🎓 Lesson 10 D5

Situational Awareness Dashboard Taxonomy

A situational awareness dashboard taxonomy is a clear, organized way to group and label all the real-time information displays that help remote operators understand what’s happening in a mine — like equipment status, environmental conditions, and safety alerts — so they can make fast, safe decisions.

🎯 Learning Objectives

  • Explain the four-tier hierarchy of the ROC Situational Awareness Taxonomy (System–Process–Event–Anomaly)
  • Design a dashboard layout that maps sensor data to the correct taxonomy level using ISO 11064-4 ergonomic criteria
  • Analyze dashboard clutter by calculating Cognitive Load Index (CLI) from widget count, update frequency, and color coding density
  • Apply IEC 62591 alarm rationalization rules to classify and prioritize 10+ real-time alerts into taxonomy-aligned severity tiers

📖 Why This Matters

In 2023, 68% of unplanned ROC downtime was traced to ambiguous or conflicting dashboard information—not sensor failure. When a haul truck’s proximity alert blinks red *while* its GPS shows it stationary inside a maintenance bay, operators waste critical seconds reconciling contradictions. A robust taxonomy eliminates this ambiguity by ensuring every pixel on screen has a defined meaning, source, timeliness, and action implication. It transforms dashboards from 'data walls' into decision-ready command interfaces—directly impacting safety, productivity, and regulatory audit readiness.

📘 Core Principles

The ROC Situational Awareness Taxonomy rests on three pillars: (1) Semantic Consistency — each data class (e.g., 'ventilation airflow') maps to one canonical definition, unit, and refresh interval across all subsystems; (2) Cognitive Layering — information is structured in four ascending tiers: System (asset health), Process (workflow state), Event (time-bound occurrence), Anomaly (deviation requiring action); (3) Role-Based Filtering — dashboards dynamically suppress or highlight elements based on operator role (e.g., shift supervisor sees fleet KPIs; geotechnical engineer sees borehole strain trends). These principles are codified in ISO 11064-4 (Ergonomics of control rooms) and reinforced by Rio Tinto’s ROC Design Standard v3.2, which mandates taxonomy alignment for all vendor integrations.

📐 Cognitive Load Index (CLI)

CLI quantifies dashboard complexity per operator workstation. Values >12 indicate high risk of missed alerts or delayed response. Used during ROC commissioning and post-incident review to validate taxonomy effectiveness.

Cognitive Load Index (CLI)

CLI = (N × log₂(T + 1)) + (C − 3) × 1.5 + (A × 2.0)

Quantitative measure of dashboard-induced cognitive strain; used to validate taxonomy compliance and ergonomic safety.

Variables:
SymbolNameUnitDescription
N Number of visual widgets count All interactive or static display elements (gauges, charts, status lights, text fields)
T Average update interval seconds Mean time between data refreshes across all widgets on the dashboard
C Distinct colors used count Number of unique hues (excluding grayscale), per ISO 11064-4 color-coding guidance
A Animated widgets count Widgets using motion (pulsing, scrolling, rotating) that increase attentional demand
Typical Ranges:
ROC Primary Command Dashboard: 4.2 - 8.7
ROC Diagnostic Deep-Dive Panel: 15.0 - 28.5

💡 Worked Example

Problem: A blast area monitoring dashboard shows: 14 widgets (7 gauges, 4 trend charts, 3 status lights); average update frequency = 2.3 sec; uses 6 distinct colors (including 2 shades of red); 3 widgets use animated pulsing.
1. Step 1: Calculate Base CLI = widget_count × log₂(update_frequency_in_sec + 1) = 14 × log₂(3.3) ≈ 14 × 1.72 = 24.08
2. Step 2: Add Color Penalty = (color_count − 3) × 1.5 = (6 − 3) × 1.5 = 4.5
3. Step 3: Add Animation Penalty = animated_widgets × 2.0 = 3 × 2.0 = 6.0
4. Step 4: Total CLI = 24.08 + 4.5 + 6.0 = 34.58
Answer: The result is 34.58, which exceeds the safe limit of 12. This dashboard violates ISO 11064-4 Clause 7.2.3 (‘visual complexity thresholds’) and requires taxonomy-driven simplification—e.g., collapsing 4 trend charts into 1 multi-series chart and reducing color palette to 4.

🏗️ Real-World Application

At BHP’s South Flank ROC (Pilbara, WA), a taxonomy misalignment caused repeated false 'cable reel overheat' alarms. Sensors reported temperature at 3 locations: drum surface (valid), motor housing (valid), and ambient air near reel (irrelevant). Under the taxonomy, only System-level (drum/motor) and Event-level (temp > 85°C for >15 sec) data were permitted on the primary haul truck dashboard. Ambient readings were demoted to Anomaly-level diagnostic logs—accessible only via drill-down. Post-implementation, false alarms dropped 92%, and mean time to acknowledge valid alarms improved from 8.4 to 2.1 seconds.

📚 References