Generative Design

Humanity goes through yet another important cycle in the history of the production of goods. Some experts compare the current moment as a new industrial revolution. Connected devices, cloud processing, borderless collaboration, and new production models based on 3D printing create an enabling environment for manufacturing disruption.

The emergence of a new class of computational tools and the cheapening (and consequent proliferation) of alternative manufacturing mechanisms tend to influence the scenario in the short term radically.

In the latest blog post about immersive engineering, we have talked about generative design. Along with the test, we comment about how the generative design is improving the Computer-Aided Design (CAD) technologies. But after all, what is generative design? Do you know?

If you are looking for a complete CAD – Download SolidFace for Free

So, on this blog post, we decided to talk more about this subject. Stay tuned! There is a lot of good content here. Check it out!

The definition of Generative Design

Many people think that generative design is only something about engineering vocabulary, but is not. Generative design is the process that uses programming (therefore, computers) to enhance the creation of products, images, etc. It offers designers, architects, artists, and designers the possibility to invent new and more efficient options beyond human capacity.

Generative Design begins with establishing the project’s objectives and then explores all possible permutations of a solution to find the best option. By using cloud computing, the software assesses thousands of design and design choices, testing configurations, and learning, from each interaction, what may or may not work.

Generative design is about designing a project without thinking directly about the final object but studying the rules that can generate that object and then “managing” these rules to obtain many variants of final objects. From there, you can make a selection and feed the process with new rules or by changing values ​​(parameters) to generate new objects. And do this successively as a way to ‘obtain’ a suitable end product.

The origin of Generative Design 

There is not a certain on when the generative design has been originated. However, there is a name that may be the generative design pioneer. His name? Celestino Soddu.

Celestino Soddu is a renowned architect and professor of Generative Design at the Politecnico di Milano university in Italy. He is one of the pioneers (or even the father) of Generative Art and Design. Its first generator software was designed in 1986 to create 3D models of infinite variations of typical Italian medieval cities.

 In 1987 Celestino Soddu created the artificial DNA of medieval Italian cities capable of generating endless 3D models of cities identifiable as belonging to the idea. In 1989 Celestino Soddu defined the Generative Approach to Architecture Design and Design City in his book “Citta ‘aleatorie.” In the call for the generative art conferences in Milan (annually since 1998), the definition of Generative art by Celestino Soddu:

“Generative art is the idea realized as a genetic code of artificial events, as the construction of complex dynamic systems capable of generating infinite variations. Each Generative project is a software concept that works by producing unique and non-repeatable events, such as music or 3D objects, as possible and multiple expressions of the generating idea, strongly recognizable as a vision belonging to an artist/designer/musician/architect/mathematician.”

Generative Design steps

In a trial of success or error, within a napkin, parchment paper, or on a black CAD background, much of an architect or engineer job is to make and redo tests, lines, shapes, copies, or even throw at the trash and start over. 

From an initial idea to a final project, there is an exhausting and long way. Designing, in your essence, is made by making infinite decisions. A change influences the whole project because, many times, there are many other elements to care about. In short, an exercise of choices and concessions. 

Whether it is possible to design with legislation as possible without impacting the environment 

According to Celestino Soddu, generative design is a process that uses organized algorithms like non-linear systems to obtain single and unrepeatable results, executed by an idea code, as in nature. The analogy with nature is illuminating and allows drawing some parallels. Taking the example of a tree with a large trunk and stronger base withstands all the weight and the bending moment caused by the wind and its weight. From there, several secondary branches are emerging and, at last, the leaves. This same logic can be used in art, design, engineering, and architecture.

Evidently, in the whole project, there are always tight deadlines and budgets, a client usually in a hurry and a limited amount of time to think about all possible combinations and if the design decisions made are, in fact, the most appropriate. This is where the concept of Generative Design has increasingly appeared.

Generative Design combines parametric design with artificial intelligence in conjunction with the restrictions and data included by the designer. 

The process begins with the definition of the parameters that will compose the characteristics of the solutions generated. The software used then creates hundreds or even thousands of variations of a solution that meets the restrictions initially imposed. Results can vary between image, audio, or three-dimensional structures files with applications in highly complex products or artistic productions. Production software and technologies vary according to the context and area of ​​each project. The essence is the same: using the processing power of computers to expand the range of possibilities in solving a problem or developing an idea.

In the field of industry, where the complexity of projects increases due to the number of requirements to be met, this technology presents itself as a great tool for designers and product engineers. Generative algorithms use precisely all the project requirements, such as weight, cost, mechanical properties, and ease of manufacture, to generate a large number of formal solutions within the defined context. This allows the professionals involved in the project to have a wider range of solutions to explore, improving the results, and the efficiency of the development process.

So, what is the connexion between Generative Design and Artificial Intelligence?

Artificial Intelligence, which you will see out there being referred to only as AI, is a technological advance that allows systems to simulate human-like intelligence – going beyond the programming of specific orders to make decisions autonomous, based on standards from huge databases. 

Artificial Intelligence, in essence, allows systems to make decisions independently, accurately, and based on digital data. In an optimistic view, what multiplies the rational capacity of the human being to solve practical problems, simulate situations, think of answers, or, more broadly, enhances the ability to be intelligent.

Using AI in the current design, art, or engineering modeling, the results can be magnificent. 

From Parametrical Design to Generative Design

The starting point is to change our relationship with computers. For decades, we have used them as tools for the development of projects, but they were nothing more than instruments until then. Generative project systems, in turn, take computers to another level. Today, they are becoming project partners, co-creators, thanks to their enormous capacity to handle large volumes of data and thus enable generative approaches. They are the cloud, big data, and artificial intelligence reaching the drawing boards, and bringing surprising results. Generative project systems, unlike their application with parametric and algorithmic technologies, are not new.

The novelty is that today we have instruments with sufficient processing capacity to make this approach a concrete reality and a viable alternative for project development.

Generative Design Applications

NASA Generative Design

American agency will use Autodesk’s autonomous design tool to create new solutions and manufacturing processes for use in space and why not in other planets.

Autodesk announced a collaborative research project with NASA JPL (Jet Propulsion Lab), intending to explore new techniques and solutions for the development of projects and manufacturing processes aimed at space exploration.

Generative Design is a technology that Autodesk has been developing and refining for about eight years now and uses machine intelligence and cloud computing to quickly generate a wide range of design solutions that fit the specific constraints defined by designers, designers, and engineers.

The most impressive thing about this process is that these options are created with little or no human intervention: machines creating machines.

For the time being, the application of generative design is still considered an R&D project within JPL. Still, just as in the past, the processing capacity of mainframes has helped the space program to reach new heights, it is believed that new technologies such as design generative will create new possibilities in space exploration, allowing us to go further and learn more about our place in the universe.

Generative Design: 3D printing and artificial intelligence help to create lighter vehicles

Using several technologies, such as cloud computing, 3D printing, and algorithms based on artificial intelligence (AI) the automaker General Motors along with Autodesk is working on the use generative design software, which quickly explores multiple combinations in part design, generating hundreds of options of geometries focused on high performance, often with unique shapes, based on user-defined goals and parameters, such as weight, material strength, manufacturing technique, and more. The user then determines the best design option for the part.

The partnership applied this new technology to produce a proof-of-concept piece on bench support. By exploring many options, they achieved an alternative that is 40% lighter and 20% more powerful than the original part. It also consolidates eight different components into one 3D printed part.

This is a good example of the result using generative design with additive manufacturing process: having the best product management, using a minor quantity of feedstock and also, leaving no waste. Cool, isn’t it?

How Generative Design could help with Architectural Projects?

At an Autodesk University event in Las Vegas in 2018, industrial-scale 3D printing, robotic design, immersive engineering, and intelligent BIM advances were featured, providing a tempting impression of how the architecture and technology will be further integrated the not too distant future. The star of the event was the fundamental concept of Generative Design. Following the foundation of the evolutionary design structures found in nature, the generative design can alter the built environment beyond recognition, and technology leaders are more excited than ever. Generative design can answer questions such as: “What form of structure would generate the most efficient bridge between two columns?” Or “What is the thinnest slab possible to overcome this gap”?

The event also illustrated more complex problems that can now be explored, like challenges related to comfort conditions and needs program. In one of the examples at the event, it displayed a scale model of a hospital, where different parameters for the aspired flow of people, light conditions, temperature, and even atmosphere could be introduced, producing thousands of possible Layouts for each wing of the hospital. This type of software illuminates generative design’s potential to help solve problems far beyond form and structure concerns.

Combining this technology with the advances in the 3D printer on an industrial and architectural scale can fundamentally change the way we think about architecture. Instead of designing building envelopes composed of separate layers for heating, ventilation, passive solar gain, and other needs, it would be possible to explain within a complex “skin” that has qualities that mimic biological organisms.

The possibilities are surprising. Water and gas could flow like “human-made veins” integrated into the walls making pipes a thing of the past. Solar data can generate transformable windows that allow light to enter the interiors at precisely the right times, according to the needs of the inhabitants. With that, blinds would be banned forever. The internal structure of the metal and carbon fiber elements will be optimized, allowing even larger spaces without columns, narrow stairs, and the most dramatic cantilever structures ever created.

Airbus Planes Case

The manufacturer has designed a new cabin division for its A320 aircraft. The goal was to reduce its weight and meet structural performance requirements to meet industry regulations. The software generated more than 10,000 design permutations for the partition, presenting them in a visual graph that made it easier to filter, explore, analyze the compensations, and, finally, select an alternative. The selected design is 45% (66 lbs) lighter than the current designs. Airbus estimates that the new design approach could have an economy of 465,000 metric tons of C02 emissions per year, the equivalent of taking about 96,000 passenger cars off the road for a year. Airbus won a Manufacturing Leadership Award in 2018 for this project.

Challenges and Opportunities of the Generative Design

However, like any new technology or something like that, at first, it sounds suspicious. There are some challenges to the acceptance of many stakeholders. For today’s product development leaders, generative design technologies present several cultural, organizational, and competitive challenges.

Probably, one of the first obstacles is a technological one: generative algorithms produce designs that can be radically different from those designed by humans. Machines taking human service might be considered “strange” or problematic, making it difficult for internal stakeholders to accept generative solutions, even though the proposed designs are technically superior. 

The use of generative design for parts aimed at end customers presents similar challenges. However, some companies are already capitalizing on the approach to create products with a unique and highly differentiated appearance.

The second major challenge is cultural. Adopting large-scale generative approaches could alter the company’s requirements in terms of human resources on the recruiting and selection process, changing the conventional talent, know-how, and resources in the product development function. For example, generative solutions can involve less time from experienced engineers and designers, allowing for shorter development cycles. This characteristic raises the question of organizational design and the allocation of resources for established players, which can be dangerous, thinking about new competitor’s entrance on the market.

The third set of issues concerns the integration of processes. Companies should consider how generative approaches will be integrated with existing engineering processes, data platforms, and toolkits. The rapid development of generative design technologies greatly affects the flexibility of companies that are likely to need to use different tools from different suppliers, changing and updating their design tools as the technology evolves. This will require open and adaptable systems and a high level of IT functions and product development.

Additive Manufacturing and CAD Software on Generative Design 

The term Additive Manufacturing represents a group of digital manufacturing technologies, which are capable of creating physical objects from a digital model.

As a common feature, all technologies work by adding layers of material, one under the other, until they form the final object. Most of all, it requires modeling software, also known as CAD. 

As we previously said, the generative design uses 3D modeling technologies to print a certain project. There is when SolidFace comes in.

If you want to know more about additive manufacturing, we invite you to read our blog post, especially about it.

SolidFace 3D: a flexible and powerful generative design ally

Let’s talk about SolidFace a little bit. First of all, SolidFace is a Free 3D modeling software. In other words, it is an excellent opportunity to, from students to skilled modeling professionals, discover this amazing tool. SolidFace has a lot of features that fit exactly what it needs in the generative design process. Check it out!

Teamwork Collaboration always in real-time: complicated designs require consistent. That is why SolidFace has a teamwork collaboration space. Having an upfront view before customer feedback can save up on materials, energy, and unnecessary repeated processes. Distributing design data efficiently and gathering feedback in real-time is not possible with your conventional CAD. Have the versatility to store your designs in the cloud or central location, granting instant access to the updated project data. Real-time design reviews, commenting, and simultaneous editing allows an accurate collaborative workflow in which everyone’s inputted information can be obtained, recorded, and implemented into the final design.

Flexibility in the Design: Our multi-purpose solution allows you to design with the flexibility to change or incorporate different configurations quickly and reduce development costs. For flexibility in design to be successful, it needs the insights of your most skilled engineers. Empower every engineer, user, or stakeholder to explore new design scenarios, using the right data, without changing each other’s work. All edit and every design change are recorded, so your partners can experiment with confidence. Errors and bad design decisions are easily undone. At the same time, the most helpful ideas can be joined back into the main design.

Product Quality and Accuracy: Quality is a decision indicator for many customers when choosing a supplier; value is meaningless if the quality of the end product is compromised. Unplanned downtime can cost you thousands of dollars per minute. Accuracy and attention to detail are crucial factors when designing a new product. SolidFace integrated modules allow related parts and subassemblies to be designed together. They take design intent and secure that every part updates correctly when changes occur. Design data is always up recent, so when another user makes a change, the entire team instantly knows about it. Errors are decreased, and every part fits together perfectly – at the first trial.

Reducing Time to Market: Providing a prototype is just the first step. Taking into consideration feedback to find the final form, fit, and function that your team matches require multiple design repetitions in the shortest amount of time. With technological innovations evolving constantly, it’s no surprise that speed is a valuable virtue to market.

Conclusion

Seeing all the examples, we realized that generative design could be a revolutionary step in many market areas like construction and automobilists production. 

While an algorithm is still just an algorithm, only a human decides which problem to solve, which objectives must be achieved, and which factors are most important to solve a problem. Computers can help organize and prioritize those decisions, but they can’t make them. Only people can decide what is important. The generative design gives architects, engineers, and builders new freedom to design and create a better world.