Informal checks and balances have been replaced by reporting standards such as NI 43-101, but mine developments continue to have a nasty habit of not working out as planned.

Project developments can miss for social (Tia Maria, Conga, Cerro Quilish) or technical reasons (Sierra Gorda, Rainy River, San Ramon), and this article hones in on the latter, given mining is something mine developers should know a thing or two about.

Failures happen for many reasons and there is always risk associated with developing a mine. To use former US defence secretary Donald Rumsfeld’s oft-used phrase, it is the "unknown unknowns" that often defeat developers.

“Problems usually start with grade, then metallurgy and then capital costs,” John O’Connell, a mining engineer and president and CEO of Victoria Gold, a company looking to build a gold mine in Yukon, Canada, told Mining Magazine.

Despite the use of Rumsfeld’s words, few failures occur for purely unforeseen and unknowable reasons.

Looking at projects through the lens of ‘Thinking, Fast and Slow’, a book written by Nobel laureate Daniel Kahneman about decision-making is useful when looking at the factors that can result in development failures.

Work quality

When a project runs into trouble, it is not long before people claim not enough work was done on the geology, metallurgy or understanding of ground conditions.

A Toronto-based mining analyst told MM: “Geology is not an exact science, so companies make their best estimation of what they have and how to build it. A lot of the time there is a lack of adequate understanding of the mineralised system, which could be because they have not done enough work, or there has been a misinterpretation of the geology.”

Paul Barbaro, business manager at a process plant engineering consultancy in Chile, said not enough attention is given to test work and so projects are developed with insufficient technical information to make key decisions; in other words, garbage in, garbage out.

“A bit of testing is done at the prefeasibility study and some companies leave it at that. We see clients, who may be mining engineers, focus on getting rock out of the ground and ignoring the fact you then have to get mineral out of the rock,” he said.

Barbaro argues the approach to testwork is often superficial or one dimensional, failing to dig into the possibilities that may be encountered via variability testing, with decisions based on relatively few data points.

“Some people use composite samples for testing but the variation is so huge that they design a plant at one end of the spectrum or the other. Ninety-two percent recovery on one piece of rock in a laboratory in ideal conditions may be 88% recovery in real conditions. That 4% can be huge and kill a project,” he said.

Worse, companies sometimes skip a prefeasibility (PFS) or feasibility study (FS) altogether. Rubicon Minerals at Phoenix (Canada) and Colossus Minerals at Serro Pelada (Brazil) chose this route.

For Barbaro, this is fatal: “There is not enough emphasis on designing a plant robust enough for the majority of operating conditions and clients are too happy to jump from a PFS to build, and miss the step of doing more testing. A PFS gets you to a point where you have ideas of where you might end up. A FS means everything is feasible,” he said.

For O’Connell, who has been called in to rescue various projects throughout his career, the more metallurgical testwork the better. “Bulk sample if you can. [Victoria Gold] spent an extra US$20 million on drilling PQ holes to get material for metallurgical testwork. We even tested in a refrigerator to mimic cold conditions,” he said.

Company size invariably plays a role in this as smaller juniors don’t have the cash to undertake the same level of testing as larger companies do.

Barbaro said tier-one miners have very strict guidelines and checklists to make sure nothing is missed, whereas "juniors are prepared to take short cuts".

Red Eagle Mining recently encountered problems with its San Ramon mine, which led to the company stopping operations to change the mining method and implement a paste backfill following falling gold recoveries.

“If we had of had unlimited capital we would have put the decline in to have tested mining underground before doing the feasibility study," CEO Ian Slater told MM.

Risk profile

Approaches to risk differ, and company size impacts how many people are involved in decision making, from technical teams of large multinationals that draw on a wealth of experience, to the comparatively limited knowledge of single-project companies.

Owners need to understand if their risk is in the underground mine, the orebody or the process plant, according to Barbaro.

"Tier-one majors put a lot of money to get from PFS to FS as their risk profile is very low. Smaller companies cannot take this approach and often have a CEO who listens to a smaller suite of experts. Unfortunately, the head technical person – with whom they may have a long relationship – may have only worked on a handful of deposits in their careers and they might be wrong having not seen enough.”

At its most cavalier, the development approach of juniors is little more than rolling the dice and hoping that any issues can be worked out en route, according to Royal Road Minerals CEO Tim Coughlin.

“This is Russian style: figuring it out once you get underground,” he told MM.

Barbaro sees the solution in companies obtaining independent risk analysis to determine what the riskiest aspects are and whether sufficient work has been done to handle them.

“When I read that companies experience unforeseen problems, it sounds counter intuitive as they should know the risks they are likely to run into,” he said.

Consultant culture

The use of consultants is widespread in the industry as they bring expert knowledge and opinion for a temporary, if expensive, period.

However, their clients often guide where they want to the consultant to get to with a study driven by considerations such as a target capital expenditure figure or production rate. This pressures them to give optimistic outcomes.

Don East, director of mining at GMI consultants in Peru, said: “Very few consultants will put their heads on the block, disagree with the owner and say that a decision is wrong. Consultants don’t stand up vociferously against things that clients want to do that they don’t believe are in their client’s best interests, which is something they should do.”

Barbaro added that this sort of interaction with clients can prove very tricky.

"They ask you to put in a lower energy cost when everyone knows it is $0.20/kWh, for example. Grinding consumes a lot of energy, so this figure needs to be accurate. Consultants have to justify what they say. NI 43-101 is very strict on that, so you need to put some rigour into estimations,” he said.

Consultant quality and experience is also a factor, with juniors again finding it hard to attract the best talent.

The consultants carrying out the mine design should be experts in that field, according to the mining analyst. "You want the A team for the kind of deposit you have and mine that you will build."

The fact many of them are not is compounded further by company management that lacks technical knowledge and experience.
“Many consultants have never built and operated a mine. Many CEOs are lawyers and accountants and, in many cases, cannot tell what is the best advice,” he said.

Contingency

All these strands come together in the approach to contingencies.

Project CAPEX estimates usually include contingencies for cost variances during the build phase, but they do not exist to cover major errors in project design that can result in failure.

As Barbaro said, some companies consider contingencies as the things they know they will have to spend, but haven’t done enough work to price in yet. Others use contingencies to cover all the unknowns by adding a percentage onto the total cost.

"As we build plants, we try to be very accurate with our costing as we want to understand a contingency and where the risk lies. If you don’t put thought into the contingency, you are not paying enough attention to areas where you can lose money,” he said.
Contingencies vary for different project elements and reflect the varied risks.

For the plant works and equipment where there are good estimates from suppliers, cost estimation accuracy should be within 10-15%. Large earthworks such as tailings storage facilities, major pre-stripping, tunnelling and other infrastructure is much more variable, with accuracy to within 25-35%. It can be the major earthworks where things go wrong like at New Gold’s Rainy River project.

O’Connell said certain costs are easy to estimate but earthworks and geotechnical works generally aren’t. "On earthworks, we have a 35% contingency. Mobile equipment is just 5%,” he said.

Projects that are not well defined require greater levels of contingency. As contingencies appear as a dollar value that project developers must raise through debt or equity, management often looks to pare them back to ensure a development is financeable.

This opens a developer to experiencing cost overruns down the line, according to East.

Management and owners

Management ultimately makes mine build decisions for financial reasons, which can make them ignore those who voice concern. This results in a planning fallacy whereby forecasts are unrealistically close to best-case scenarios without anticipating the "unknown unknowns".

For instance, the building of a mill; the mining analyst said companies often work backwards from the production they would "like to have", build a mill that can produce this and try and get the orebody to deliver.

"When the mill is too big, all the money goes into keeping the mill fed,” he said.

Personalities also play a role, said East, and is one of the reasons why consultants can have a muted voice. “When management is stubborn and opinionated, it is very difficult for them to be open minded about getting the best solution,” he said.

Finance

Competition for capital sees a focus on the financial estimates as companies seek to benchmark their PFS and FS financial estimates against peers to obtain favourable financial comparables. And scenarios are often moulded to the most favourable scenario to get better numbers.

The vice president of exploration for a growing mid-tier gold producer told MM that 80% of the work he reviews in project studies is “rubbish".

"It is not good enough quality for our exploration team to take an internal project to our development team,” he said.
While study information is in the public domain and collated by services such as SNL Metals and Mining, crucially, data on true project development costs and their variance from FS estimates are not collected. This means there is no real reference for forecasting.

Ken Kluksdahl, the former head of AngloGold Ashanti in Colombia, said classic benchmarking with the PFS and FS of other projects doesn’t take into account site specific issues such as productivity rates, labour law, local skills sets and the true development capability of the EPCM contractor.

"Companies also miss on things like steel and cement price estimates as most engineering companies don’t have very good estimating skills,” he said.

Improvements

To ensure risks are fully considered prior to execution, Gary Klein suggested in a Harvard Business Review article that companies undertake a pre-mortem before making key investment decisions such as building a mine.

This sees a group that is knowledgeable about the decision imagine they are a year or two in the future after the plan has been implemented and was a disaster. The pre-mortem asks them to write a brief history of why it was a disaster and is a tool to overcome groupthink and legitimise doubts.

Unfortunately, mining is a very conservative sector and one not best disposed to talk about its misses.

Chile has taken a step into the brave new world of honest and open reflection about past failures with the inauguration of ‘Fuckup Nights’ organised by the Alta Ley (high grade) initiative headed by Mauro Valdez, the former vice president of corporate affairs for BHP Billiton’s base metals business in Chile, and Alejandra Wood, director of copper study group Cesco (who worked with Valdez at Escondida).

The first night, in June, saw 150 people hear three speakers from the mining sector talk about their failures and obstacles to overcome to achieve success.

“In an industry that is adverse to risk and very conservative like mining, it is necessary to shake people up a little and challenge them," Valdez told MM.

"The mining sector tends to be very prudent but innovation has to embrace risk and the possibility of failure to progress.”