Stornoway Diamonds had already deployed TerraTraax to capture structured downtime events and failure information for its mobile fleet. When a series of hydraulic hose ruptures caused multiple fire events on haul trucks, the maintenance and reliability team used TerraTraax data—combined with equipment engine hours from onboard systems and PM compliance data from a separate scheduling tool—to identify the root cause.
This integrated analytical approach revealed a clear failure cluster between 7,000–8,000 engine hours and exposed critical gaps in preventive maintenance timeliness. These insights enabled Stornoway to introduce a mid-life hydraulic hose replacement strategy and refine inspection protocols, resulting in the complete elimination of hydraulic-related fires.
The haul truck fleet at Stornoway was in its first useful lifecycle, with no major component overhauls completed. Overhaul targets were approximately 14,000 hours.
Despite a strong safety culture, the site experienced four hydraulic hose rupture events over a short period—each creating fire conditions. Fire suppression systems activated correctly and no injuries occurred, but the recurrence indicated a deeper reliability issue requiring structured analysis.
1. Early Hydraulic Hose Failures
Hoses were failing significantly earlier than expected. Each failure involved high-pressure fluid release, contact with hot engine surfaces, and ignition. Suppression systems successfully contained the events, but recurrence signaled a systemic issue.
2. PM Compliance Issues (from Scheduling Tool)
Analysis of PM scheduling data revealed that roughly 30% of preventive maintenance tasks—especially hydraulic inspections—were completed late. This led to missed early-warning indicators such as hydraulic weeping, abrasion, or fitting wear.
3. Need for Correlated Failure Pattern Analysis
To uncover the true root cause, the reliability team merged three datasets into a single analytical view:
· TerraTraax downtime & failure event data
· Engine hour readings from CMMS
· PM compliance data from their scheduling tool
This integrated dataset provided the complete picture needed for discovery.
1. TerraTraax Provided the Structured Failure Dataset
TerraTraax delivered consistent downtime classifications, detailed failure modes, and safety-related event tracking. While TerraTraax did not record engine hours or PM completion metrics directly, it provided the structured failure event backbone required for deeper analysis.
2. Merging TerraTraax with Engine Hour Data
By aligning TerraTraax failure records with engine hour readings, the team discovered a clear cluster:Most hose failures happened between 7,000 and 8,000 hours.This discovery contradicted the expected 14,000-hour overhaul interval and revealed a premature wear pattern.
3. Correlating Failures with PM Compliance
Linking TerraTraax failure events with PM scheduling data showed that late preventive maintenance were a significant contributor. Missed inspection windows meant early leak indicators were not caught in time.
4. Power BI Analytics Unlocked the Insights
By bringing all datasets into Power BI, Stornoway was able to visualize:
· Failure clustering
· Condition-based deterioration signals
· PM delays
· Root cause patterns
These insights enabled rapid decision-making and targeted reliability interventions.
1. Improved PM Timeliness
PM planning was adjusted to ensure critical inspections occurred on time, reducing missed early-warning signs.
2. Updated Inspection Checklists
Existing hydraulic inspection checklists were refined to include clearer indicators, and technicians were coached to identify deterioration patterns more effectively.
3. Condition-Based Maintenance Mindset
The team shifted from “checklist completion” to true condition assessment. Detailed technician comments were recorded, improving future detection and trending.
4. Mid-Life Hydraulic Hose Replacement at 7,000 Hours
Based on the clustering analysis, Stornoway introduced a proactive mid-life replacement strategy at 7,000 hours, which:
· Cost approximately $65,000 per truck
· Avoided an estimated $74,000 per unplanned failure event
This intervention directly addressed the primary risk window discovered through data analysis.
