Mission Critical IT in Mining: Keeping Underground Operations Running

Mining has come a long way from pickaxes and canaries. Today's underground operations run on sophisticated IT infrastructure that monitors everything from air quality to equipment health in real-time. When you're operating hundreds of metres below the surface, reliable technology isn't a luxury—it's a matter of life and death.

The Unique Challenges of Underground IT

Mining environments are brutal on technology. Underground IT systems must contend with:

• Extreme temperatures and humidity — Equipment rated for office environments simply won't survive
• Dust and debris — Fine particulate matter infiltrates everything
• Vibration and shock — Blasting operations and heavy machinery create constant physical stress
• Limited physical access — You can't just pop down to the server room when something fails
• RF interference — Rock, metal ore bodies, and heavy equipment create challenging wireless environments

These conditions demand ruggedised hardware, redundant systems, and careful planning that goes far beyond standard enterprise IT.

Critical Systems That Can't Fail

1. Underground Communication Networks

Voice and data communication underground saves lives. Modern mines rely on:

• Leaky feeder systems — Coaxial cables that "leak" radio signals along tunnels, providing continuous coverage
• Mesh WiFi networks — Self-healing networks that route around failures
• Through-the-earth (TTE) communication — Emergency systems that can penetrate solid rock when conventional systems fail
• VOIP and digital radio integration — Unified communications that work seamlessly above and below ground

When an emergency occurs, these systems must work instantly. There's no "please hold while we troubleshoot."

2. Personnel Tracking and Safety Systems

Knowing exactly where every person is underground isn't optional—it's a legal requirement in most jurisdictions and a moral imperative everywhere.

• Real-time location systems (RTLS) — Track personnel movements throughout the mine
• Cap lamp tracking — Smart cap lamps that report location and can detect if a miner is stationary too long
• Refuge bay monitoring — Ensure emergency shelters are properly equipped and functional
• Access control — Prevent unauthorised entry to hazardous areas

These systems generate massive amounts of data that must be processed, stored, and made accessible to surface operations centres in real-time.

3. Environmental Monitoring

Underground air can kill you in ways you won't see coming. Continuous monitoring tracks:

• Gas levels — Methane, carbon monoxide, hydrogen sulphide, and oxygen depletion
• Ventilation flow — Ensuring adequate fresh air reaches working areas
• Temperature and humidity — Critical for both human safety and equipment operation
• Seismic activity — Early warning systems for rockbursts and ground movement
• Water ingress — Detecting flooding before it becomes catastrophic

A single sensor failure in a critical location can have fatal consequences. Redundancy and fail-safe designs are non-negotiable.

4. Equipment Telemetry and Automation

Modern mining equipment is essentially a data centre on wheels:

• Load-haul-dump (LHD) vehicles — Many now operate autonomously or semi-autonomously
• Drilling rigs — Precision drilling guided by 3D geological models
• Conveyor systems — Kilometres of belts monitored for wear, alignment, and throughput
• Pumping systems — Keeping the mine dry requires constant monitoring

The data from these systems feeds into predictive maintenance algorithms, production optimisation, and safety systems.

Network Architecture for Underground Operations

Network infrastructure is the foundation everything else depends on. Get this wrong, and every system built on top of it inherits those weaknesses. Get it right, and you have a platform that can support current operations and future expansion.

The Fibre Backbone

Fibre optic cabling forms the high-speed backbone of any serious underground network. Unlike copper, fibre is immune to electromagnetic interference from heavy machinery, blasting operations, and the electrical systems that power underground equipment.

Fibre selection considerations:

• Single-mode vs multi-mode — Single-mode fibre supports longer distances (up to 10km+ without amplification) and higher bandwidths, making it ideal for shaft and main tunnel runs. Multi-mode works well for shorter lateral connections where cost is a factor.
• Cable construction — Armoured cables with steel wire or corrugated steel tape protect against crushing, rodent damage, and accidental impacts. Gel-filled loose tube designs prevent water ingress.
• Fibre count — Always install more fibres than you currently need. The cost difference between 12-fibre and 48-fibre cable is minimal compared to the cost of pulling new cable later.

Installation best practices:

• Route cables through dedicated cable trays or conduit, never alongside power cables
• Use mine-rated splice enclosures at junction points—standard enclosures won't survive the environment
• Install fibre in protective innerduct for future-proofing and easier replacement
• Maintain detailed documentation of every splice point, cable route, and fibre assignment

Network Topology Design

The physical and logical layout of your network determines both performance and resilience.

Ring topologies provide automatic failover—if a cable is cut, traffic routes the other way around the ring. This is essential for main backbone connections where a single failure cannot be allowed to isolate sections of the mine.

Star topologies from distribution points to end devices simplify troubleshooting and reduce the blast radius of failures. A failed switch affects only devices connected to it, not the entire network.

Hybrid designs combine both approaches: resilient rings connecting major distribution points, with star connections to individual devices and work areas.

Switching and Routing Infrastructure

Underground switches and routers must be industrial-grade equipment rated for harsh environments:

• Extended temperature ranges — Standard IT equipment is rated for 0-40°C; underground conditions often exceed this
• Conformal coating — Circuit boards coated to resist moisture and dust
• Ruggedised enclosures — IP65 or higher ratings for dust and water resistance
• DIN rail mounting — For installation in industrial cabinets

Key features to specify:

• Managed switches with full SNMP monitoring capabilities
• Support for link aggregation (LACP) to bundle multiple connections
• Spanning Tree Protocol (STP) or Rapid STP for loop prevention in redundant topologies
• Power over Ethernet (PoE+/PoE++) to power cameras, access points, and sensors without separate power runs

VLAN Segmentation

Not all traffic is equal. Proper VLAN (Virtual LAN) design separates traffic types to improve security, performance, and manageability:

• Safety systems VLAN — Gas monitoring, personnel tracking, emergency communications. Highest priority, strictly isolated.
• Operations VLAN — SCADA, equipment telemetry, automation systems. High priority, limited access.
• Voice VLAN — VoIP traffic with QoS guarantees for call quality.
• Video VLAN — CCTV and surveillance. High bandwidth but tolerant of minor delays.
• General data VLAN — Office applications, email, internet access. Lowest priority.
• Management VLAN — Network device administration. Restricted access, never routed to general networks.

This segmentation contains security breaches, prevents broadcast storms from affecting critical systems, and allows granular quality of service policies.

Quality of Service (QoS)

When bandwidth is constrained—and underground it often is—QoS ensures critical traffic gets through first.

Traffic prioritisation hierarchy:

1. Emergency/safety alerts — Absolute priority, pre-empts everything
2. Voice communications — Low latency requirements, small packets
3. Real-time telemetry — Equipment and environmental sensors
4. Video surveillance — Important but can tolerate brief degradation
5. General data — Best effort delivery

QoS must be configured end-to-end, on every switch and router in the path. A single misconfigured device can negate the entire QoS architecture.

Wireless Infrastructure

Wireless connectivity extends the network to mobile equipment, personnel, and areas where cabling is impractical.

WiFi deployments:

• WiFi 6 (802.11ax) — Better performance in high-density environments, improved power efficiency for battery-powered devices
• Industrial access points — Rated for temperature, dust, and vibration; often with external antenna connections for directional coverage
• Mesh capabilities — Self-healing networks that maintain connectivity when individual APs fail
• Seamless roaming — Critical for tracking systems and mobile devices; look for 802.11r/k/v support

Private LTE/5G:

Increasingly, mines are deploying private cellular networks for reliable wide-area coverage:

• Dedicated spectrum (in South Africa, ICASA has allocated spectrum for private networks)
• Better penetration through rock and around obstacles than WiFi
• Native support for massive numbers of IoT devices
• Carrier-grade reliability and security

Leaky feeder integration:

Legacy leaky feeder systems for radio communications can coexist with modern IP networks. Gateways bridge analogue radio traffic onto the IP network, enabling unified communications and recording.

Edge Computing Architecture

Latency matters. A safety system that takes 500ms to respond might as well take 500 years. Edge computing places processing power underground, close to where data is generated and decisions must be made.

Edge node design:

• Ruggedised servers — Fanless designs with solid-state storage survive vibration and dust
• Local processing — Safety algorithms, autonomous vehicle control, real-time analytics
• Store-and-forward — Buffer critical data when surface connectivity fails
• Synchronisation — Automatic sync with surface systems when connectivity restores

Typical edge deployment:

• Primary edge nodes at major level stations (every 500-1000m of depth)
• Secondary nodes at key production areas
• Micro-edge devices integrated into autonomous equipment

Bandwidth Planning

Underground bandwidth requirements are growing exponentially:

A medium-sized operation can easily require 1-2 Gbps of backbone capacity, with growth projections pushing toward 10 Gbps as automation increases.

Latency Requirements

Different applications have different latency tolerances:

• Emergency stop systems — <10ms (often hardwired, not networked)
• Autonomous vehicle control — <50ms
• Voice communications — <150ms (one-way)
• Real-time video — <200ms
• Telemetry and monitoring — <1 second
• General data — Best effort

Network design must account for these requirements, with appropriate QoS and routing to meet each application's needs.

Physical Layer Considerations

The physical infrastructure is often overlooked but determines long-term reliability:

• Cable routing — Dedicated pathways away from water, vehicle traffic, and blast zones
• Junction boxes — IP66-rated enclosures with proper cable glands
• Grounding and bonding — Essential for equipment protection and safety
• Labelling — Every cable, port, and device clearly identified; critical for troubleshooting under pressure
• Spare capacity — Pre-installed dark fibres and empty conduit for future expansion

Network Monitoring and Management

You can't manage what you can't see. Comprehensive monitoring provides:

• Real-time visibility — Dashboard showing network health at a glance
• Alerting — Immediate notification of failures, threshold breaches, or anomalies
• Historical trending — Capacity planning and problem pattern identification
• Configuration management — Centralised backup and deployment of device configurations
• Automated diagnostics — Self-healing networks that can reroute around failures

Modern network management platforms can integrate with mine operations centres, providing IT visibility alongside production and safety dashboards.

Integration with Surface Networks

Underground networks don't exist in isolation. Clean integration with surface infrastructure requires:

• Demarcation points — Clear boundaries of responsibility between underground and surface teams
• Firewall placement — Security controls at the surface/underground boundary
• Bandwidth management — Preventing underground traffic from overwhelming surface WAN links
• Unified management — Single pane of glass visibility across the entire operation

Redundancy by Design

Mission-critical underground systems typically feature:

• Dual-path connectivity — Independent physical routes to surface (e.g., main shaft and ventilation shaft)
• Diverse media — Fibre plus wireless backup for critical links
• Uninterruptible power supplies — Battery backup at every network node, sized for extended outages
• Hot-standby equipment — Automatic failover for critical switches and routers
• Graceful degradation — Systems that continue operating with reduced functionality when components fail
• Geographic separation — Critical equipment in separate locations to avoid single points of failure

Edge Computing Underground

You can't rely on surface data centres for time-critical decisions. Edge computing nodes positioned underground handle:

• Real-time safety system processing
• Local caching of critical data
• Autonomous equipment decision-making
• Store-and-forward when surface connectivity fails

Cybersecurity in Mining Operations

Mining operations face unique cybersecurity challenges:

Operational Technology (OT) Security

Much of the equipment underground runs on industrial control systems (ICS) and SCADA platforms that weren't designed with modern cybersecurity in mind. Attackers who compromise these systems could:

• Disable ventilation systems
• Override safety interlocks
• Manipulate sensor readings
• Take control of autonomous vehicles

Converged IT/OT Networks

The drive for efficiency has pushed mines to connect previously isolated OT systems to enterprise IT networks. This convergence creates attack paths that didn't exist before.

Best practices include:

• Network segmentation — Air gaps or strict firewalling between IT and OT
• Anomaly detection — Identifying unusual patterns in industrial protocols
• Secure remote access — Vendor access that doesn't compromise the network
• Regular vulnerability assessments — Many ICS devices can't be patched; compensating controls are essential

South African Mining Context

South Africa's mining sector operates some of the world's deepest mines, with operations extending below 4,000 metres. This creates additional IT challenges:

• Extreme geothermal heat — Refrigeration systems for both people and equipment
• Long cable runs — Signal integrity over extreme distances
• Legacy infrastructure — Many operations have IT systems spanning decades
• Load shedding — Eskom's unreliable grid requires robust backup power

The industry is also navigating the transition to mechanisation and automation—a shift that dramatically increases IT dependency.

Planning for the Worst

Disaster recovery in mining isn't theoretical. Rock falls, flooding, fires, and equipment failures happen. Your IT systems need plans for:

• Communication system failure — Alternative methods to reach underground personnel
• Surface data centre loss — Underground edge systems that can operate independently
• Extended power outages — Beyond what UPS systems can handle
• Cyber incidents — Isolation and recovery procedures for compromised systems

Regular drills and testing are essential. A disaster recovery plan that's never been tested isn't a plan—it's a hope.

The Role of Managed IT Services

Mining companies face a choice: build internal IT expertise for these specialised systems or partner with providers who understand the unique demands of the industry.

A good managed services partner brings:

• 24/7 monitoring — Because underground operations never stop
• Rapid response capabilities — Including personnel cleared for underground work
• Vendor relationships — Access to specialised mining IT equipment and support
• Compliance expertise — Understanding of mining regulations and safety requirements
• Technology roadmap planning — Helping operations evolve without disrupting production

Conclusion

IT in mining isn't about the latest consumer gadgets or trendy cloud services. It's about reliable, ruggedised systems that work when lives depend on them. The stakes are higher, the environments are harsher, and the margin for error is essentially zero.

As mining operations become more automated and data-driven, the line between IT systems and safety systems continues to blur. The organisations that thrive will be those that treat their underground IT infrastructure with the seriousness it demands.

Need help with mission-critical IT infrastructure? Dexani provides managed IT services for demanding environments where reliability isn't optional. Contact us to discuss your requirements.