Auteur: Erik Pols
Property development is inherently risky, and of all the phases the feasibility phase is undoubtedly the riskiest. Critical decisions need to be made without full information, and the consequences can have major impact on the (financial) success of the project. For example, governments require developers to commit to energy efficiencies as part of the permit process, even though a small change of plans in the design phase can mean 10 solar panels less or more. Another example is the effect of sun and shadow on surrounding buildings, or the fact that many parking regulations require preliminary architecture designs to determine feasibility.
Property developers have recognised the advantages of standardised building processes for a long time now. Many construction companies have experimented with modular building, and even though that has not yet taken off on scale, most do have forms of standard designs that they apply to building projects. Many of the advantages have been applied to later stages of the development, with clear advantages in logistics, consultancy costs and risk management.
At Vellum, while working with our clients we have come to recognise the trend towards standardisation as a major opportunity to reduce risk in the feasibility phase. Standardisation in itself is only marginally helpful to reduce this risk, because many factors still change per project. Every site still has its own regulations and neighbouring buildings that influence the final designs. However, by applying generative design algorithms to standardised designs, we suddenly are able to provide a wealth of analyses based on assumptions that have a high level of confidence.
Reduce regulation and planning risk Generative design algorithms can make sure designs stick to planning regulations, for example by making sure buildings conform to building codes, contain enough parking in or outside the building, sufficient storage, etc.
Increase confidence about intangible requirements. Governments often have abstract urban visions that aim to support political goals. For example, a city might require buildings to be tall and narrow, to have an ‘open’ plan, or require a certain amount of green. Although understandable, these factors can be abstract but have big impact on projects. Generative design can take these requirements and translate them into tangible ratings that create a great basis for discussion with local governments.
Commit to energy efficiency labels with higher confidence. By generating buildings with standardised building designs, algorithms are able to make confident assumptions about many parameters needed to calculate energy labels. Things like insulation, type of heating, number of solar panels etc can all be included in the algorithm, and generate energy ratings with confidence.
Avoid planning risks by doing shadow studies early. The feasibility of a project is often heavily influenced by the situation of neighbouring buildings. By doing shadow studies early, it becomes easy to understand quickly whether a project adheres to the regulations set in the zoning plans.
Maximum value by comparing more concept designs. Generative algorithms are able to compare many variants of designs quickly. Applying analyses on view distance, available daylight etc to finished designs was already part of the standard process, however now it is possible to do this in a matter of hours without the use of expensive consultants that need to integrate the designs in advanced software solutions. Also, by doing the analyses in later phases, many choices would already have been made, making it hard to change the outcome retrospectively. With generative design, it is possible to find the best solutions much earlier.
We believe we've only lifted the tip of the veil of what generative design can do. The future certainly is exciting!