Reducing Unplanned Downtime by 15% with Predictive Maintenance
Customer profile
Industry
Global confectionery manufacturing
Footprint
High-throughput, 24/7 production lines
Operations
Complex continuous lines serving growing global demand
Maintenance maturity
Traditional preventive and manual inspection model transitioning to predictive maintenance
The challenge
As Darrell Lea's operations scaled, traditional maintenance methods struggled to keep pace. With a growing production footprint and an increasing number of critical assets, the maintenance team faced mounting pressure to maintain reliability without proportionally increasing headcount.
- Rising unplanned downtime as asset count and utilisation increased
- Limited capacity for preventive maintenance with technicians unable to inspect all critical assets frequently enough
- Reactive maintenance cycles increasing the risk of unexpected failures
- Manual inspections consuming valuable technician time
- Difficulty detecting operator-related and process-driven faults early
The approach
Darrell Lea partnered with Factory AI to modernise maintenance using predictive maintenance and continuous condition monitoring, integrated with existing workflows for rapid adoption.
- Continuous monitoring using vibration and temperature sensors
- AI-driven analysis to detect early signs of abnormal behaviour
- Automated alerts to flag developing issues before failure
- Actionable dashboards designed for maintenance teams
- Integration with existing maintenance workflows
Factory AI gave us earlier visibility into developing issues that we simply couldn't see before. That allowed our team to plan work, reduce unplanned downtime, and scale maintenance without adding headcount.
Engineering Manager, Darrell Lea
Results
Following deployment, Darrell Lea achieved a measurable reduction in downtime and a more scalable maintenance operating model.
Unplanned Downtime Reduction
15%
Meaningful cost savings and improved production stability
Maintenance Scalability
No added headcount
Capability scaled with production growth
Manual Inspection Burden
Reduced
Routine manual inspections were eliminated
Maintenance Posture
Reactive to predictive
Earlier intervention before downtime events
1. 15% reduction in unplanned downtime
- Unplanned equipment downtime reduced by 15% after deployment
- Earlier detection of developing faults improved intervention timing
- Improved identification of operator-related issues
- Better prioritisation of maintenance work across critical assets
2. More efficient maintenance operations
- Elimination of routine manual inspections, freeing technicians for higher-value work
- More efficient maintenance scheduling and resource allocation
- Reduced emergency repair costs due to earlier intervention
- Improved asset reliability across critical production equipment
3. Scalable maintenance without added headcount
- Condition monitoring and fault detection were automated
- Maintenance capability scaled with production growth
- No proportional increase in maintenance workforce required
- Actionable data supported prediction and pre-emption of downtime events
Challenges and learnings
- Predictive maintenance created earlier visibility into developing issues that were previously missed
- Automating detection reduced manual inspection burden and improved technician leverage
- Integrating insights into existing workflows accelerated practical adoption
What's next
- Continue expanding predictive coverage across additional critical assets
- Further refine operator and process fault detection patterns
- Use accumulated data to improve planning precision and reliability outcomes
Why it matters
- Delivers measurable downtime reduction
- Improves maintenance efficiency in fast-growing operations
- Reduces reliance on manual inspections
- Supports consistent production quality at scale
- Protects throughput and controls costs for continuous, high-value production lines
Want results like this in your plant?
Book a tailored walkthrough and see how Factory AI can surface faults earlier, reduce downtime, and improve maintenance planning.