I've always thought that we don't do as good a job sharing our learnings (aka "mistakes") as we do our successes. Glad to hear I'm not the only one who thinks that!

What does Hatch plan to do with the results of the survey? Will they be shared in a public forum or will they be used internally only by Hatch? Would be happy to contribute and to help encourage other possible contributors but I would hesitate if the results and the conclusions were not shared and open to scrutiny and comment from our metallurgical community in general. No two metallurgists have exactly the same perception, especially when it comes to a "root cause".

I also have reservations about improving Hatch's marketing pitch. I'll be happy to contribute if it leads to a published document which the community can share. It's been some time since such a review of new projects was last published.

The aim is to produce a paper similar to the one I am sending you, and is not going to be an internal Hatch document. The aim is currently to present at the CIM risk conference in the fall and hopefully have a good discussion. The group working on this with myself started because we were interested in the topic. If you have any more questions please feel free to contact me.

Back in the days when i was young, still am, and working in South Africa, the engineering companies used to actually go out to projects six odd months after start-up and do an audit of what worked and what did not. Strange concept. Unfortunately the growth of litigation and opportunists has resulted in this obviously sensible concept to be swept aside. Nowadays plausible deniability is the name of the game. So I fear you are going to have a hard time getting what would be wonderful information. I have plenty to share, but should it end up in the public domain it would be the end of me as a consultant. I fear that you are going to be hamstrung by similar concerns.

1y

An interesting conversation however as you are discovering....gone are the days when people (engineers & clients) could sit down and review a project that has been completed with a no holds barred approach without fear of legal action or being sued for professional negligence.

As mentioned, the idea of post installation audits was a powerful tool in Southern (Sef) Africa engineering community and people did learn and adopt changes form the audits.

I think your own employer should have a wealth of project audit data from their southern african office that you could use.

I don't think that you will get anyone 'fessing' up with a specific project. Career limiting as pointed out. Modern day projects are 'managed' by people who not as experienced in project development as they were, dare I say, in the 'old' days and driven by immediate goals such as share prices, bonuses/options and a general lack of funding.

Indeed how EPCM companies operate these days has changed significantly, many of them now listed on stock exchanges and affected by the same disease.

While it is excellent practice to find out what went 'right' and 'wrong' 6 months after a project has been commissioned (more about internal education for the "EPCM"), I'd rather have a project basically designed and implemented correctly in the first place. However there are many reasons why a project goes 'wrong' or 'fails' and you will need to be satisfied with more generic data for your paper.

I would have thought that Hatch would have a sufficient information from their own projects and experiences (or is this problem!) on this topic as Ewan has pointed out - rather than having to canvass the general metallurgical community for 'failed' projects undertaken by other EPCM companies.

Big subject and there are plenty of candidates. Murrin, Bulong, Cawse. Ravensthorpe (BHP) and Talvivaara would be a possible short list on the nickel side at least - although some of these projects are over 10 years old now.

One additional issue I have seen is let's call it "emotional attachment" for want of a better description. The inability to "let go" a project when it is truly marginal. This leads to "rose coloured glasses" or selective data usage - best recoveries, best grades, (no talc aka Mt Keith !).

This disease gets worse over time when a company or a project manager has been involved with the project for a long time perhaps years. It affects CEO's right down to project engineers. Maybe we could call it the "Dead Parrot" effect!

1y

1)Australian Magnesium Project. _ Stanwell Qld early 2000s. Project abandoned after spending $600 million

2)Hatches lithium Project for Galaxy in China where a recent plant design failure about 18 months ago killed and injured workers and almost killed Galaxy.

3)Gorro Nickel project in New Caledonia for Vale. Huge project blowout and nameplate has never been met.

There's 3 good ones to take a look at. Greg Sheehan from Hatch in Brisbane should be able to fill you in on details on all 3.

Magnola Magnesium (economics) and Bougainville Copper Project (local population against it). Also, you might want to look at the cash flows of the recent aluminum smelters build in the Middle East.

All ingredients for failure: RMG - Rosia Montana Gold project

MMX Serra Azul Iron Ore concentrator expansion in Brazil reached 40% engineering and 5 % construction then stalled in April 2013 due to owner financial difficulties.

So far all of the comments are on which projects failed, but we need to know why they failed.

There are gate systems that should be used to develop a project and assure that the engineering, costs and risks are in line with the expectations of the stakeholders. I would like to know if any of the aforementioned project failures had quality plans that were developed in the first phase of the project and were the quality plans followed?

Magnola Magnesium cost twice as much as Noranda had anticipated. Then the Chinese producers dumped magnesium into the market and drove the market price down 30% where it held for several years. The plant was dead within 18 months.

Bougainville was a while ago, but the main problem was rebels and local opposition.

And the Middle East smelters were built when aluminum prices were expected to be in the $2200 to $2400 per ton range. Not the $1600 per ton range. While their cash costs are among the lowest in the world the all in (capital costs) give them poor returns. 

No failure list would be complete without Reko Diq!

While following the discussion the short comment of your, in my view, is most comprehensive, even it's too short. Actually Reko Diq was a wonderful project but that has been victimized by incompetence, greed and surely biased intention of international business/political groups.

First Coal: A 10 Million dollar piece of equipment that was under designed for the project, First nations not on board, Micro managed out of head office which built a number of kingdoms in the company. Back biting, undermining authority. I could go on.

Anglo's current Minas Rio project is a classic; insufficient due diligence + fast tracking = huge cost blow-out and project delay.

Ravensthorpe Nickel springs to mind as project with all the ingredients. In the initial early study stages there were well known engineering organizations involved. The estimate of $1200-1400 million in the detailed study blew out to an installed cost of approx. $2600million, production was not achieved with substantial schedule overruns. These serious overruns ultimately lead BHP-Billiton to sell it off at a massive loss.

On the Goro project, the major contributors included project team turnover as the project was stretched and delayed many times (including complete turnover of assigned teams at Engineering firms on the project), significant new technology, and major environmental issues which changed as the project proceeded. Local social issues were significant as well.

I suggest Talvivaara- bioleaching for a complex multi metal deposit was not only novel, but it also appears that there were a number major environmental and regulatory issues that were perhaps not foreseen, impacting production levels drastically. So it’s definitely a "problem project" and I don't think the technology was the MAIN reason for failure.

This got me thinking about a project management process chart I first saw almost 30 years ago that had the following project stages:

1)Enthusiasm

2)Panic

3)Disillusionment

4)Search for the guilty

5)Punishment of the innocent

6)Bonuses for the non-participants

Technology is often used as an excuse for failure and to cover up the real reasons. The salient question about Daniel’s link is that everyone knows this at the start of a major project, and yet there are still a high percentage of failures.

One cause seldom mentioned is that executive compensation is often based on stretch growth targets, and severance packages are both locked in and large. This promotes a 'river boat gambler' style of leadership in up cycles, with executives making very high risk decisions that are not always in the best interests of the shareholders. If they fail, the company suffers and they move onto another company toting a hefty severance package. Am I being too cynical?

I was seeing few projects going straight down even in the stage 1 and 2 everything was looking ok still. I have been seeing one project who once lost the political support was never saved. I was seeing on the other hand one project saved by market even they try to reinvent the wheel, not once but many times. I was seeing another one where human factor was reduced to numbers ... and they failed twice. Closer you look at this project you start to wonder why they have been started ... What I learned is expertise is expensive but deserve it every money spend it on it.

Your comments were insightful and not cynical at all. I have seen it many times in my many years in this business. Your note makes me think that perhaps we should consider adding to the list of factors: people, organizational culture, alignment in incentives, and all those soft elements that drive behaviours and decisions and ultimately make a huge difference in project success and failure. I know this is hard to evaluate and measure but how can it be ignored if we believe it’s a large factor?

I would just like to add that being impactful as an expert is an art especially when so many things are going wrong in a project.

Most projects never include appropriate life of mine planning for closure and effective rehabilitation at feasibility so are effectively under resourced for the life of the operation and their true viability within the geospatial and social landscape is never assessed properly. As a consequence most will eventually fail because important cost factors are omitted. This causes massive legacy issues with no resources or corporate motivation to effect a solution.

How about BHPs Port Headland HBI plant. It was built in 92 or 93, started out at US1.2b or there about, finished at about US$3b and after never meeting nameplate, or coming close it was shut down. As well with its sister plant in Venezuela. Murrin Murrin Nickel was another one. The most famous job that has come close recently is Pascua Lama. There are many reasons for that one. In my opinion the principle reasons are not enough experienced people working on project during FEL1 and FEL2, the HR organisation having too much power inside a company and not wanting to pay for experienced people and more importantly very senior client side project sponsors or resource owners who have little or no experience and then make them in charge. The truth is you can take a mining or metallurgist who might be absolutely brilliant and mining and metallurgy, but he won’t get the same result or make the same decisions as a project management professional with 25 years’ experience working in projects.

Would you get a very experienced orthodontist to give you open heart surgery after 10 mins of reading about the procedure in a text book? Obviously not! So why then do resource companies give a few billion dollars to a mining engineer and expect them to get excellent results with their project after reading a book on project management. The result is the same. Heart failure and fame and loosing lots of money. There are many factors that go into a successful project and in my opinion the largest that does not make the news is the ability of HR policies and practices and procedures that are not project friendly. Now the HR folks will dispute this and will say that those policies are set by management and they are correct. If a resource company does not have separate pay scales for project types and a HR organisation that can handle the massive amounts of hiring that has to take place early on a project for non-mining people, then for the project will never have enough people early enough to be able to nail down the key factors that determine project success.

In my limited experience every time you hear a HR person say “No that person is far too expensive, or we can’t afford this guy with xx years of project experience" Then you are about 12 to 18 months from a disaster and it’s time to sharpen your CV and hit the road before you know what hits the fan. One of the best and most difficult projects, also one of the most remote on the planet that I have heard of had separate pay scales for project personnel and mining/operations, and the HR management and staff had their personal bonuses linked to acceptance rates over 80%, and an annual retention rate over 95%. See if you can get your HR manager to volunteer for that? That project from what I heard was an outstanding success and everyone remembered it well despite the fact it was incredibly remote and had temperatures up to minus 45C. In contrast to another project that I heard of which is currently famous where the rejection rates for Client side positions was 80% and the staffing levels were at about 55% for most of the job and retention rates were about 45%. 

As an outsider to the HBI project I think the major reason for its failure was that it was under the control of BHP Minerals rather than BHP Steel. If it has been run as a Steelworks project it would have been straight forward and the Steel Division had plenty of in house expertise to make that project work. Their track record over 100 years was excellent. Instead it was owned by the Minerals Division whose only experience was in dirt moving projects, dig it up, put it on a train then load it on a ship. They in turn used BHP Engineering to manage the project who again were mainly dirt movers moving iron ore or coal. Big companies with several divisions really need to think hard about what part of their organisation is best suited to manage the project instead of automatically keeping it in house within their own division. Getting project people with experience in the correct industry division is critical. Oil and gas people are fine for oil and gas projects but don't put them in charge of building a magnesium plant! Similarly a HBI plant needed to be run by steelworks people rather than iron ore miners.

I knew some people who were on that job years ago. It was one of the jobs that half the people on Perth worked on but 6 months later on no-one had it on their CV. There are a few jobs around like that.

There are many types of project failure and you need to group them by category. Some projects don't meet the original expectations but somehow are successfully because either the environment changes or they lead their owner company on to bigger and better outcomes. Also many reasons for failure given will be just the symptoms of the failure and not the root cause. Often within projects only one or two people actually know the real situation and have all the facts. The troops can't see the forest because of the trees. Some total failures can become disguised because commodity market conditions might improve or takeover merger activity occurs which is favourable. Sometimes it's the second mouse that gets the cheese!

As an investor in the resources sector, and formerly in project/corporate development of a large mining firm, I have concluded that the industry as a whole has sacrificed standards best practices and failed to audit completed projects and modify future approaches from lessons learned. Ultimately, the liability of project falls on the owners of the project/mining companies - that is, the shareholders - not the Board, not the CEO, and certainly not the EPCM's save for typical contract provisions to do "re-work" of the defective services only.

The investment funds I deploy (i.e. the pension money of retirees, insurance money coming from our policies, family office money who have worked hard for their family fortunes), and I am one of the few niche investors performing due diligence beyond a desk in Wall Street, Bay Street and London City - that money goes to pay handsome salaries and stock options to Management, and in the case of say small-cap mining companies with one development project, the balance of funds are then paid on a cost-reimbursable basis to the likes of Hatch to perform a Scoping Study, Pre-Feasibility Study, Bankable Feasibility Study (and I would argue the term "Bankable" is very deceptive - implying to a non-mining investor that the project MUST be bullet-proof and will definitely be delivered as defined in the BFS), and ultimately EPCM or EPC or E/P/or CM services. Also, does industry go to the lengths of actually defining and agreeing a common standard as to what are "low, medium, and high" quality studies? I say the answer is no, and apologize in advance if indeed there is a common standard to which all mining companies can use when engaging an consultancy to perform a study to low/medium/high quality.

I am more interested in knowing how the mining companies get their resource estimates so wrong, how engineering firms get it so wrong (I would ask the group if 50% is the ball park number representing the number of projects that are not delivered within the expected cost estimate, schedule and scope of work?).

Further, it would be interesting to see how many projects in your database for this report were actually audited and how the lessons learned from those audits are implemented/incorporated into the future development approaches of mining projects.

Lastly, I would ask if CIM or JORC or N-43 101 actually provide a rigorous set of standards and requirements for the studies. The study from a layman's perspective tells me

I do have plenty of case studies, details of original vs final costs, schedule, throughout changes/mine plan production rates (as not all cost increases are a result of failures to perform - sometime extra CAPEX buys you more production or increases recovery = improved project economics).

Happy to provide information and contribute to Hatch's work.

In short, and at risk of being shot down which I would happily welcome (!), the industry has failed to successfully establish and prove delivery strategies, methods, approaches, systems and instill the confidence in the very people that fund them to do what the mining companies and its contractors promised to deliver. The focus of the study, in my view, should be forward-looking based on past project failures and lessons learned, not just a history book of what how when and why did the failures occur. The much more experienced professionals than me, who have contributed in this forum have already written that book in this discussion thread alone.

I think your comments / observations are spot on. We conduct independent reviews on Feasibility Studies and one of the areas we see as constantly lacking is in the delivery strategy / methodology process of selection. We also find that project owners rarely have the tools / processes to accurately assess engineering contractor’s estimates and schedules. I was at a presentation at Mines and Money in Hong Kong recently and they showed the figure to be less than 15% of mining projects achieve the objectives established in the BFS!

There are a number of reasons why this occurs but in our experience it comes from lack of control from the project owner, deviation from the agreed delivery strategy signed off in the BFS, risk being poorly managed and project changes not being approved by the correct DOA.

We have developed our project management solution (PMS) for project owners that creates a control environment and a governance framework for compliance to process and management of risk. Process compliance is critical for managing risk and provides confidence in performance measures. Our PMS is a gated process, linked to a DOA where staged audits provide reports on level of completeness and require a signed Justification for Capital to proceed to the next stage. The PMS is underpinned by principles and rules. Principles are the intention to do something and the rules provide how the process and procedures will be governed. This creates a level of control for project owners and should form part of a contract. As the opportunity progresses compliance scores immediately identify areas of the project that need to be improved or require management intervention.

GRS have successfully deployed our PMS with various clients across the globe on numerous projects, creating measurable improved project performance and delivering the clients business objectives. Success comes through analysing an opportunity correctly, selecting optimum delivery strategies and contractors, having a robust project management framework in place to manage risk and control the outcome of the project and most importantly having experienced and competent management.

Excellent analysis. Here in Nicaragua, the main problem was the intent of implementing an American vision in Nicaraguan soil without knowing and/or respecting the Nicaraguan rules. This led to a break in partnership with myself and my American partner who was on a poorly preplanned budget without knowledge of a real Nicaraguan execution time-frame (taking vacations, employee workdays, government regulation fulfilment time, etc.) which led into frustration by the American side which promised investors a previously committed return/plan of action.

Also, the company tried to implement Merrill Crowe here without knowing chemical and raw materials import law (requirements), and based their profit solely on sampling analysis, commodity market prices, FOB machinery costs, and proven gold reserves.

I would like to follow-up on the BHP HBI project comments. I delayed in hopes that someone involved in the project would weighed in, but as Matthew pointed out, probably nobody wants to admit they worked on it. It is easy to forget in these days of multi-billion dollar disasters, that a +100%, +$1 B overrun in the 1990’s was absolutely unprecedented.

I was doing some business development work in the Pilbara at the time the plant was being built, and our team flew over the construction site to have a look. The material handling system appeared to be an absolute nightmare, with conveyors running every which way, and flows seeming to reverse for no apparent reason. We actually circled the site a few times to try and figure out a method to the madness, but gave up and flew on.

I don’t think involvement by BHP Steel would necessarily have been a positive. I was (unfortunately) involved in a failed HBI plant with steel company involvement, and they could not wrap their heads around how sticky HBI is and what an operational and maintenance headache the briquetting machines can be. HBI/DRI was the flavor of the month back then, to be built anywhere there was a cheap supply of natural gas. Everything was driven by gas pricing.

No-one knows what actually caused a project to failure; it is usually due to a combination of features, usually poor design and poor implementation...which can be put down the quality and experience of the people involved. These days lack of sufficient funds, as implied by other commentators, is probably the fundamental reason why projects are poorly done, coupled with a management with its eye on the share market.

Just commenting on some of the failed projects, unfortunately as an outsider:

* Australian Magnesium Project. Stanwell Qld early 2000s. Like many western magnesium projects at the time, relied on an unproven electrolytic process. Was never going to make it, despite having a superb resource, competing against the thousands of 'back yard' Pidgeon processes in China. Poor understanding of competitors.

* Hatches lithium Project for Galaxy in China. Can't comment on this one...but you will no doubt be highlighting the deficiencies Rachael.

* Gorro Nickel project in New Caledonia for Vale. This was an inherited project - a colleague who reviewed the operation said that it would never work due to the adoption of so many unproven technologies and circuits. I guess he was correct!

* BHP's HBI project. Should have worked. They have an operating plant in Venezuela to base it on but the local BHP project management apparently refused to do so. Came down to simple gas flow and contact requirements...basic chemical engineering principles.

* BHP's Been up mineral sands project south of Perth. A classic - full of clays - a guaranteed problem project before the design even began and widely known in the industry. As generally understood, the upper management were on a career path to directorship....and only 'yes' men were retained on the project. Fortunately only cost BHP $300 million.....

And so it goes on.

The possibility for failure, including cost overruns and design errors, seems to increase with the size, cost and complexity of the project.

Unusual to see small gold leaching and copper flotation projects 'fall over'.

As a final comment, many of these large industrial mineral projects rely on massive infrastructure e.g. railways. It would be interesting to learn about the number of projects that fail through infrastructure 'failure'.

Excellent comments, as far as I know all big mining companies have its White Elephant, as already mentioned the Noranda Magnola Project was one big casualty of not knowing what China was doing. As a result of this, to prevent the go ahead of the so called Pet Projects, Noranda implemented Six Sigma and then eliminated by its new owner. Falconbridge had its pyrrhotite pellet project, at Sudbury, that also failed. Some technologies work in one place (Korea) and not in other like the Lead Smelter built by Teck Cominco an now replaced by Kivcet lead smelter. I am looking forward for the publication of this paper by Hatch. We have to learn from others mistakes.

Barrick's Pasqua Lama is a project for this study. As well as Tia Maria and Conga, Santa Ana in Peru.

Except a few, most of the projects do not achieve the targets as regards production, qualities, schedule, environmental norms, cost etc. Actually most of the projects are not executed as planned and run into various problems. You name any, you will hear the problem.

Having an engineering company update the paper is not unbiased because engineering companies are but one of the reasons for project failures! Having said that it is my belief that the following are main causes of failure.

1 Incompetent Owners Team, at the end of the day the owner’s team is ultimately responsible for project delivery.

2 Too much focus on financial models that make unjustifiable assumptions.

3 Too much of doing the same thing and expect a different result!!

4 Poor equipment selection to reduce capital.

5 Disconnection with landowners\stakeholders

6 Risky technical assumptions to reduce testwork costs.

I could go on and on having been on all sides of the project execution and delivery fence, perhaps a good topic for a book.

The mining industry has been infiltrated by too many non-mining people recently, people who are focused on making short term money and moving on to the next industry whether it be exotic fruits, chocolate, software production, lingerie, tourism etc to them it doesn't matter as long as they see a line to make money!, As a result I believe the mining industry is now entering a regrouping phase largely because of the failures this environment has endured (too bad if you are one of the shareholders in these companies).

Engineering companies don't help they fill the chairs in the office and hand out timesheets to the client for payment instead of sourcing reputable people with solid track record.

A recognition that a mining lease does not empower the mining company to "own the land" especially in undeveloped regions.

Copper Mountain Mine in Princeton, BC should be a consideration.

Specific projects which have not already been mentioned would include:

Tampakan - Philippines

Olympic Dam Expansion-South Australia,

Botswana seems to be right up there with the failures at African Copper, and Discovery Metals, projects.

Brazil weighs in with the Mirabella Nickel Project

Colossus Minerals in Brazil.

These cover a wide size, as well as geographical range, just to name a few.

1y

The Gindalbie magnetite project in West Australia is also worth including, depending on how far back you wish to go, the Port Headland Hot Briquette DR plant could also be included.

The Laterite Nickel Leaching projects - Ravensthorpe, Cawse, Bulong, Minara would also be very interesting!

I work with a team of Metallurgists at IMO ( Independent Metallurgical Operations) and there are a number of different factors influencing a project's success resulting in issues which include :

Initial testwork has not been done

Initial testwork completed inaccurately

Incorrect flowsheet and process option choices

Final flowsheet does not reflect the testwork

Wrong equipment selection

Changes and inconsistency during engineering / construction

Operational surprises

There are so many examples of projects which have experienced a number of concerns to list them all. Our team of metallurgists at IMO, assist clients to work through and resolve issues from the testwork and engineering design phase to commissioning and operation into production. Sometimes issues can arise from plant operators cutting corners, using the wrong blends and quantities of reagents and or feed which will also affect the production and recoveries.

It is very clear even from the limited comments made so far that many projects, dare I say most (?), don't go to plan in some way. I noted a few items in Christine R-Cs comment that hit a chord with my experience a) test work b) final flow sheet did not reflect test work. Test work is not always done on representative samples, and even worse, sometimes not done at all due to time or other constraints listed by the client. Sometimes a dominant client side or design side personality will over-ride key decisions, selections or advice for a variety of reasons and sometimes this results in a poor outcome.

The Hatch update would appear very timely and I would love to see a copy of that report / article when it becomes available.

Indeed, with a success ratio of about 0.25% form grass-root through commercial plant, the industry and investors alike would welcome a change. Certainly, I would be privileged to contribute, please feel free to contact me directly.

I think a major factor which i cannot see shown above is the complex relationship between investor enthusiasm & global commodity prices. The largest comment I have heard historically is "the money isn't available for {xxx} so we're [delaying]". I imagine however this is the crux of the referenced paper.

On the ground in execution phase delay primary drivers are in my opinion all sourced from conflict between the diverse interests of the owner/engineer/operator/equipment supplier over the concept of what "value" and "best practice" actually are. Aligned vision & alignment control strategies are the solution of course.

It is interesting that in much of my experience of large projects in other industries these delays and delay sources are very well managed by proper use of Process Licensor and 3rd party PMC contracts. I have never seen this in M&M to the same extent.

As others mention i look forward to reading the finished product when/if published.

It may be an easier task to identify the projects that were completed on time and within budget, in recent years there have been very few of these. However, you should consider the Boleo Project as a recent example of a project with problems. Concerning reasons for failures, not only could inadequate testwork, flowsheet design and equipment/process selection be factors, but you should also look at inadequate commissioning planning and construction completion criteria. Too often the Owner is pressurized into production before the plant is even close to being ready.

The Sino Iron (Citic Pacific) and Minas Rio (Anglo American) projects should make for interesting case studies. I cannot claim to have any particular insight to these projects, but it would seem, given the cost overruns, that the list of problems would be significant.

The Sino Iron is definitely the one worth of mentioning. I've kept an eye on this project from its very beginning and a few of the major project participants are my close acquaintance.

While the owner mainly attribute the failure to the lack of experience of CM Company in international project, I think there is much more to explore.

1. under-estimated the project risk at the very beginning.

2. The owner is mainly a financial investor, with limited experience on mega-mining project, especially iron ore project.

3. Lack of understanding of the local culture and could not resolve culture conflict problems effectively.

4. The owner did not have an experienced project team that could effectively monitor the progress and resolve issues arising.

5. EPCM did not have experience on major international projects.

6. Problems on equipment supply.

I thought I could be a perfect one to bridge the cultural and project experience gap for the EPCM, but they even did not bother to invite me although my name was mentioned during their engineering work (I did not actively looking for this opportunity for some reasons).

"For the want of a nail the battle was lost". Frequently a single weak component in the chain causes cash flow distress and project loss. It could be failure of a critical 5 hp motor, or inappropriate suite of spares in stock.

Wow, I am going to watch this topic with extreme interest. From a Commissioning perspective, I have been involved in (usually way too late) or watching a few of the failures mentioned above. I have also been involved in a number of very successful projects (i.e. under budget, and ahead of schedule). There are significant reasons, and differences in management style of both.

Finance folks don't necessarily see the reason for commissioning, some almost ignoring this critical part of a project till it’s too late, or just end up skimping on commissioning budgets i.e. run out of money. They forget a $US2M per day process plant equates to lots of $$$ per hour when it’s not running, due to lack of spares, or poor project planning.

Another failed project is The original Tonkin Springs.

Obviously a hot topic judging by the commentary, and a good initiative, well done, this is a really important area and I agree with what people are saying above.

An additional suggestion I have is for you to reflect on the actual post construction review process. In my experience, once plants are completed, the engineers rarely return to site after several months of operations to review the design/build/operation (for reasons like; the original budget and schedule were blown, the relationship deteriorated with the owner towards the end, engineering team burns out, no budget to undertake a follow up review, engineer too busy or don't want the engineer back, personnel have moved on, not interested? And even if there have been learnings, they tend not to be locked in and the mistakes are repeated again in the future.

This assessment of yours is a good example that generally these reviews are undertaken when the engineers are quiet, otherwise as Enzo pointed out, there is a bit of a bums on seats mentality, particularly when it is really busy.

I once read a paper called 'Taking the Sting out of Project Start-up Problems'. It is from the early 80's and it is frightening how little has changed in some ways!

Look at the mining projects in the Fort McMurray, Alberta area. Here there have been many projects which have failed dating as far back as the 1940's (possibly before), and the latest round of failure due to overpriced inputs (i.e., wages) and the market crash within the past 5 or 6 years.

A little expansion on my comment about the "Owner’s Team” having ultimate responsibility for project delivery. My observation in some of the failed projects is that the owner’s team does not have the experience and intimate knowledge of project execution to pull together the numerous engineering houses, testwork laboratories, specialist consultants, financiers, government organisations, non-government organisations, landowners. In some ways this is related to my previous comment about too many non-mining people infiltrating the business recently.

The owner’s teams I speak of concentrate on appointing the "accredited suppliers" and think the job is done! Having covered their behinds if something goes wrong!! Unfortunately this is exactly what the non-mining infiltrators want- give me "brand name" and fancy presentations to convince financiers and innocent shareholders to provide funds, by the time it becomes obvious that things are not going well the perpetrators have fled the scene, leaving behind a mess which is not always easy to retrieve.

A couple of other useful references you might want to look at are Terry McNulty's paper Minimization of Delay's in Plant Start-ups (SME Plant Operators Forum 2004) and R&D Prospects in the mining and metals industry by Filippou and King - both references are readily accessible online or for purchase from Elsevier or SME. Several examples of unsuccessful projects and contributing causes are noted in each.

Some other projects I can think of that have not already been mentioned are Hartley Platinum (Zimbabwe), Syama? Gold (Mali), Beenup Mineral Sands (WA), all of which were undertaken by BHP in the 1990's. (Enzo already listed BHP's HBI project.) These project failures amongst other causes drove a major organisational shakeup of BHP in the late 1990's.

With only a few exceptions, project "fast track" will lead to some kind of project failure.

Getting the concept studies or pre-feasibility front end wrong is definitely a major contributor. On the mining and processing side, as mentioned by others above, failure is contributed by non-representative sample selection for ore body definition, metallurgical and other testing as is failure to understand longer term design variations that might be needed as ore body characteristics change with time.

Inadequate ore body (geo-tech, grade, impurity, metallurgical (throughput, grade, recovery)) characterization and variability test work on bench and pilot or demonstration scales to support appropriate initial mine and plant design, mining and plant process flow sheet and equipment selection can all drive project failure. One wonders how not selecting representative samples gets past the lenders due diligence team if the project gets as far as FEL4 and requires external financing?

Another form of project failure, at least from an operator’s point of view, is a plant layout that looks good on paper but limits ability to upgrade/expand or retrofit new equipment into a plant including installing equipment that has been cut during initial plant design and construction to "save" costs. These cuts might even be as extreme as modifying the flow sheet without considering production, operability or long term expansion flexibility implications of these changes. Other issues might be ability to access, isolate and maintain the plant.

Take a look in South African Anglo Platinum BMR Expansion and Spain Cobre Las Cruces. The both are projects that have huge difference between the promises of the engineering company and the final results on production and costs.

This is a topic of interest to academics as well as we try to train our future mining and metallurgical engineers.

There was some discussion with letters to the editor and articles on this on the Mineweb.com site. A podcast was also done in August. Reference there is made to a paper I gave at the COM meeting in 2012 at Niagara Falls which extends Terry McNulty's analysis of what goes wrong with new technology projects. Paper title is "Red Herrings on the path of technology development - Capital destruction in the pursuit of a bad idea". I will also have a paper at Hydromet 2014 which provides examples of success stories - from 50-70 years ago!

Based on the Mineweb items, I got a considerable amount of feedback for a wide variety of people who had been involved in "technologically challenged" projects, many very good examples. The two things that are really troubling is that (a) most due diligence reviews miss the potential culprits and fatal flaws, and (b) even when they might have caught them, management ignored them and proceeded anyway. There are many examples of failed projects, but it is difficult to get the full story on them, and even so in more projects than should be the team is just kicking the can down the road, hoping (one assumes) that the same problems won't occur again. I call that the "pin ball machine" protocol for technology advancement. Get a big bag for the money you will need to finance this endeavor.

Whilst the McNulty paper is still the best overview to date addressing the general causes of ramp-up failures, actual published data collected during the past 15 years suggest that the root-cause for technical failure (including cost over-runs, slow ramp-ups or complete write-offs) is the reliance on assumptions as basis for process and equipment design criteria. More specifically, the key underlying issues consist of handling and separation difficulties and usually in conjunction with feed variability. Process chemistry and general conditions are relatively easy to optimize in isolation. But this isolation means that the process physics (as a resultant of the process conditions) is generally overlooked, and this is essentially the problem. Mixing the slurry commonly occurs without major hurdle, but pumping, thickening and filtering almost are almost always difficult. And a plant stops working when any single process stream cannot be pumped or separated. Contrary to the general perception, establishing the design criteria package for a certain deposit (new or expansion) is quite straightforward and comes at a fraction of the cost compared to the speculative design approach.

My angle is "why are these assumptions allowed to find their way into the design criteria?" My opinion is that often the engineering company does not have the operational know how to decipher which assumptions are not justifiable, and if you have an owner’s team that believes the engineering company knows it all then you are in real strife!!

A good engineering company will take you where you want to go on time and on budget, unfortunately if you are heading for a cliff they will still take you there on time and on budget!!

The owner’s team has responsibility for developing design criteria, process description, operating philosophy etc. and to scrutinise assumptions made during the design stage to make sure they are not heading for a cliff.

An interesting case – the failure is not so much due to the failure of the technology per se but to unusual circumstances and their impact on equipment performance. Such as very heavy rainfalls throwing the water use and flow out of balance (impacting carbon efficiencies and thickener performance). And such as power outages causing plant trip outs and requiring auxiliary power supplies to be provided in critical areas.

It might be asked why the design team did not include these sorts of potential problems in their design, problems which were not unknown for the area. Unfortunately, this failure may lead to a much shorter project life than originally projected.

Perhaps this is another example of the Red Herring effect – management fixated on the new process and its potential, thereby being diverted from careful design procedures, not creating an objective and thorough design basis for the whole plant and allowing defective (or ignoring) “haz-ops” reviews.

Company projections seem to be that adequate corrections to the plant cannot be made to being the technology into a commercially economic status, shortening its life (and thus profitability). Which says that probably if design considerations had been more thorough and objective up front, the plant would have never been built.

And that is how capital is destroyed by defective technology implementation strategies.

1y

In my experience reliance on assumptions is a major underlying reason for min-met-plant failures and the unwillingness to challenge them. In my practice, about 90% of the metallurgical engineering data requests come with predefined targets. Many of them with full Metsim models, and yes, with near-frozen flowsheets. Now, what is coming out once all real data are generated? Well, sizing corrections on all handling equipment, pumps, thickeners, filters, pipes, drives, power draw, and lifters. The only items requiring little or minor changes consist of straight-agitated mixing tanks, reactors, cells, etc. So the general assumption is that once the feed is ground and fed into the first circuit, everything else will work by the book. No, it will not - the cart is already running miles ahead of the horse. Regardless of the under-performing plant/scenario I look at, I always stumble into a sequence of wrong assumptions in chain-reaction. No exceptions, really...and then the escalations start...based on new assumptions of course. This a proven mechanism for the vicious circle; the no return-money-pit.

Following is an example of how a project with a bad technology choice falls apart. The company in question is ZincOx, the technology is a rotary hearth furnace to process EAF dust, and the plant is located in S. Korea. The author has critiqued this technology and this project earlier, most recently in a paper, Red Herrings, in the proceedings of the 2012 MetSoc meeting at Niagara Falls. Another paper appeared in the Indian journal SteelTech for January 2012. These papers predicted the eventual failure of the project, pretty much following the technical problems they are experiencing now: refractory damage, hearth accretions, off gas handling problems, inability to reach design throughput, low heat transfer to hearth, and zinc in residues.

We will be looking at the 2013 annual report, issued June 3, and other news releases from Korea and England. They began the project by calling the RHF revolutionary and a game changer, ignoring several prior failures by others. They also chose to operate this hazardous waste processor without charging a treatment fee, which indeed had never been done. They have since found out a great deal about both of those claims.

Reading this annual report, as well as press releases and fawning accounts from financial advisors and stock market publications, one finds much that is not said, and that which is rarely contains substantive data. However, this report contains a number of “mentionable” which should have been seen as a challenge to any wistful attitude toward the future of the project:

(1) The stock price has fallen precipitously since its entry into the S. Korean business at the beginning of 2012, from 80p down to 12p, indicative of what the larger investment world thinks of their chances,

(2) Returns for the first two year have all been losses, which were roughly equal to revenues ($27.2 million revenues, $26.3 million losses for 2013), which means expenses have been twice revenues, even though dust processed has risen from 30% of design (200,000 tpy) to 50% in 2013,

(3) One of the largest single line items on expenses is Administrative Costs, about 40% of 2013 revenues and 100% of 2012,

(4) Zinc price realized is 55% of LME price (Korea Zinc is the buyer),

(5) Why so low? – it seems more than a smelting charge or shipping,

(6) They now must pay for disposal of the contracted dust they cannot treat (which last year was half of the total capacity); but costs for this item were only $1.8 million, maybe 20% of what might be expected, which makes one think #5 and #6 may be interrelated – i.e., does difference represent garnishment of zinc revenues by the mills to pay part of the landfill costs?,

(7) The process rejects iron, which ZincOx considered a “high value iron product” and was originally stated as contributing significantly to making the operation profitable. It cannot be sold, so must be landfilled, another $1.7 million expense, and no buyer has been identified, just “maybe’s”,

(8) Their main competitor, Befesa, with the world standard waelz kiln technology, is right next door, and they have plenty of dust to fill their 105,000 tpy unit, and have been filling it since the first of the year,

(9) ZOX had to give up the second phase dust contracts (the project had two phases, KRP-1 and KRP-2) – since they passed the deadline for the second plant to be on stream last October, with no reasonable outlook of being able to build the second plant anytime soon,

(10) There is a large gap in information and explanations needed (the “unmentionables”?) to realistically support a positive outlook of ever reaching even close to design levels.

(11) The literature reports no similar plant has ever reached RHF design performance ZincOx claims to be obtainable,

(12) Their optimistic outlook remains the same, in spite of the continual setbacks (talking the walk, but never walking the talk, one or two good months always followed by one or two bad months, many breakdowns and equipment failures, signs of systemic problems)

(13) Korea Zinc made them a “Development Loan” of $15 million, which is due in Feb. 2015; current projections are that they will not have the cash to repay it, and no viable action plan yet identified,

(14) Cash reserves have been dwindling by 50% per year for three years in a row,

(15) Each fixit shut down has three impacts: repair costs, lost zinc revenues, and paying to send that dust to landfill,

(16) The many shutdowns and repairs have required several new loans and stock sales,

(17) Application for a large loan from Standard Chartered Bank had to be terminated a year ago since they could not meet bank criteria: operate for three months in a row with adequate cash flow (maximum period was two months, but still low production and negative cash flow),

(18) Properties and equipment for earlier attempted dust processing projects in Yemen, Turkey and the US had to be sold off,

(19) ZincOx states that their zinc oxide product can be upgraded to Rubber Grade Product, and a plant was built in Belgium; that is now up for sale (all the oxide from KRP-1 was contracted to KZ, so could not have even run RGP plant until KRP-2 was on-stream),

(20) Obviously their business model is flawed: the technology was not ready for prime time, they had much to learn about the process (and did not learn from earlier projects), and that a good reason exists for a tipping fee,

Still only bits and pieces of news, primarily only the positive type - last month a record for zinc sales, but what about the other months of 2014?

It is fascinating how credulous even long-time investors are towards an operation that has never delivered on its promises. For me, a better use of the money would be to let my grandkids use it to toast marshmallows. (Or better yet, send them to University).

Here is the final paper https://www.911Metallurgist.com/C/GEO/Top%20Reasons%20for%20Project%20Failures%20and%20Examples.pdf

Thanks to everyone.

Hi Carl

I came across this very interesting information trail on your website. I tried to use the link to access the "final paper" but it is not activated. Is it possible to get a link to the final paper that was put together from this input? 

kindest regards

Ian Benning, Pr.Eng. Johannesburg