How relevant is ISO 9001 quality assurance to architectural design activity?

“Doing things right the first time, on time, all the time.” It’s a common phrase in the quality assurance lexicon. Undoubtedly for most of the activities of an architectural practice that edict is applicable. But not for design.

Which seems odd, because pretty much everything we do as architects flows from design. If we haven’t done a design most of the consequential activities simply won’t happen. No need for a planning application and there won’t be any construction documentation. Most communications won’t be needed, we won’t need to go to meetings, financials won’t need to be totted up and billed. And forget project marketing.

A more accurate description of architectural design could be “Doing things right in the end having explored other possibilities and having exhausted resources, time and money.” But I’d reckon we’ve never seen that written in an architect’s quality assurance system.

Having a certified quality assurance program can be necessary for architects when pursuing projects. And it’s true that a well-configured QA system can greatly assist the work of an architectural practice by analysing and making explicit those actions and documents which are common to that practice across projects.

Unfortunately it is usual for project procurement programs to assume that architectural activities, including design, occur in a waterfall model- a linear sequence of steps. Quality assurance programs typically follow such a logic. While acknowledging the iterative nature of design by incorporating feedback loops into the design brief, it is nevertheless common to assume that the formation of the design brief precedes the design activity.

However design is neither linear nor consequential. Design typically does not receive definitive inputs, as the knowledge arising out of the act of design will inevitably replace those inputs. Because those inputs may be changed repeatedly there is usually no clear definition of a successful design against which to test the preferred design. And those changing inputs and expectations means there is no accurate way of predicting how long the design process will take- which will inevitably be different from the length of time that has actually been scheduled.

The International Organization for Standardization notes that “Implementing ISO 9001 means your organization has put in place effective processes and trained staff to deliver flawless products or services time after time.” For almost all architectural activities that is entirely achievable. For design, however, it is decidedly problematic. Design can be suggested, but not dictated. There are design methodologies which can be explored, but simultaneously highly individualised explorations deliberately or unintentionally incorporating techniques such as analogical thinking, replacement postulation and opposition will be required. The act of design will inevitably be influenced by each individual’s social and cultural background, education, knowledge, confidence and skills.

And so the singular most important architectural activity- that of project creation- finds itself unable to be assured. But perhaps that is not such a bad thing? The world is changing at an accelerating rate, and recycling design solutions that used to work won’t answer the need for innovative solutions. Allowing the act of design to be unrestrained by the imposition of quality assurance dogma may well be a better outcome than attempting to corral the necessary free thinking.

Let’s take a diversion for a minute or two and use a real- and quite typical- architectural project as a way of examining and further exploring that thesis. The site for the project is located in the central activities zone of a suburban activities centre in greater Melbourne. The site dimensions are defined by the land title. The planning scheme defines building setbacks and the absolute maximum height is set by shadow-cast onto a residential precinct south of the site. The preferred maximum height is defined in the planning scheme by a diagram illustrating urban form stepping down from the highest buildings in the centre of the activities area to lower buildings at its periphery.

At face value therefore a maximum building volume quickly and clearly reveals itself. If we go with that we can quickly get on to designing the building. If, however, our approach is to use good design first principles to analyse each constraint, the result will be completely different.

The “stepped height” diagram is an approach much loved by planning authorities because it is easy to explain and seems at face value to be logical. But the way we experience, explore, understand and enjoy urban environments is very different. The stepped heights approach can and should be challenged. If successful in our argument, the height of the building will be higher than shown in the diagram. Therefore we are confronted with two very different heights and therefore potentially two very different design outcomes.

To mount that argument will require presentation of first principles with accompanying research and design exploration. We are therefore designing before we know our inputs, unless we consider the input for height is “dunno”.

Simultaneously we consider the setbacks. The front setback is intended to create a clear delineation between a podium height- common with adjoining buildings- and the tower above. The podium has no setback requirement but the tower does. Initial design explorations show, however, that a better urban outcome would result if the tower was pushed to the northern boundary, which fronts a major road with little pedestrian movement, and further away from the southern boundary facing onto a smaller, more intimate street with greater pedestrian occupation. The planning scheme is lazy: it only considers one way of defining the podium and the nebulous benefits of that have overridden the quality of the environment south of the site. But there are other ways to delineate that podium, and so we consider that we can mount an argument to achieve a zero setback on the north accompanied by a larger southern setback.

So now we effectively have no inputs for height or setbacks. As fluid as that situation is, there are further complications. Aware of increasing demand for housing in the locale our client is proposing a primarily residential building. Like most planning authorities, however, the council is understandably very concerned that residential use is displacing other uses in central activity areas and will prove to be very difficult or impossible to reconfigure in the future if different needs arise. To mount the arguments of increased height for urban form and reduced north setback for pedestrian benefits will be difficult if we are also seen to be barely responding to the council concerns about employment opportunities. We therefore need to incorporate commercial space to a greater extent than our client desires. How much will be determined by how we think we can best negotiate with the council.

Now we are designing without any guidance for building dimensions or agreed mix of uses. The council ultimately accepts both of the propositions we put to them, and will provide subsequent planning approval for the scheme based on those. The now-in-place design constraints which have been suggested by the design explorations are very different from those contained in the planning scheme. The notion inherent in quality assurance systems of assembling inputs which provide informing conditions for our design, and against which design quality can be assessed, has completely gone out the window.

Nothing in what is described here is unusual. Back in the last century I had a saying “the latest brief is the latest drawing” as a response to our clients whose instructions? were “I have this site- what can I do with it?” Architecture often encounters complex scenarios without obvious solutions at the start. We can be faced with interlocking possibilities and constraints that evolve over time. We may not properly understand what we are facing until we develop and assess solutions- information may be missing because we haven’t discovered it yet. There are many stakeholders who may not mutually agree. Constraints can change over time and there is usually no definitive solution, but more likely a range of optimal solutions yet to be discovered.

The only strange thing about that is that quality assurance systems don’t seem to recognise that fluid and necessarily unrestrained nature of design. Instead of the brief leading the design, the act of design produces critical elements of the brief. Might it be better for the QA system to acknowledge that? Should the system also acknowledge that assembling the brief and the act of designing best occur simultaneously and osmotically? Should quality assurance forget the fiction of repeatability of key aspects of the design process? Should it instead be recognised that divergent design outcomes will be inevitable when different architects with different experience, skills, confidence and personality undertake that design?

The design approach described above was a response to the input of the planning scheme and how that could be questioned and those inputs changed. But that response was seeded by the knowledge and experience of the architect which revealed the alternatives, and their confidence and chutzpah to be convinced that they could persuade both client and council of the correctness of that approach. Is it possible for a quality assurance program to promise flawless services time after time when the success or otherwise of such critical activities is reliant on the individuality of the architect?