Where Prefab Fits — and Where It Doesn't

Off-site construction pays when a project has the shape a factory thrives on, and the first and most important feature is repetition. A factory's advantages — jigs, learning curves, quality control, amortised setup — are realised by making the same thing many times. A project that repeats a unit (a hotel room, an apartment, a student bedroom, a hospital ward, a classroom) dozens or hundreds of times lets the factory climb its learning curve and spread its fixed cost, which is exactly when off-site gets good and can get cheap. Closely related is volume: enough units to keep the factory busy and justify its setup. Repetition without volume (a beautifully standardised design of which you build only three) rarely pays; volume without repetition (many units, all different) does not either. You want both — many copies of a few well-designed things.

The second favourable feature is a cellular, repeatable layout — buildings naturally made of similar, room-sized, enclosed cells. This is why hotels, apartment and social housing, student accommodation, care homes, hospital and ward accommodation, custodial buildings and barracks are the classic off-site building types: their very programme is a stack of similar cells, which maps beautifully onto volumetric modules or repetitive panels. The building type is doing half the work of suiting off-site before the designer touches it.

Third is a tight programme — a genuine premium on speed — because off-site's parallel working (factory and foundations at once) is one of its strongest, most repeatable advantages. Fourth, access-constrained or disruptive sites: dense urban infill, over or beside live operations (a working hospital, a school in term, a railway), or remote locations where on-site labour and logistics are hard — here the ability to do most of the making elsewhere and assemble quickly on a constrained site is worth a great deal. Fifth, and cutting across all of them, early design certainty: off-site rewards a brief and design that can be fixed early and will not churn, because the factory needs frozen decisions. When several of these align — a repetitive, high-volume, cellular building on a tight programme and a constrained site, with an early-frozen design — you are looking at a strong off-site candidate, and that is no accident: it is the shape of almost every celebrated modular success.

The mirror image matters just as much, because choosing off-site for the wrong project is a common and expensive mistake. Off-site punishes the conditions that deny the factory its advantages, and the first is the bespoke one-off: a building (or every part of a building) designed uniquely, with little or no repetition. Here the factory gets no learning curve, no amortised setup, no economies of scale — you pay for the factory, the transport and the cranes without the repetition that makes them worth it. A single bespoke house, a unique cultural building where every space differs, a one-of-a-kind form — these typically cost *more* off-site, not less.

Second, large open spans and big clear volumes — auditoria, sports halls, large atria, column-free industrial or exhibition space. Volumetric modules are, by nature, room-sized boxes with structure at their edges; they are poorly suited to creating large uninterrupted spaces, which usually want long-span structure better delivered by other means (though components and panels may still help). Third, highly irregular or sculptural forms: geometry that refuses a grid, with every element a different size and angle, fights the standardisation and dimensional discipline off-site depends on. Curves, twists and one-off angles can sometimes be prefabricated, but they forfeit the repetition that justifies the method.

Fourth, and perhaps most dangerous because it can ambush a good candidate, is a high likelihood of late change: a client who will keep revising, a brief not yet settled, approvals or requirements still in flux. Because the factory needs decisions frozen early and cannot improvise like site trades, late change in off-site is far more disruptive and costly than on a site-built job — it can mean re-tooling, scrapped modules, or losing the very speed that justified the choice. Fifth, very small projects: below some threshold, the fixed costs of engaging a factory, transport and craneage simply cannot be spread thinly enough to pay. When a project shows several of these — bespoke, long-span, irregular, change-prone, tiny — off-site is usually the wrong tool, and forcing it produces the slow, expensive, compromised results that feed prefab's bad reputation. Recognising the anti-pattern is as much a professional skill as recognising the good fit.

These rewards and punishments become useful when you turn them into a decision framework you can run early and defend. A practical sequence of filtering questions — each a 'no' that pushes away from full off-site — works well:

One: Does a unit or room repeat, in enough volume to matter? If there is little repetition or volume, the factory cannot earn its keep, and off-site is likely the wrong tool (or confined to a few repetitive components). Two: Can the design and brief be frozen early? If late change is likely — an unsettled brief, a client who revises, requirements in flux — off-site's need for early decisions will be violated and its advantages lost; this alone can disqualify an otherwise good candidate. Three: Do the site access and logistics allow it? Can modules or panels actually reach the site, and can a crane operate there? If transport or craneage is impossible or prohibitive, even a repetitive design may not suit. Four: Is the form compatible? Is the building cellular and grid-friendly, or dominated by long spans and irregular geometry? Only when a project passes these — repetitive and voluminous, freezable, logistically feasible, dimensionally disciplined — is it a strong candidate worth taking to the manufacturer and engineers to test properly.

But the most important practical insight is that the answer is rarely a pure yes or no. The hybrid is usually the right answer. Most real projects are neither wholly suited nor wholly unsuited to off-site; they have repetitive, cellular parts that suit it superbly and bespoke, large-span or irregular parts that do not. The mature strategy is to apply off-site where it fits and site-built where it does not, within one building: volumetric bathroom pods in a site-built apartment block; repetitive bedroom modules stacked above a bespoke, site-built ground floor, entrance and roof; panelised external walls on an in-situ frame; a modular superstructure on a conventional podium and foundation. This concentrates the off-site benefit where repetition earns it, and keeps site-built flexibility where the design needs it. The framework, then, does not just sort projects into 'prefab' or 'not prefab'; it locates *which parts* of a project suit off-site and *which do not* — and that finer judgement is exactly what a DfMA-literate designer brings. As always, the framework yields a design-stage candidacy, not a verdict: the manufacturer's tested system, the engineers and the cost consultants confirm whether the case actually stands.

The framework is a guide to judgement, not a substitute for it, and two honest qualifications keep it from hardening into dogma. The first is that the thresholds are relative and system-dependent. How much repetition is 'enough', how small is 'too small', how tight a programme justifies the premium — these depend on the specific system, the manufacturer, the materials, the region and the market. A mature off-site supply chain with a hungry factory nearby changes the arithmetic from a thin, distant one. So treat the rewards and punishments as directions and strong tendencies, not fixed numeric rules, and test the real thresholds with the people who actually run the factories.

The second qualification is the regional reading, carrying forward the last lesson. In India, where the labour-saving driver is weak, the fit question leans even harder on the other rewards: a project earns off-site here mainly through scale, speed, repetition and quality, so the 'does it repeat, at volume, with a real speed premium?' questions carry extra weight, and the classic fits — large repetitive housing and institutional programmes, precast for repetitive structures and infrastructure, bathroom pods, quick-build schools and healthcare — are precisely where Indian off-site has traction. A small bespoke project that might just about justify off-site in a high-labour-cost Western market is even less likely to in India; a large, fast, repetitive programme is a strong candidate for reasons that have nothing to do with saving labour.

So the discernment this lesson teaches is not a rule but a habit of honest matching: read the project's shape against what off-site rewards and punishes, run the go/no-go filter, look hard for the hybrid that puts off-site only where it earns its place, weight the drivers regionally — and then take a genuine candidate to the specialists to confirm. Do this well and you avoid both errors that dog the field: forcing off-site onto projects it punishes (the expensive, compromised results that feed its bad name) and missing it on projects it would richly reward. That balanced judgement — neither evangelist nor sceptic, but a clear-eyed matcher of tool to task — is the whole aim of this module, and the foundation for the systems, the DfMA and the disciplines the rest of the course builds.