I take a cue from a fascinating article by Paul Kremer that helps to clarify a hot topic within the wood industry.
The assessment of pros and cons of the various levels of automation in giving shape to a CLT plant.
This analysis becomes useful if you think about what is happening overseas: a real ‘mass timber rush’ with mega hyper-automated plants ready to go.
So, thank you Paul, to let the question arise.
Are there alternative solutions to a massive investment, involving a lower risk level and more affordability?
The answer is found in Paul’s article, but at this point I ask myself: why should we keep building another CLT plant?
This may seem to be a provocative or meaningless question. It should be clear that if you intend to build a CLT plant, there are economic solutions upstream that can make it convenient.
However, the question is not entirely meaningless, especially if there are some considerations attached to this matter that I will try to briefly explain in this article.
CLT represents the building block of a more complex structure and is manufactured through an industrial process with a level of adequate automation.
The designer who stays upstream of the industrial process has to use dedicated software to communicate and send the shop-drawings directly to the CNC plants of the CLT plant. All of this entails a design performed by a structural engineering firm with proven experience in mass timber.
Furthermore, to define a structure, panels have to be joint by connections ad hoc, calculated and designed along with a series of membranes that make them waterproof and acoustically and thermally insulated.
This implies the presence of additional players within the same supply chain.
If the above mentioned conditions weren’t all met, we would not obey a fundamental rule inherent to this constructive methodology: “What you draw is what you get on-site.” as explained by Eng. Franco Piva. That is why we can define CLT as an engineered product.
Allow me to add another critical and quite underestimated element to the supply chain: assembly.
[I will soon dedicate an entire section to it that I am not mentioning here for a matter of space.]
In a nutshell, the construction of the timber structure identifies a series of supply chain members that, as a result of a shared digital environment, are even more increasingly interdependent.
Perhaps this industrial process, intrinsic from the start to the delivery of the building, represents the main innovation of the mass timber construction industry.
The industrial process – unlike the process of conventional construction – is, by its very nature, highly efficient and able to produce with millimeter precision what required within the agreed times.
This manufacturing phase per se, involves a series of technological, logistical, and time constraints that alter the supply chain both upstream and downstream, within the process.
Here is a reproduction of a chart, from the paper: “Insight into the Global Cross Laminated Timber Industry” (1) as a perfect representation of the CLT supply chain.
(1) Society of Wood Science and Technology: Insights into the Global CrossLaminated Timber Industry – Dec. 2017 Lech Muszynski ´ 1 *, Eric Hansen 1 , Shanuka Fernando 1 , Gabriel Schwarzmann 1 , and Jasmin Rainer 2 1 Department of Wood Science and Engineering, College of Forestry, Oregon State University, 119 Richardson Hall, Corvallis, OR 97331.
The constraints of a CLT plant
A CLT plant is a peculiar industrial process: it does not manufacture standard products, neither it works on ‘make to order’ , but only on project. Each order is merely different from the others.
A highly flexible management of this process is made possible by the CNC machines equipped with a digital programming ‘controlled’ by the design phase.
Design and manufacturing can be geographically distant from each other, since their contact occurs mainly by means of files.
The CLT plant is also bound by:
During the planning phase of the plant, a break-even-point simulation is run to make sure of meeting the waterline.
A backlog is provided to secure continuity during the process flow.
The question: “Have we competent technicians and sellers who can provide us with this data with reasonable assurance?” is asked repeatedly.
The availability of raw material upstream is checked to ensure a consistent supply.
The kind of raw material depends on the degree of integration upstream.
The CLT plant will be more advantaged if it owns a sawmill. In the case of an upstream forest, it would be even more beneficial — the smaller is the integration, the weaker the ability to deal with raw material availability at competitive prices.
A careful analysis of upstream resources is, therefore, highly advisable.
Technical and Skills Constraints
A CLT plant, based on numerical control systems, cannot start working if someone does not activate a set of information or a digital network that can make them communicate.
As we said, one of the advantages inherent to CLT is the file transfer method: from the final stage of design to CNC machines.
Only those with the required engineering skills can implement this set of information, and this automatically excludes all other structural engineers who are not proficient with timber or who do not have digital access codes to transfer this information.
From a recent Swedish study emerges a scarce availability of technical skills on the mass timber market concerning architects and structural engineers. (2)
The criteria of success of any industrial process are performance and quality. Efficiency is nothing more than a ratio between a given output and a unit of time. And it is the time factor that forces any component of the supply chain to comply with the conditions imposed by the industrial process.
Today most process manufacturers operate according to “Lean Manufacturing” rules as a direct application of “Lean Thinking” introduced in the early 1950s by Taiichi Ohno of Toyota Manufacturing.
Therefore, I hope that future CLT plants will adhere to this method.
One of the Lean mantras is to avoid any output of “muda” (a Japanese word meaning waste).
Muda is any activity, material, or investment that along the phases of a process does not produce any added value. Muda can be identified in time waste, stock of material, unjustified waste, and above all, unsaturated production capacity.
So how do we avoid muda output?
There is more than one applicable method, but to remain relevant to the theme, the one best suited to our context is called “the pull method.” The process flow is determined by its last phase rhythm (takt time), which is the “pulling” part. Therefore all the previous phases have to synchronize with the last one.
In the case of an uneven flow between two intermediate steps, we can have either a semi-product excess or a stop in the line. In both cases, it’s waste.
When considering a CLT plant output as an intermediate step of the whole supply chain, CLT is not a finished good anymore, but a semi-product of a broader process, aimed to deliver a building.
In this case it becomes useless to maximize the production of a specific phase.
The production capacity must be commensurate with all the other elements of the supply chain, having as its final objective the fulfillment of market expectations in terms of on-time delivery of the building and customer satisfaction.
What is your ultimate business?
Not having a clear idea of what an outlet market is, can lead to misinterpretations and wrong strategic decisions.
In the 1930s and until the early 1950s, American railway companies created an oligopoly regime on long-distance transportation. As this happened, end users felt the need to have a more efficient means of transport.
The companies were unable to see their inefficiency, along with the structural limits that the railway entailed. They did not pay much attention to the emergence of a competitor. Indeed a series of competitors – interstate highways, car manufacturers that provided more comfortable and faster cars, airplanes manufacturers and airlines – developed, to create, together, the conditions for a better and more dynamic range of means of transportations.
The US railways, too focused only on increasing their market share, had failed to recognize themselves as a subset of a broader market.
I used this example to clarify that CLT is not the outlet market, but always the same old construction market, only, it has to be seen with new eyes.
These considerations lead me to ask myself further questions.
Why, then, in the United States and Canada is there a race to install CLT plants with a high production capacity, that seems not to be justified by real demand?
Why, despite this high production capacity, is the CLT price abroad, even higher than in Europe?
For how long will CLT be the ideal structural solution?
The independent variable of the technological speed doesn’t seem to allow much longevity.
Are we sure that the times from now up to technological obsolescence, are sufficient to have an acceptable ROI for those who have invested?
Economies of Scale or Economies of Skills?
I agree with Paul Kremer and his implicit criticism regarding people who don’t mind investing massive amounts to climb as fast as possible the walls of a market, of which, the top is still not visible.
Sometimes people want to pursue economies of scale by all means, without respecting the physiological rhythms of a company, and maybe they think that the automation of each phase is the only solution to get to the top faster. The more you try to automate, the more variables you enter, the more complicated the process control becomes. So it ends up that economies of scale become diseconomies of scale and waste reigns supreme.
I believe that economies of scale should be opposed to ‘economies of skills.’ But that’s still another story.
So once again, why build a CLT plant?
And if it was more convenient to buy an existing production capacity on the market, without having to bear the fixed costs that each plant has to take?
This will certainly appear as a temporary solution, but, perhaps, it would determine a more comprehensive approach within an industry that needs no speculative bubbles.
(2) Roos, A., Woxblom, L. & McCluskey, D. 2010. The influence of architects and structural engineers on timber in 2 construction – perceptions and roles. Silva Fennica 44(5): 871–884.
Thanks to Ledinek for the photos in this article.