The image above is a conceptual representation of the generative design process. The design requirements and existing conditions are inserted into a parametric model that a computer uses to generate an optimized design based on the design goals, mainly cost efficiency and CO2 emissions reduction.
How is digitalization advancing project goals?
James Melvill: Digitalization of the design process helps to reduce risk and errors by ensuring that the latest design data is available to be used in the following design works. Having access to the latest design geometry from the 3D model or results from the latest analysis, will ensure that work can be done right the first time. This reduces the need for rework. Assumptions can be more accurate, and the design can progress at a faster pace, thus reducing project timelines.
Digitalization of the design process is a key enabler of parametric and generative design—the ability to change design parameters to influence the design and then optimize the design by generating many options and testing their effectiveness.
What other benefits is generative design bringing today?
James Melvill: Generative design helps to refine the design toward the selected solution more quickly. Traditionally, an engineer would make an assumption about the design, complete the calculations and then repeat this process until the design was optimized. Generative design uses automation to allow many more options to be evaluated, and the best one can then be chosen in less time than before. This can even lead to a completely different solution from that expected by the engineer.
Dennis van Heeren: Generative design is especially useful as the engineering design process usually consists of many iterations. Engineers examine the local conditions and determine the client goals to establish a complete picture of the design problem. That information is then used to iterate through different design options and make changes until all design standards and goals are met. Traditionally, these design iterations have consisted mainly of manual tasks, which, in practice, means that only a small number of options can be examined within the available time and budget. Given the limited number of options considered, it makes sense to question how well the outcome is suited to solve the design problem and meet all client goals.
Generative design is speeding up the design process, increasing both efficiency and quality by taking out the repetitive manual tasks as much as possible. This capability allows the engineers to focus more on creative thinking and integration with other design disciplines; it also frees up resources to dive deeper into the unique aspects of the design challenges, such as complex connections and corners.
At the same time, generative design can benefit our clients by providing more insight into the relationships between the design options and different, possibly conflicting design goals, allowing the client to trade-off between these goals. The large number of iterations and design options evaluated using generative design can add more weight to the choice for a best design option, knowing that it is picked out of thousands of lesser options.
In this way, the generative design process uses the best capabilities of both digital and humans. The engineer sets up a parametric model based on the design problem and incorporates a scoring mechanism that measures the design goals for a design option. The computer, leveraging artificial intelligence, then uses an algorithm to generate and optimize solutions in this model using the performance of the options to find the best possible solutions. The engineer can analyze the data and pick the most preferred design option, present the possibilities to the client and other project stakeholders to choose from or change the input of the model and run more iterations in search of even better options.
In a world that is trying to deal with an increasingly extreme climate, generative design allows engineers to minimize the embodied carbon impact of the designs we create for the future.
James Melvill: Overall, generative design enables engineers to better address complex design challenges and create more cost-efficient and sustainable designs, goals that are increasingly prioritized by client organizations.
