Blend Compliance Deviation Index (BCDI)

GMI-MIN-003MiningIntegration-Required

Definition

Deviation between dispatched ore blend ratios and target blend specifications, measuring whether autonomous fleet dispatch is meeting metallurgical feed requirements.

The Measurement Gap

Mine planning teams spend significant effort creating hourly and shift-level blend targets that balance crusher feed grade within the narrow range the mill needs for optimal recovery. Dispatch systems then assign trucks to shovels based on an objective function that typically prioritizes shovel utilization and truck cycle efficiency. Legacy metrics track blend compliance at the crusher using head grade sampling, but they report the deviation without attributing the cause. When the crusher feed grade deviates from plan, the geology team is questioned first, the shovel operators second, and the dispatch system last.

Traditional metrics like tons per hour, shovel utilization, and even head grade variance cannot separate three distinct sources of blend deviation: geological variability within the block being mined, shovel operator dig sequence choices, and dispatch sequencing decisions about which truck goes to which shovel and when. Physical Availability, MTBF, and equipment-centric metrics are entirely silent on this question. Even cost per ton, the universal mining metric, treats blend deviations as an invisible cost buried in downstream recovery losses rather than an upstream dispatch decision.

Decision Supported:

Is the dispatch algorithm sacrificing blend quality for shovel utilization, and is the mill paying for it in recovery losses?

Formula

Plain Language: Calculate the root-mean-square deviation between actual crusher feed grade and planned grade across each time interval in a shift. This gives the overall BCDI score. Then, for each interval, compare the grade that was available at the shovel the dispatch system chose against the grade that was needed for blend compliance. The proportion of total grade deviation that traces back to dispatch sequencing (rather than the geology being different from what was expected) is the Dispatch Attribution score.

Notation:

BCDI = sqrt(Sum((Grade_actual_t - Grade_planned_t)^2) / N)

Dispatch Attribution: DA = Sum|Grade_available_at_dispatched_shovel_t - Grade_required_for_blend_t| / Sum|Grade_deviation_total_t|

Variable Definition Unit
Grade_actual_t Actual crusher feed grade at time interval t Grade units (% Cu, g/t Au, from mill sampling)
Grade_planned_t Planned crusher feed grade at time interval t Grade units (from mine plan)
N Number of time intervals in measurement period Count
Grade_available_at_dispatched_shovel_t Grade of material at the shovel the dispatch system selected at time t Grade units (from grade control)
Grade_required_for_blend_t Grade required from that truck assignment to maintain blend compliance Grade units (derived from mine plan)
Grade_deviation_total_t Total absolute grade deviation at time t Grade units
DA Dispatch Attribution, proportion of blend deviation caused by dispatch sequencing Percentage (0-100%)

Proposed formula, pending validation pilots.

At a Glance

Unit
Percentage
Trend Direction
Lower; approaching 0
Reporting Frequency
Per-shift
Category
Autonomous Fleet Intelligence & Dispatch Economics
← Rock Characterization Validation Score (RCVS)Exclusion Zone Proportionality Score (EZPS) →

Assess Your Organization

The AI-GPS Assessment evaluates your AI governance maturity and recommends KPIs like this one for your implementation roadmap.

Take the Free AI-GPS Assessment