Construction - Modular Building

Next: Secondary Effects and Houses that Walk

Its time to change topic and look at various industry verticals. What will digitalisation or what it could mean for them? With digitalisation we mainly mean automation, modular solutions, open interfaces, ecosystem building, service design and agile development that tend to go with it.

TL;DR Modular building is a blueprint for a house where apartments are manufactured in factories indoors based on repeatable designs and later moved to construction site and slid into the scaffolding of the buildings.

Modularity enables reuse of the same design over and over. This creates economics of scale lowering costs and allowing much larger investment into original design. Similarity of designs leads to fleet learning helping both during the lifetime of buildings as well in designing next generation.

How will concepts like automation and modularity affect different industry verticals? It’s time to let the rubber hit the road. Since there are big differences between industries, technologies and practices they use and industry structure in general, it is likely that the impact and end results will differ as well. This is the scope of the second part of this substack series (and the unpublished book that this series is based on).

Second part tries to test the hypothesis that decentralised self-standing communities powered by globally shared open design repositories are possible. By looking at practical problems of different industries it is possible to get a feel where the limits of decentralisation lie. The real world is after all the place where beautiful theories go to die.

Industry Structure

Construction projects start when a developer buys a lot of land and hires an architect for the planned new development. Before that there is usually dialogue with the city zoning on changes proposed by the developer. Developer company can outsource building to a construction company or do it themselves. During architecture work different specialist companies participate in HVAC, electricity, elevator etc. design. 

Many construction companies act as a project management office and outsource different construction aspects to companies focusing on specific part of the project. These in turn use their own internal subcontractor networks resulting in long supplier chains. Tens of small companies work on the construction site. The end result is a web of relatively small companies each focusing on their specialist domain. The project manager sees how these companies relate but the participating companies do not, they only see their direct counterparts.

In Finland industry transitioned into this structure after the depression in the beginning of 90s. Many companies went bankrupt. Surviving ones dismantled large organisations into an endless chain of subcontractors and just left small project office for themselves. This was seen as a way to protect against next recession (and to stay alive). As a result, no one really bears responsibility for quality.

What would it mean to introduce innovation to this ecosystem? As an example, assume there is a new system of laying out electric cables that is 50% cheaper.

Would a company doing electrical installations willingly cut their revenue by that amount and fire half of employees, especially if they are a doing just fine now? How about other parties? They all depend on the expertise of electrical companies and this change would make them very angry. Nor do other companies know know what types of relationships and interdependencies the electrical companies have in the background. Rather than making vital partners angry, it is much easier for all to continue as before, pretend innovations does not happen and just add that extra little bit to the price of the final product as always.

No one has either the economic muscle or controlling position in the system to drive any major change through the operating model. This is why construction is one of the least digitalised verticals where change is slow.

This concept of past decisions tying industries down and preventing to embark on a new journey is called “Path Dependency”.

Only possibility for change is someone from outside of the industry enters and changing the mode of working. This is what happened in the car industry and now everyone is into electric cars.

Modular Building

After a depressing start, what opportunities lie in construction?

Every building built today is a prototype, designed specifically for that specific lot. Almost same work is redone for each building with little cumulative learning.

What’s the alternative?

In shipbuilding each cabin is a well-designed module (because space is premium) separately built and finally placed inside the hull.

Same concept is applicable to construction.  In modular building concept, the building is divided into a scaffolding and apartment modules. Apartments are built elsewhere indoor at a factory eliminating all the typical problems of materials getting wet during construction time or last-minute adjustments. Being standardised, they are easier to build and quality control is easier at the factory leading to significant reduction in errors. Finally, they are slid into the scaffolding (or scaffolding is hoisted at the same time).

Since modules are produced in volume, much more effort can be spent into the design and doing user studies creating results that optimise the use of space while meeting user needs. Volume additionally brings economics of scale leading to lower construction costs.

Lower costs mean that it is possible to instrument these apartments with sensors that monitor the condition during lifetime. This monitoring helps to detect starting problems like humidity in structures. This further reduces the total lifetime costs.

Lower cost means that effort can be spent on making the apartments adaptable so that apartments can be restructured (walls moved or removed for example) when changes evolve with age. Modularity also means that when needs change, I could sell my old module and purchase a new one with a design that meets my new needs better (to compensate age related disabilities as an example).

Modules that are easy to slid into the scaffolding are also easy to slid out. If designed properly that owner can move with the module if employment opportunities are better somewhere else. For example, in Finland there is a large set of 70s buildings in areas that are permanently losing inhabitants. These apartments are practically worthless. Modular apartments protect the investment to a degree.

Overcoming industry structure limitations

Construction industry enterprises are unlikely to do anything like this due to industry structure as described above. Such designs could instead be done as part of university projects, governmental funded programs to achieve some important policy goal (such as resource efficiency and sustainability) or a crypto-led experiments to introduce crypto-incentives on physical infrastructure on a wide scale (city or nation level).

Open sourcing the designs would achieve maximum effect.

The modular apartment itself can be made up of smaller modules like toilets, kitchen etc. These are independent building blocks and can be open sourced to be part of the open repository. This allows to use them in retrofitting older buildings. Smaller teams could produce them in formerly abandoned factory buildings in cities and turn those old buildings from the inside into residential living spaces. This modularity also allows in care phase of modular apartment itself. After all different components have different expected lifetimes. A unit that is easy to disassemble eliminates the need to break anything during replacement and saves costs and time.

Signalling Building Quality and Dynamic House Reputation

For people with different purchasing price, different modular designs make sense. The alternative is to create a “perfect” solution that is so expensive that only a small minority can afford.

Constructing modular houses with different initial purchasing prices creates a problem as house prices in a given area tend to smooth out. The area has bigger impact on sale price than the actual house condition unless differences are very clear like building age implying an overdue plumbing renovation.

If it were possible to build cheaply and sell expensively, most companies would gravitate to the low-cost end and general building quality would go down. This is the reason why construction industry at least in authors home country is not regarded as a shining beacon of quality. You do not know what is incubating behind the fancy leaf-thin decorative bricks until it is too late and warranty no longer is in effect.

This could be helped by having a public classification scheme. Today many home appliances have energy classification and in similar manner modular buildings could have a classification telling whether this is an economy model or more expensive version.

This concept can be enlarged. The collected data over lifetime allows to develop machine learning models that assign to each building a quality index indicating how well it was designed and built and how well the subsequent repairs have improved its quality. Each building can have a ‘reputation’. And each individual building reputation could contribute to the reputation of the developer and construction companies – i.e., company reputation directly from faults and correction costs of their work.

Smart contracts can be used to change the payments model as well. The true quality will reveal itself only with time, so house buyers could utilise automated contracts where the payment of purchase goes to contract and initial payment covers the direct construction costs and rest is paid back on yearly instalments during the next 10 years or so.

This is unlikely to happen since in the industry structure individual house buyers do not have any influence, the only real option would be to create an aggregate of buyers that is large enough to attract some developer to accept such terms.

Turning tables, a constructor who believes in the craftsmanship of their work might sell the apartments with a meagre profit as long for each subsequent change of ownership they get a share of the price increase. This way both the builder and the owner have shared interest in seeing the property well managed.

Side note: Smart contracts can also work to benefit companies in other ways. It is typical that during repair work customers order all kinds of extras that would be nice to have but once they see the final bill, the memory starts deteriorating and they do not remember agreeing to extra items nor to the charged prices. With automation a smooth payment schedule can be implemented where customers move funds to the repair company at well-defined and commonly agreed milestones (technically via a payment channel). And these payment channels can be programmed to accommodate extensions that make agreements visible and payments automated.

Next post we’ll continue and consider what this modularity will mean in secondary effects.