The domain of engineering draws heavily from mathematics and its applications. Every engineer knows no matter how creative they become, in the end, the various applications of mathematics help shape their ideas. This very same relationship between engineering and mathematics enables
AI
, especially
GenAI
, to play a crucial role in modern-day product development and engineering design.
GenAI potentially has several important applications here: Innovative and novel materials generation, rapid prototype generation, efficient, optimised, and sustainable design generation.
Traditionally, some evident gaps in engineering design are high vehicle development time owing to high digital and physical validation, or the requirement to meet stringent regulations, under-utilisation of historical or cross-functional data, and loss of knowledge.
AI-powered design can go one level deeper by creating designs for complex machines that address three core capabilities:
At the component level: Where individual parts can be designed us ing GenAI. For instance, 3D design of a single component such as a bracket, a beam, a crossbar, or something similar. As the component design through AI is instant, this results in reduced component development time, sustainable design as the component is designed with optimum material, and reduced iteration with
CAE
.
At the sub-system level: Predicting multiple design parameters that affect system-level performance, such as all the different parts that can ensure improved functional performance. Predicting the design parameters at an early phase of development and adjusting those parameters in design would benefit in significant cost reduction involved in modifying the system in later stages.
At the system level: A prescriptive model that recommends iterations in product design to meet functional or regulatory requirements. For instance, if a product is not satisfying a functional parameter or regulation, GenAI can help predict which sub-systems and components need modification.
Benefits range from significant reduction in number of design iterations and thereby, in product development time. This will also result in considerable cost, effort, and material saving.
There is a fair chance that engineers can overor under-design components. By training the machine with optimum design parameters, the result will be closer to what is needed as a final iteration.
Expanding human possibilities
The design topics listed above have been cited to stress the way AI (and GenAI) is being practised today and the challenges faced by the product engineering community. An engineer knows that along with meeting all requirements within a given set of constraints, ingenuity plays a crucial role when designing and engineering a component or system.
Therefore, no matter how creative AI becomes, unless it starts developing an understanding of the principles of engineering, mathematics, and socio-economic ideas behind a design, it cannot be a replacement for human ingenuity and expertise. Designers remain essential in providing the critical thinking, aesthetics, and problem-solving skills as well as socio-economic context that shape the final design outcomes. GenAI in design is most effective when it collaborates with human designers, augmenting their abilities and expanding the possibilities.
On the horizon
Ethical issues around GenAI such as misinformation, plagiarism, copyright infringements cannot be ignored. But the benefits this technology can bring in the engineering space, in the safety net of a closed product development environment, is unprecedented. In 1950, British mathematician Alan Turing asked, “Can machines think?” and look where we are today.
At this point in 2023, AI has already reached a level where it can give an optimised design at the component level. Now is the time to ask – “In the next three to five years, can machines really design engineering systems on par with humans while addressing functional performance along with regulatory compliance?” The answer to this question is an emphatic yes.
“Can GenAI ever be able to introspect in case of a suboptimal design?” – This is a question for the AI community to answer.
Age of thinking machines
Technology has always been the driving force behind human progress, shaping the way we live, work, and interact with the world. In the 70s, the development of expert systems marked a significant step in the evolution of thinking machines. Since then, AI has become a focal point for the development of thinking machines and continues to evolve rapidly.
As we stand on the brink of a new era, several advancements are taking the world of tech by storm. Large language models (
LLMs
),
quantum
and GenAI are leading from the front.
English (US) ·