Assembly Sequence & the Programme

A modular building is erected in a sequence, and the sequence is mostly dictated by a simple fact of gravity and geometry: modules are stacked, so lower ones must be placed, levelled and often secured before the ones that bear on them. Assembly generally runs bottom-up and outward from a starting corner, each module landing on the prepared bearings of the structure or the module below, checked for level and alignment, connected, and only then ready to receive the next. This creates a web of dependencies: module B cannot be placed until module A is down and secured, because A carries B, or because the crane needs A in place to reach B, or because the connection between them can only be made from a position A creates.

These dependencies are the heart of sequence planning. Some are structural (you cannot land a module on a support that is not yet there). Some are about access (a joint between two modules must be reachable by the person who makes it — if the next module seals off that access, the joint must be completed first). Some are about the crane (you cannot place a far module if a nearer, taller one now blocks the swing or the reach). And some are about services: the pipe and cable connections between stacked or adjacent modules frequently have to be made *before* the next module closes over them, because afterwards there is simply no hand-access to the junction.

The consequence is that the order is not free. There is usually a small number of workable sequences and a large number of impossible ones, and finding a good sequence is a design task done with the manufacturer and the assembly contractor, often rehearsed virtually in a BIM model before a single module is made. The sequence is typically captured as a numbered erection drawing — module 1, module 2, module 3 — that the factory, the transport and the crane all work to, so that a truck delivers module 17 at the moment the crane is ready to place module 17.

For the designer, the lesson is that buildability in the right order is a design property, not something the site discovers. A plan that looks fine on paper can be unbuildable if some module can only be reached after the thing it depends on is already sealed. Designing with the sequence in mind — simple stacking logic, connections that stay accessible until they are made, a clear starting point and direction — is what makes the rehearsed assembly possible. The detailed erection engineering, temporary stability during assembly and the final connection design remain, as ever, with the structural engineer and the manufacturer's system.

The most visible promise of modular construction is speed on site, and the assembly sequence is where that speed is cashed in. Because the modules arrive near-finished, the on-site programme is not the long, sequential chain of a conventional build — dig, found, frame, make watertight, first fix, plaster, second fix, finish, each trade waiting for the last — but a short, intense assembly phase: land the modules, connect them, weatherproof the joints, stitch in the services across the interfaces, commission, hand over. A building that might take many months of site trades can be stacked in a matter of weeks, with storeys going up in days.

But the compression is real only if the conditions for it are met, and they are demanding. The design must be frozen early enough for the factory to build to it — a late change cannot be absorbed by site trades because there are almost none. The sequence must be worked out and buildable so the crane never waits. The logistics must deliver just-in-time in sequence. And the connections must be designed to be made fast and reliably — a modular programme can be wrecked by an interface detail that is slow or fiddly to complete, multiplied across hundreds of joints. The on-site time is dominated not by 'building' in the old sense but by connecting and sealing, so the speed of the programme is really the speed of the interfaces.

This reframes where the designer's effort pays off. On a site-built job, design effort spreads across the whole construction; on a modular job, a huge share of the on-site programme — and its risk — lives in the joints between modules and the sequence of making them. Shaving time off each connection, or removing a step from the sequence, compounds across the whole building. This is lean thinking applied to assembly: reduce the steps, standardise the connection, make the sequence flow without waiting.

The honest framing matters here too. Modular's headline speed is the *on-site* speed, and it is genuine and valuable — less disruption to neighbours, faster handover, earlier revenue for a client. But it is bought with a longer *front-end*: the early design freeze, the factory tooling and production, the detailed sequence and logistics planning. The total time from decision to occupation is not always dramatically shorter than a good conventional build; what changes is *where* the time sits — more in planning and factory, far less on the open site. Whether that trade is worth it is exactly the kind of honest, project-specific judgement the course keeps insisting on, and the binding programme and commercial case belong to the project team, not a rule of thumb.

The single biggest time lever in off-site construction is not that any one task is faster but that two slow tasks happen at the same time. While the factory is building the modules, the site crew is doing the groundwork and foundations — excavation, piling, the ground slab or podium, drainage, service connections into the site. In a conventional build these would be sequential: you cannot frame until the foundations are in. In modular, the foundation work on site and the module-making in the factory run in parallel, overlapping for weeks or months, so that by the time the modules are ready the foundations are waiting for them.

This parallelism is why modular programmes can genuinely compress even though assembly itself is only part of the story. The weeks the factory spends building modules are not added to the site programme; they are *hidden behind* the site's own groundwork. The moment of truth is the handover from groundwork to assembly — the foundations must be finished, accurate and ready at exactly the point the first modules arrive, because the whole just-in-time choreography of the previous lesson depends on it. A foundation that is late, or wrong, collapses the advantage: the modules arrive with nowhere to land, or land on a base that does not match them.

For the designer and the planner, parallel working has concrete implications. It means the foundation and podium design must be resolved and started early, in parallel with the module design, rather than waiting for the superstructure to be finalised — a reversal of the usual rhythm. It means the interface between the site-built substructure and the factory-built modules (the subject of the next lesson) must be coordinated between two teams working simultaneously. And it means the programme is drawn as two streams — factory and site — that must converge precisely, rather than one long line.

There is also a risk to name. Parallel working assumes the design is settled enough that the factory can commit to production while the site commits to foundations — both expensive, both hard to reverse. If a change emerges after both streams are running, it is costly on both sides at once. This is another face of the course's through-line: off-site rewards early certainty and punishes late change, and parallel working is precisely where that certainty is spent. Used well, with a frozen design and a coordinated interface, it is the mechanism behind modular's most compelling programmes. The binding programme logic, the critical path and the commercial consequences of the overlap are for the project managers, engineers and the manufacturer to own.

Everything in this lesson converges on one principle: the assembly sequence is not discovered on site, it is designed into the building from the earliest stages. Because a modular building cannot be improvised and cannot be un-built to fix a buried mistake, the order of assembly — and the accessibility of every connection at the moment it must be made — is a design requirement with the same weight as structure or fire. Teams increasingly test this in a 4D BIM model, adding time to the 3D geometry so the whole erection can be rehearsed virtually, clashes and dead-ends found, and the sequence proven before any steel is cut.

What does 'designing the sequence in' actually look like for the designer? It means choosing a clear stacking and starting logic so the crane and the teams have an unambiguous order to follow. It means keeping connections accessible until they are made — never detailing a joint that can only be completed from a space the next module will seal, unless that joint is genuinely completed first. It means designing service interfaces that can be connected in sequence, with access panels and routes that respect the order of assembly. It means coordinating the weatherproofing sequence so the building is protected as it rises, not left open to the monsoon between lifts. And it means thinking about temporary states: a half-assembled building must be stable and safe, which can govern propping, bracing and the order itself.

The payoff is large and concrete. A building whose sequence is designed in assembles quickly, safely and predictably, with the crane fed and the teams never blocked — the balletic speed that makes modular worth doing. A building whose sequence is an afterthought discovers its dead-ends at the worst possible moment, with a module on the hook and the programme burning, and may need expensive rework or improvised site trades that erase the off-site advantage. The difference is almost entirely upstream, in design.

This is why logistics and assembly, which can feel like post-design site matters, are in truth core DfMA concerns — the 'A' in Design for Manufacture and Assembly. A module that is easy to make but impossible to assemble in a workable order is a failure of DfMA. The designer's task is to hold the factory, the truck, the crane *and the sequence* in mind from the first sketch, and to work the buildable order out with the manufacturer, the structural engineer and the assembly contractor. The binding erection engineering, temporary-works design and connection details stay with those specialists; the design that makes a good sequence possible is the architecture.