top of page
Image by SJ Objio
Title: Enhancing Operational Excellence and Sustainability through Advanced Analytics: Value Driver Tree Application
Organization

Newmont - IBM

Period

July 2023 - January 2024

Status

Completed

Est. Budget

USD 1.2M

Est. Business Benefit

USD 10.0M

Indicative Value-to-cost

8.3x

Keywords

Value Driver Tree, operational excellence, scenario analysis, KPI governance, explainable AI, sustainability analytics

Benefit basis

Estimated 12-24 month value from improved prioritisation of production, cost, recovery, energy, and sustainability initiatives across selected operations.

Business Challenges 
Newmont required a practical way to connect operational variables, financial outcomes, and sustainability commitments across a complex global mining portfolio. Operational performance was affected by ore-grade variability, mine and plant constraints, changing market conditions, supply-chain complexity, workforce variation, and regulatory expectations. The business challenge was not only to report performance, but to explain what operational drivers were creating or destroying value and to support faster decisions under uncertainty.
Expanded Technical Solution
The completed solution was a mining-specific Value Driver Tree (VDT) application that decomposed enterprise value into linked operational, technical, financial, and sustainability drivers. The model mapped production, recovery, equipment availability, energy intensity, processing cost, reagent consumption, maintenance performance, emissions, and product value into a transparent decision tree. Machine-learning and statistical layers were used to identify sensitivity, nonlinear relationships, bottlenecks, and driver contribution. Scenario-analysis capability allowed users to test changes in throughput, ore grade, recovery, cost, energy, and emissions before decisions were implemented in the field.
Technical Work Packages/Methods
  • Developed a hierarchical driver-tree model linking operational KPIs to EBITDA, unit cost, production volume, energy intensity, and sustainability performance.
  • Integrated operational historian data, production reporting, maintenance data, laboratory/assay data, planning data, and finance reference tables into a governed analytical layer.
  • Implemented driver attribution using regression, correlation, sensitivity analysis, explainable-ML methods, and scenario simulation.
  • Designed user workflows for executives, operational excellence teams, plant managers, and site improvement teams.
  • Created dashboard views for current value leakage, driver ranking, bottleneck prioritization, scenario comparison, and estimated improvement opportunity.
Key Deliverables
  • Validated VDT taxonomy for mining operations and sustainability performance.
  • Analytical data model and KPI dictionary for value-driver calculations.
  • Interactive decision-support prototype/dashboard and scenario-analysis framework.
  • Governance model for maintaining calculations, assumptions, and ownership of business drivers.
Results and Value Created
The project produced a practical bridge between operational data and executive value conversations. It improved transparency of value leakage, strengthened prioritisation of operational-excellence initiatives, and provided a repeatable analytical framework for evaluating cost, throughput, recovery, and sustainability scenarios. The estimated benefit reflects avoided value leakage, improved prioritisation of production and cost initiatives, and faster executive decision-making across selected assets.
bottom of page