A strategy of continuous improvement that maximises productivity without large capital investments
‘Productivity’ and ‘continuous improvement’ have had a raised profile in mining recently. It seems as if all the major players are either chasing it or talking about it. It sounds simple, but finding a sustainable and systematic way of achieving and driving continuous improvement has proven difficult to achieve in practice. Could a methodology that has been very successful in other industries, such as the ‘theory of constraints’ (TOC), be the key to unlocking success in this very important area? This article explores the potential application, benefits and roadblocks to the acceptance of TOC in the mining industry.
What is the theory of constraints?
The theory of constraints is a broad continuous improvement methodology most widely adopted by the manufacturing industry (Goldratt and Cox, 1984). TOC can be used to drive improvement in both strategic and tactical planning. It is based on the premise that there is one key thing (a constraint or bottleneck) that is controlling the rate at which profits are generated. The terminology is normally viewed in a production line metaphor. This might be a particular machine, the availability of a certain part or a specific operating policy. The theory asserts that addressing this constraint first will deliver the most improvement to the operation in terms of profitability. Conversely, improvement projects based around non-constraint areas are unlikely to provide much, if any, improvement.
Unlike other more well-known continuous improvement methodologies such as Six Sigma and Lean, TOC focuses on:
- The entire system, putting the most effort in determining what to change rather than how to change. There are a number of examples where a hybrid TOC, Lean and Six Sigma approach has been applied successfully in industry, where TOC is used to define what to change and Lean and Six Sigma are used to determine how to change and monitor it.
- Increasing throughput and leveraging fixed costs, rather than cost-cutting or minimising variability. TOC states that there is an upper bound to cost-cutting and that when it is breached, further cuts may reduce productivity. This consequently reduces profit and is therefore seldom sustainable.
- The change that will make the most positive difference, rather than making lots of small changes.
- Exhausting the least expensive improvement options before considering expensive options.
- Using buffers to deal with uncertainty, rather than planning for perfection.
- Applying the theory of constraints
The application of TOC uses five focusing steps:
Identify the system’s constraint. This constraint is what is stopping the business from generating cash at a faster rate. This could be physical capacity (eg a particular piece of equipment) or a policy. For most businesses that have not applied TOC, policy constraints are the most common. A key symptom of a policy constraint is the so-called ‘wandering constraint’, where it appears that a number of physical constraints exist. Policy constraints must be addressed before physical constraints.
Exploit the constraint. This step requires all effort to be focused on getting the most out of the constraint. For example, the constraint in a manufacturing process may be a particular machine. The machine can be exploited by eliminating operator breaks through hot-seating, offloading ancillary operator duties not associated with the constraint, finding alternative process routes for some products or prioritising the highest-margin products (per machine hour).
Subordinate everything else to the constraint. In order to fully exploit the constraint, a number of other processes need to be reoriented to ensure that no time is lost at the constraint. If the machine is operating for more time, there may be a need to increase orders for parts or improve the efficiency of upstream or downstream processes.
Elevate the constraint. Only after all other options have been exhausted should additional capacity be purchased. In the previous example, this would be the purchase of an additional machine.
When the constraint has been broken, return to Step 1. Note that this may happen before Step 4.
Stay up-to-date with all the latest news and analysis by joining minerals industry professionals in the AusIMM LinkedIn community.
Wait a minute…mining is different
Just like TOC, Lean and Six Sigma were adapted from a manufacturing context. These methodologies can be applied to a mining context because a mining operation can be viewed as an abstract manufacturing and distribution system. There are raw materials (rock) that flow through various tasks (eg develop, drill, blast, load, haul, crush, grind, leach) to produce a product. Before we are chastised for this apparent oversimplification, we acknowledge that mining and processing is different in that the raw materials can be highly variable and are finite. However, the goal remains the same: to generate the most profits now and into the future. Therefore, the concepts and methods of TOC are almost directly transferrable to mining projects.
TOC has been used successfully in mining by the authors. We will demonstrate the potential applicability and impact of TOC for mining with two diverse real-life examples. The first case study shows the application of productivity improvement within a large operating underground mine. The second case study identifies optimal project strategy at the design stage.
Case study one – improving productivity for a large underground mine
We were asked to analyse and make recommendations to improve the productivity of a large underground mine operating in conjunction with a large open pit mine. The underground ore is of a substantially higher grade and delivers a higher margin than the open pit ore. As a result, the operation’s preference was to process as much of the underground ore as possible. This put a lot of pressure on the underground operation to maximise ore production. Therefore, the underground mine’s capacity to produce ore is considered the constraint of the operation and, if improved, would improve the overall profitability.
As is often the case in underground mining, the complex interaction of processes meant that the inspection of the mining operation found no immediately obvious constraint. In this instance, it is useful to select a likely constraint and work through the process. This will typically either confirm the constraint or make another constraint apparent. In this case, underground truck haulage was selected.
Following the five focusing steps, we proceeded to step two and identified a number of opportunities to exploit haulage. These were:
- Routing. As there were two possible mine exit points (a shaft and a portal), trucks were strictly routed to the shortest cycle time alternative (rather than historically preferred options).
- Payload. The trucks weren’t being loaded to full capacity. By installing weightometers on the loaders to ensure the trucks were fully filled, the tonnes per cycle were increased.
- Shift length. A range of options were considered to reduce the length of the pre-shift meetings and prioritise transport of loader and truck operators at shift commencement.
- In addition to exploiting the constraint, we uncovered a number of downstream and upstream processes that could be subordinated (step three) to further enhance productivity:
- Traffic system. Analysis of truck time allocations found a significant portion of a truck’s shift was spent waiting for other traffic to clear in the decline, including waiting for light vehicles. There was an opportunity to change the traffic system and traffic policies to prioritise haul trucks.
- Surface bins. At the top of the shaft, the skip needed to wait until a surface truck was available before dumping its load. Installing all-weather surge bins allowed the hoisting to be decoupled from the surface haulage, which resulted in an increase in hoisted tonnes per day and decreased underground truck delays waiting for the skips.
- Shaft uptime. The shaft had downtime while materials were delivered. Re-routing materials through the portals meant that shaft downtime was reduced, so trucks could remain on the shorter hauls to the shaft rather than being re-routed to the portal during material movement runs.
In combination, these opportunities were modelled to increase system productivity by 33 per cent without any significant investment in capital or time required, as determined by a calibrated discrete event simulation. Further improvements to the exploitation of the constraint would be to truck higher-grade material (ie to increase the revenue through the constraint). This can often lead to the largest potential gains.
Case study two – optimising nickel project strategy
In our second case study, we were part of a team looking to increase the value of a nickel laterite project using a hydrometallurgical process to produce a nickel hydroxide product. An increased value would make the project more attractive to potential financiers.
As in most hydrometallurgical processes, acid is one of the most expensive components of both the capital and operating costs. Therefore, it was decided to make this the system constraint. Note that in the planning process, it is about deciding where the constraint (or bottleneck) should be, whereas for an operation, the constraint has to be identified.
Exploitation of the constraint considered a number of strategic adjustments:
- Variability modelling. We modelled the consumption of acid as a function of the chemical composition of the ore. We then prioritised the ore with the highest nickel grade to acid consumption ratio for scheduling.
- Beneficiation. In order to further leverage the ratio of nickel to acid, beneficiation was considered to remove some of the acid-consuming gangue. This was able to reduce acid consumption by 50 per cent while only reducing average recovery by 25 per cent. The net result was a much higher rate of nickel production, outweighing the lower total nickel inventory.
- Reduced acid addition. A final step was to consider adding less acid. Variability testing showed a step change in acid addition to recovery, indicating a change in mineralogy. Adding these lower levels of acid was able to reduce the acid consumption by 40 per cent for a further 20 per cent loss in nickel. However, the net overall economic benefit was significant.
In order to facilitate the exploitation of the acid constraint, a number of processes were subordinated. In particular, mining had to subordinate to processing. Rather than optimising the efficiency of mining, it was important to have flexibility in the mining sequence and rate so as to supply the preferred material, and a large stockpile of lower-value ore was built for later processing.
Of course, any decision that prioritises short-term production over total inventory (or recovery) needs to be considered within the context of the life-of-mine value. We modelled this project with a linear programming formulation to ascertain the right balance of process settings to maximise net present value. The benefits in this case were significant: a 77 per cent improvement in net present value relative to the original project strategy.
Why is the theory of constraints not widely applied in the mining industry?
Hopefully, we have managed to show the potential of the application of TOC techniques in answering two important needs for the mining industry: maximising the profitability of operations and maximising the value of mining development projects.
There are a couple of typical responses we have received in our application of the TOC and the case studies:
- these solutions are obvious and you don’t need TOC to come up with them
- if TOC is as simple and effective as you say it is, why hasn’t it been embraced yet?
To address the first response: yes, the solutions are not game-changing innovations and they are usually obvious. Yet if they were so obvious, why hasn’t every project maximised its value and why hasn’t every operation maximised its productivity? TOC talks of the concept of ‘uncommon sense’ (ie that solutions are only obvious once they are identified). The fact that the solutions are simple is a powerful benefit. TOC is also excellent at directing attention to the issue that is going to give the most benefit (similar to identifying a single tree in the forest); it’s at this point that other continuous improvement methodologies are employed, focusing on the identified constraint.
Regarding the second point, we definitely hear the term ‘TOC’ used frequently within conversations. It is applied in pockets within the industry. You may have noticed that debottlenecking is a widely used (and abused) term. Unfortunately, typical debottlenecking only looks locally and does not adequately consider subordination across the entire process, meaning that the possible gains are limited. And with good reason: subordination is hard to coordinate and it’s harder to motivate staff to consider the total picture. Upon significant reflection, and based upon experience of others in the TOC world (Goldratt and Goldratt-Ashlag, 2008), there are two important explanations:
- People do not believe a step change is possible. Most will think that, particularly for mature projects and operations, improvements can only be made incrementally and that there is no silver bullet, so why try to find something that (they believe) does not exist?
- People do not believe there is a simple solution to a complex problem. It is painful to think that a problem that you have mulled over for years has a simple and obvious solution. Surely, the solution must be just as complex as the problem!
Notice the use of ‘believe’ in these two statements. The hardest thing to change is not the operation or project strategy, but the beliefs of those working within it. Without changing beliefs, the improvements cannot be fully realised.
The greatest strength, and greatest weakness, of TOC is its inherent simplicity. It remains to be seen whether the industry is ready to apply simple solutions to its complex problems. However, the risks associated with giving this a chance are significantly lower than the benefits, and we would encourage more mining businesses to educate themselves and trial this approach. The fact that the process targets smart solutions by making the most of the current system, rather than spending additional capital, aligns well with the current needs of the market.
Goldratt E M and Cox J, 1984. The Goal: A Process of Ongoing Improvement (North River Press: Great Barrington).
Goldratt E M and Goldratt-Ashlag E, 2010. The Choice, revised edition (North River Press: Great Barrington).