Back to Blog

How Mining Operations Achieve 99.9% Connectivity Uptime

The engineering behind mission-critical mining communications

Modern mining operations run on connectivity. Autonomous haul trucks, real-time fleet management, remote operations centres, safety monitoring systems, SCADA networks—all depend on communications that simply can't fail.

Achieving 99.9% uptime means less than 9 hours of downtime per year. In a 24/7 operation where a single hour of downtime can cost $150,000 or more, that target represents serious engineering and operational discipline. Here's how leading mining operations achieve it.

Understanding 99.9%

Before diving into how, it's worth understanding what 99.9% actually means:

UptimeAnnual DowntimeMonthly Downtime
99.0%87.6 hours7.3 hours
99.5%43.8 hours3.6 hours
99.9%8.76 hours44 minutes
99.99%52.6 minutes4.4 minutes

Each additional "9" represents a tenfold improvement in reliability—and typically requires proportional increases in investment and operational sophistication.

The Multi-Path Architecture

Single-path connectivity—whether VSAT, Starlink, or any other technology—cannot reliably achieve 99.9% uptime. Every technology has failure modes. The solution is multiple independent paths.

Path Diversity

Effective multi-path architectures combine technologies with different failure modes:

  • LEO satellite (e.g., Starlink, OneWeb) — Excellent bandwidth and latency, but susceptible to rain fade and congestion
  • GEO satellite (e.g., VSAT) — Proven reliability but higher latency, different failure characteristics
  • Terrestrial (e.g., microwave, 4G/5G) — Where available, provides diversity from all satellite failure modes

The key insight: technologies that share failure modes don't provide true redundancy. Two LEO services from the same constellation may fail simultaneously. LEO plus GEO provides better diversity because they fail independently.

SD-WAN Orchestration

Multiple paths require intelligent orchestration to achieve seamless failover:

  • Continuous monitoring of all path health
  • Sub-second failover when degradation is detected
  • Application-aware routing (latency-sensitive traffic to appropriate paths)
  • Load balancing across healthy paths
  • Graceful degradation under partial failures

Physical Redundancy

Beyond path diversity, physical infrastructure redundancy matters:

  • Redundant power supplies with UPS and generator backup
  • Spare equipment on-site for rapid replacement
  • Diverse cable routes where possible
  • Equipment rated for environmental extremes

24/7 Monitoring

High availability requires constant vigilance. Issues caught early cause less impact.

Network Operations Centre

A dedicated NOC provides:

  • Real-time monitoring of all connectivity paths
  • Automated alerting when thresholds are exceeded
  • Engineers available 24/7 to investigate and respond
  • Escalation procedures for critical issues

Proactive Intervention

The best NOCs don't just respond to failures—they anticipate them:

  • Trend analysis identifies degradation before failure
  • Predictive maintenance based on performance data
  • Weather monitoring to prepare for conditions that affect connectivity
  • Capacity planning to stay ahead of utilisation growth

Site Integration

Effective monitoring extends to the site level:

  • Local technicians or trained personnel for physical intervention
  • Remote access to equipment for diagnostics and configuration
  • Spare parts and equipment available on-site
  • Clear procedures for common scenarios

Operational Best Practices

Technology alone doesn't achieve 99.9%. Operational discipline is equally important.

Change Management

Many outages result from changes gone wrong. Disciplined change management includes:

  • Testing changes before production deployment
  • Scheduled maintenance windows for non-urgent changes
  • Rollback procedures for every change
  • Post-implementation verification

Incident Management

When issues occur, rapid effective response minimises impact:

  • Clear severity definitions and escalation paths
  • Runbooks for common scenarios
  • Communication protocols with stakeholders
  • Post-incident review and improvement

Regular Testing

Redundancy only works if it's verified:

  • Periodic failover testing to verify redundancy works
  • Disaster recovery exercises
  • Performance testing under load
  • Equipment and procedure validation

Mining-Specific Considerations

Pit Operations

Open pit environments present unique challenges:

  • Pit walls can obstruct satellite signals at low angles
  • Dust affects radio propagation and equipment
  • Mobile equipment requires in-pit connectivity solutions
  • Blasting creates temporary communication blackouts

Underground Operations

Underground mines require different approaches entirely:

  • Leaky feeder or WiFi networks underground
  • Reliable surface-to-underground connectivity
  • Redundancy at the surface gateway
  • Emergency communication systems

Remote and Fly-Over Sites

Sites with no terrestrial options must achieve diversity through satellite alone:

  • Multiple satellite technologies (LEO + GEO minimum)
  • Multiple providers where possible
  • Higher on-site spares inventory
  • More extensive on-site technical capability

The Investment Case

99.9% uptime costs more than 99%. The investment case depends on downtime costs:

For a mine with $150,000/hour downtime cost:

  • Moving from 99% to 99.9% saves approximately 79 hours/year
  • Value of avoided downtime: 79 × $150,000 = $11.85 million/year
  • Multi-path connectivity might cost $500,000-$1,000,000/year more than single-path
  • ROI: 12-24x return

Even with conservative assumptions about improvement, the investment typically pays for itself many times over.

Frequently Asked Questions

Can Starlink alone achieve 99.9%?

No single technology can reliably achieve 99.9% uptime. Starlink doesn't offer SLAs and has documented outage events. For best-effort use, it's excellent. For mission-critical operations requiring 99.9%, it should be one path in a multi-path architecture.

What's the minimum for 99.9%?

At minimum: two diverse connectivity paths with automatic failover, 24/7 monitoring, and on-site support capability. The specific technologies depend on location, but path diversity is non-negotiable.

How do you measure uptime?

Uptime should be measured at the service level—can users access the applications they need? Not just whether equipment is powered on. Measurement should be continuous, automated, and agreed in service level definitions.

Engineer Your Uptime Strategy

Our team has designed connectivity solutions for some of Australia's largest mining operations. We can help you understand what's required to achieve your uptime targets.

Discuss Your Requirements
Back to Blog