“Project Management for Mining: A Practical Guide”

What makes project management for mining different?

Mining project management differs because the project’s viability depends on variables that a standard project plan does not track: ore grade, recovery, strip ratio, permit conditions, and the price of a commodity you do not control. A software project can absorb a 10% scope change. A mine can be rendered uneconomic by a 10% change in head grade.

The second is study-gate discipline. Scoping, pre-feasibility, feasibility, and execution are decision points where capital is released, each with a defined estimate accuracy and technical assurance. Skipping a gate does not save time; the uncertainty resurfaces during construction, when it costs far more.

The third is physical. People, equipment, fuel, and consumables must reach a site that may be 300 km from a paved road, at altitude, with a four-month sealift window. A day of weather can cost a week of production, and camp beds constrain headcount in a way office desks never do.

Which project phases must a mining plan cover?

A mining project should be planned across five phases: scoping, pre-feasibility, feasibility, execution, and commissioning/ramp-up. Each phase ends at a gate with a defined decision, a defined estimate class, and a defined set of technical studies.

The table below maps phases to their purpose, indicative duration, and the AACE estimate class that should accompany the gate. Durations are indicative and vary with scale, jurisdiction, and commodity; they are useful for sequencing, not for promises.

Two rules follow. The study budget should be proportional to capital at risk; a $600 million project spending $4 million on feasibility is under-studying the decision. And a gate should not open until the study matches its class: a Class 4 estimate presented as Class 3 is how projects get approved with a 30% cost gap nobody owns.

How do you plan around permitting and social license?

Permitting and social license belong on the critical path, not in a parallel approvals column. In most jurisdictions the permitting sequence is longer than the construction sequence, and it cannot be compressed by adding contractors.

Build a permit register with four fields per approval: legal instrument, issuing authority, required studies, and demonstrated duration from comparable projects. Then link each permit to the construction activity it gates. The result is usually uncomfortable: the water permit, tailings approval, and power connection gate different parts of the schedule, and the latest sets first ore.

Social license is harder to schedule but equally real. Consent obligations, land access agreements, and grievance mechanisms are conditions that can suspend a project, not public relations. Treat each agreement as a deliverable with an owner, date, and definition of done, and track community commitments as rigorously as engineering ones. The IEA notes that social metrics, including worker safety, have improved more slowly than environmental indicators across major miners, which suggests persistent under-investment.

A scenario: a $450 million project with a 30-month permitting sequence and 24 months of construction. The permit is the critical path, so a six-month permit slip is a six-month project slip. Crashing construction by three months might cost a 12% premium on $300 million of remaining spend, or $36 million, to recover half the delay. Spending $2 million earlier on baseline studies is cheaper insurance, and it is the decision most often deferred.

How do you manage a remote site without losing weeks?

Remote-site management is a logistics discipline with a schedule attached. The plan must cover transport modes and their seasonal limits, camp capacity and roster rotations, fuel and reagent supply, spare-parts strategy, and communications.

Start with camp capacity, which constrains everything else. If the camp holds 600 beds and the plan peaks at 780 workers, the plan is not executable. Reschedule the peak, increase off-site fabrication, or expand the camp; each has a cost and lead time. Discovering this during peak construction leaves two contractors competing for beds while both charge standby.

Then model the supply chain by season. A site served by a seasonal road may have four months to move heavy equipment and a year of consumables, which makes warehouse capacity, working capital, and fuel storage project decisions rather than procurement details.

Here is the arithmetic that makes the constraint concrete. Eight haul trucks with a 45-minute cycle, 22 productive hours per day, and a 90-tonne payload move about 17,950 tonnes per day at 85% availability. The calculation: 1,320 available minutes divided by 45 gives 29.3 cycles per truck, times 90 tonnes is 2,640 tonnes, times eight trucks is 21,120 tonnes, times 0.85 is 17,952 tonnes.

Now cut the cycle to 41 minutes through better dispatch and road maintenance. Cycles rise to 32.2, and daily tonnage rises to about 19,700, a 9.7% gain with no new equipment. On a project moving 6 million tonnes a year, that is roughly 580,000 additional tonnes of capacity, which is often cheaper than a ninth truck and its operator, fuel, and maintenance.

How should safety be organized inside the project plan?

Safety in a mining project should be organized as critical control management for fatal risks, with verification activities scheduled as work packages. The ICMM approach is the reference most major miners use: identify the material unwanted events, define the critical controls that prevent or mitigate each one, assign accountability, and verify that the controls work.

The project implication: safety work is not toolbox talks layered on the schedule. It is a set of deliverables with dates, including ground-control assessments before development, ventilation commissioning before underground work, isolation systems before energization, and emergency readiness before the first blast.

Three practices separate projects that manage this well from those that do not:

• Critical controls appear in the schedule. Without an owner and a verification date, a control is deferred the moment the schedule tightens.
• Contractors use the same control set. One standard prevents the “two rulebooks” problem between an EPCM team and the owner’s operations group.
• Leading indicators are reviewed weekly. Verified checks and corrective-action close-out rates say more about next month’s risk than last month’s injury rate.

Safety also affects schedule realism. Planning 12-hour shifts for 28 consecutive days in a remote camp plans for turnover, fatigue incidents, and a workforce that does not return after the first break. Roster design is a project decision.

How do you control cost and schedule on a mining capital project?

Control cost and schedule with earned value against a control-budget baseline, measured physically rather than financially and reviewed often enough to act. The baseline must be the Class 2 or Class 1 estimate, not the feasibility estimate.

Earned value gives you three numbers that matter: cost performance index, schedule performance index, and the estimate at completion. On mining projects, the estimate at completion is the number executives actually need, because it converts today’s variance into a forecast. A cost performance index of 0.94 on a $450 million project at 40% complete implies a forecast overrun of roughly $16 million if the trend continues, which is a conversation worth having early.

Progress measurement is the weak point. Counting “percent complete” from invoices measures spending, not physical progress. Measure installed quantities instead: cubic meters of earthworks, tonnes of steel erected, meters of development advance, commissioning checklists signed off. Each has a defensible unit rate and together they produce a curve comparable across contractors.

Contingency and escalation need explicit treatment. Commodity-linked costs, currency, and labor agreements can move a budget more than controllable variance. Separate the base estimate, contingency for identified risk, escalation for market movement, and a management reserve held outside the project. Blending them hides which risk you actually carry.

A delay scenario shows why. A $480 million project delayed nine months at 70% complete carries about $8.1 million in escalation (6% per year on the remaining $180 million), $9.9 million in owner standby ($1.1 million per month), and $7.5 million in financing (5% on $200 million for nine months). That is roughly $25.5 million, or 5.3% of capex, before lost revenue.

What does operational readiness and ramp-up require?

Operational readiness means the mine can run at nameplate when construction ends: people trained, procedures written, spares on site, systems commissioned, and an organization ready to own the asset. Ramp-up is the climb to nameplate, and it belongs in the project plan with its own budget.

Three gaps cause most of the pain: staffing, where operators are hired too late to finish training; spares, where one critical pump without a backup stalls commissioning for weeks; and data, where a missing maintenance system or grade-control model leaves the team unable to measure ramp-up.

Plan ramp-up as a curve, not a date. Commissioning moves from mechanical completion to cold commissioning, hot commissioning, first ore, then a production climb over months as availability improves. Each stage needs exit criteria; declaring success at first ore while availability sits at 55% is how projects look complete and perform badly.

Grade control deserves a place in readiness planning because it determines revenue from day one. Dilution is the quiet killer: a 2.4 g/t reserve diluted by 18% waste at 0.4 g/t yields a head grade of about 2.04 g/t, a 15% reduction. At 8,000 tonnes per day, 365 days, and 91% recovery, that is roughly 174,000 ounces instead of 205,000, a difference of about 31,000 ounces in the first year. At $2,400 per ounce, the gap is roughly $74 million, which is larger than most project contingency lines.

How should contractors and the EPCM model be structured?

Choose the contracting model based on design certainty at the gate, not fashion. EPCM suits owners who want design control and pay for services; EPC suits mature scope where risk transfer is worth the premium; multiple smaller contracts suit local-content requirements and faster mobilization.

Whichever model you choose, the interface register keeps it working. Every physical and informational interface between contracts needs an owner, a date, and a rule for resolving conflicts. On a typical mine, delays concentrate between the plant, tailings facility, power supply, and water management, because each is delivered by a different party with a different incentive.

What are the most common mistakes in mining project management?

The same six mistakes appear across commodities and jurisdictions.

1. Treating the feasibility estimate as a budget. It is a decision tool with a stated accuracy range, not a control baseline.
2. Permitting in parallel with everything. Permits gate construction; if they are not on the critical path, the schedule is fiction.
3. Ignoring camp and logistics until mobilization. Beds, fuel storage, and sealift windows are hard limits that cost far more to fix late.
4. Measuring progress by spend. Invoices show what was paid, not what was built.
5. Deferring operational readiness. Training, spares, and systems take longer than expected and flatten the ramp-up curve.
6. Separating safety from the schedule. Unscheduled critical controls are the first thing sacrificed when a project falls behind.

Which tools fit mining project management?

Mining project controls traditionally live in Primavera P6 for scheduling and spreadsheets for everything else. P6 remains the standard on large capital projects because it handles resource-loaded schedules and complex calendars, but it is expensive and needs trained planners. Microsoft Project suits smaller studies and owner’s-team coordination. Power BI usually handles reporting, and an ERP such as SAP handles cost and procurement.

The gap is the middle: the weekly execution layer where permit actions, contractor interfaces, control verifications, and readiness deliverables need named owners. That work lands in spreadsheets because the scheduling tool is too heavy and the ERP too rigid.

Doitify, which we build, targets that layer: multi-level tasks and sub-tasks, checklists for control verification, WBS dependencies, Gantt and calendar views, risk and constraint registers, meeting notes, and workload reporting in one workspace, with a Copilot that drafts a work breakdown from an objective. It is not a replacement for P6 on a $1 billion schedule, and we do not position it as one. It is a practical home for detail that otherwise lives in twelve spreadsheets.

Conclusion

Project management for mining succeeds when the plan respects five realities: studies are decision gates with accuracy requirements, permits sit on the critical path, logistics constrain the schedule, safety is a set of scheduled controls, and ramp-up is part of the project. Generic templates cover none of these, which is why mining projects often look well managed until the first gate review.

Start with the phase table, build a permit register linked to construction, model camp and logistics limits before mobilization, and measure progress in installed quantities. If the plan survives those four tests, the rest is execution discipline, which is easier to buy than a second chance at feasibility.

If this post on project management for mining was helpful, you might also enjoy How to Consolidate Your Project Management Tools (2026 Guide) and How to Replace Spreadsheets for Project Management.