-time machine data was streamed into a central S3 data lake, transformed with EMR, and used to train ML models that recommend optimum machine speeds and detect risk of failure. The solution also helped Georgia-Pacific consolidate disparate production data and expert knowledge into a centralized analytics approach that more experienced operators and central support teams could use to improve production decisions.

Organization: Georgia-Pacific
Industry: Manufacturing
Location: United States
Published: June 2026

Reported outcomes

60-90 days

failure prediction horizonTime & speed

−40%parent-roll tears reduced−30%waste reduced

Strategic outcomes

New product / capabilityPredicted equipment failures in advanceNew product / capabilityOptimized converting line speedsBetter decisions & insightCentralized production decision makingBetter decisions & insightConsolidated production data and expertise

Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

Parent-roll tears reduced: 40% decrease

AWS Solutions Case StudyJun 9, 2026Customer storyExplicit claimMedium evidence strength

eliminated 40 percent of parent-roll tears during the converting process

Normalized claim

Waste reduced: 30% decrease

AWS Solutions Case StudyJun 9, 2026Customer storyExplicit claimMedium evidence strength

seen a 30 percent reduction in waste associated with the chipping process

Normalized claim

Failure prediction horizon: 60-90 days increase

AWS Solutions Case StudyJun 9, 2026Customer storyExplicit claimMedium evidence strength

predict equipment failure 60-90 days in advance

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: Georgia-Pacific
Provider: AWS
Maturity: Unknown
Linked source: AWS Solutions Case Study

No explicit deployment-stage evidence found.

Customer identity supportedSource describes one deploymentMaturity evidence evaluated

Primary read

Use case focus

Showing 3 of 3

1Predictive Maintenance
2Production Optimization
3Industrial Analytics

Reduce paper tears and breaks in converting lines
Predict equipment failure 60–90 days ahead to reduce unplanned downtime
Consolidate data and site-specific knowledge from distributed experts

Built an AWS-based advanced analytics solution using Amazon S3 as a central data lake
Used Amazon EMR to transform and structure manufacturing data
Used Amazon SageMaker to build, train, and deploy ML models that recommend optimum converting line speeds and anticipate equipment failures
Streamed real-time machine data from manufacturing equipment into the centralized data lake
Enabled real-time feedback to operators and centralized support decision making

Technologies

Amazon S3 Amazon EMR Amazon SageMaker

For one converting line, eliminated 40% of parent-roll tears
Increased profits by millions of dollars for one production line
At one OSB facility, reduced waste by 30% and increased annual profits by millions of dollars
Can predict equipment failure 60–90 days in advance to help reduce unplanned downtime
Rapidly scaling the approach across similar facilities

Architecture

Real-time manufacturing data from machines is streamed into Amazon S3 as a central data lake. Amazon EMR transforms the data into structured outputs, and Amazon SageMaker trains and deploys models that generate machine-speed recommendations and failure predictions for operators and support teams.

Sources & evidence1

Evidence: Medium50/100Evidence strength