Several times recently, I’ve had discussions where people are surprised that Lean manufacturing is something that would really benefit the offsite sector. Prior to that I’ve had colleagues lamenting the lack of a standard product and how we need have a standard module design, house type or flat. The perception is that lean is about producing a high volume of matching products, “you can have any colour you like as long as it’s black”. Thing is, that quote is from Henry Ford and the world of mass production. Lean manufacturing, on the other hand, originated in a much smaller market with diverse product offerings. Lean was born in post World War 2 Japan from the principles of lower volume, high variability with significant space and investment constraints preventing Kiichiro Toyoda from simply copying western plants. I find that this tendency to lump all manufacturing together and the lack of a clear distinction between mass production and lean manufacturing produces the misconceptions around lean and offsite. This is why I used this idea to open our exploration the lean house over the last two posts and proposed a clear philosophical line between industry 2 and 3 as well as the technological one.
Following this, I delved into the power of Throughput, Inventory, and Operating Expense as key metrics, and how Value Stream Mapping helps identify areas for improvement. We've also discussed the ideal of Single Piece Flow, Just-in-Time (JIT) and the reduction of waste for a truly efficient system. I finished up pointing out that Toyota had decades to build TPS and have been refining it for the last 40 years. Achieving perfect single piece flow and JIT can be a challenging feat, especially when starting out. Real-world manufacturing environments often face demand fluctuations, equipment breakdowns, and supplier issues.
So, how do we get started on our Lean journey without getting overwhelmed? I’m a keen advocate of the idea of getting started even if it’s not perfect and then using problem solving and continuous improvement to ramp up. In this pragmatic approach we can bend or even break some of the rules to allow our starting point. But ultimately when deciding where to bend a rule or even go and break it you need to be clear on your destination and the core principles or culture you’re building.
One common example of this is the use of buffers, where a stock of inventory is kept between processes to ensure that the follow-on process won’t be forced to wait should there be an issue downstream. In the offsite volumetric factory this is common between the subassembly of 2D elements such as floors, walls and ceilings and the box build process.
Buffers also help to balance uneven workflows, a prime example in volumetric construction being the site install of modules. A factory will build up an inventory of modules which then get dispatched to meet the install plan, often the whole build is ready at the factory before site install is started even when this might require space for 100 modules or more.
The same goes for stock levels, much of the construction industry supply chain isn’t ready to delver anything close to just in time so the offsite factory requires space to have sizable deliveries and store it as it consumes the materials into the product.
It is reasonable to deliberately create a controlled form of waste (inventory) to prevent another form such (waiting) if by doing so enables you to keep your flow steadier and get the product to the client sooner. The problem comes when the inventory isn’t well controlled by overfilling it or not recognising it as waste that needs to be reduced. Looking at the example of the subassemblies, should there be an issue upstream which prevents the buffer being drawn from, pushing more inventory into that buffer and overfilling it is overproduction so now you’re generating two wastes in an attempt to prevent one. Managing your buffer levels is therefore critical and setting clear minimum and maximum levels controlled with a pull-based system like Kanban will help with this control. One thing to be very clear on is how you manage these levels as output ramps up. It’s easy to set targets for a day’s worth of production and as you ramp up output this level will increase with it, which from 1 module a day to 2 is easy but 2-4? What about 16 modules a day? Your now firmly in the realms of overproduction with oversized buffers and the associated risks such as storage costs of that inventory.
Part of your plan for ramp must include for a clear assessment of maintenance needs, risks and existing performance. If you have evidence that means you are regularly needing buffers at your current rate for a full day’s downtime you’re not ready to ramp up, make your next improvement project focus on the need for the buffer so you can keep it the same size or only slight larger after ramp up as it was before. Then once the ramp up is delivered and running, plan to target your buffers further so you can reduce them especially if they were increased in the ramp up process.
While the use of buffers is about breaking the ideal Just in time system using small batches breaks the pure single piece flow. I always used to joke with a manufacturing director that his batch build processes built a restaurant in 8 week but with a flowing process I could do it in 8 days. Which with the correct prep, jigs and fixtures and process development a flow would be able totally achievable. However we still have some arears of the process where batching has benefits to help us 'get started', the most common is to manage drying times. Tape and jointing or Tiling have very long waiting periods where further work can’t be achieved, so while setting up it’s reasonable to designate these tasks to a batch process. Once you’re up and running improvement projects and work in these areas should be carried out to reduce the size of these batches and there is a real R&D potential for industry to look at replacing old fashioned traditional materials in these areas.
 |
| Like the high speed wheel change in F1 SMED allowed for the efficient change over between products |
By taking this approach the goal is to move towards the more ideal Just in time single piece flow system and is why understanding the underlying principles of Lean is so important to our industry instead of just knowing the tools or jargon. Lean is more than Just in time and Pull is more than just Kanban both things that are more common misconceptions like those above. When you dig into the history of Lean it was SMED or Single Minute Exchange of Die that was the real game changer. It allowed Toyota the ability to develop single piece flow in an industry that traditionally used batch sizes in the hundreds of parts. It moves away from the idea of economies of scale to ones of whole system efficiency and allowed Just in time to come about. In our industry we don’t have complex setups between products, instead we are often faced with difficult product balance as some parts of a building are just fundamentally more complex than others. So, we need to be going back to the basics of the lean house and finding our own enabler like SMED was for Toyota while taking advantage of over a century of learning manufacturing offers us.
One particular challenge unique to our industry is the transition from factory to site, be it a small plumbing subassembly, 2D cassettes or fully fitted volumetric modules. Getting the keys into our customers hands relies on us managing this process, there is no point in producing the best factory products in the world if they go to site, get damaged or wet and take months to finish. To really reap the benefits of offsite and MMC getting the site side sorted and aligned with lean thinking is our enabler. Having recently spoken with Paul Richards from GUR build we discussed this at length and how important it is to bring site thinking into alignment, because right now that disconnect often causes conflicts with factories and the MMC processes. There is so much that can be gained by bringing ‘the factory to the site’ as opposed to a tendency to have a building site in a shed for the factory.
Over the last few posts we've explored many of the principles and ideas underpinning lean, today I applied some of those in a practical manor to a portion of the system. In my next post I’ll be taking this further going into the steps to lean implementation, looking at the practical application of those steps and how they can apply both to the offsite factory and the building site.
And if you find these articles interesting and would like to discuss more, please feel free to message me on LinkedIn.
