AI is transforming numerous industries — but for many mechanical engineers, its role still feels uncertain. Is it a productivity booster, a passing trend, or a threat to job security? This series explores how AI is being integrated into mechanical design and manufacturing, beginning with what it is, what it does, and why it's worth understanding now.
Reasons for skepticism of AI include concerns about job security, data security, data ownership, and reliability of AI output. While these concerns are legitimate, they can be addressed effectively with education and training. The potential benefits of AI are monumental — capable of propelling the performance of organizations and individuals to new heights. By viewing AI as a partner, rather than a threat, mechanical engineers can be at the forefront of reaping the benefits of AI.
Note: This is the first in a series of articles on artificial intelligence for mechanical engineers.
AI includes machine learning, automation, and more. How will you use this technology in the future? Image source: everythingpossible/stock.adobe.com.
What is AI?
Definitions of AI vary, but in general, AI is an umbrella term for strategies and techniques, including rule-based algorithms, that allow machines to mimic human behavior and intelligence. Given a set of human-defined objectives, AI can make predictions, recommendations, and decisions much as humans would, given the same objectives and a relevant knowledge base.
CAD programs have been incorporating AI technology into productivity-boosting tools, such as the SOLIDWORKS Design Assistant, which learns how users work and offers suggestions based on particular workflows. Image source: SOLIDWORKS.
Under the AI umbrella is machine learning (ML), which uses algorithms and statistical methods to allow machines to “learn” from data and improve their performance without explicit programming. Applications of machine learning include product recommendations on retail websites, automatic email sorting and spam filtering, and certain automated or predictive features in CAD software such as SOLIDWORKS.
Deep learning, a more advanced form of machine learning, uses large multi-layered neural networks to learn patterns from vast amounts of data and potentially create new outcomes a human may not have considered. Deep learning is used in widely available tools such as ChatGPT and Google Gemini (formerly Bard), virtual assistants such as Apple’s Siri or Amazon’s Alexa, and advanced graphics technologies such as the Denoiser in SOLIDWORKS Visualize.
The Denoiser in SOLIDWORKS Visualize uses machine learning to filter out noise from unfinished and “noisy” images. The image on the left shows typical results with Denoiser off, and the image on the right shows results with Denoiser on. Image source: SOLIDWORKS.
AI can be categorized in different ways, and opinions vary on how to subdivide the technology. A subset called generative AI looks promising for design engineering and technical applications. Generative AI learns patterns from existing data, then uses this knowledge to generate output, such as text, graphics, 3D geometry, computational results, and other similar data. Some familiar generative AI tools include OpenAI’s ChatGPT for text generation and Dall-E for image generation, Google’s Gemini for text and image generation, Adobe’s Firefly for image generation, and Microsoft and OpenAI’s GitHub Copilot for code generation.
Other noteworthy types of AI include physical AI, agentic AI, and predictive AI:
- Physical AI enables autonomous systems such as robots, self-driving cars, and other devices to understand and perform complex actions in the real (physical) world.
- Agentic AI uses sophisticated reasoning and iterative planning to autonomously achieve goals and solve complex problems with minimal human guidance. Still in relative infancy, agentic AI develops solutions based on large language models (LLMs), which are models trained on a vast amount of textual data. The end user provides an AI “agent” with natural-language prompts, which are used to develop solutions based on available data.
- Predictive AI uses statistical analysis and ML to identify patterns, anticipate behaviors, and forecast upcoming events. In product design, it can inform decision-making by analyzing user interaction data, simulating performance under various conditions, and optimizing design iterations based on projected usability and market response.
Benefits of AI
AI can provide numerous benefits to mechanical engineers, such as time savings, financial benefits, productivity gains, and quality improvements. By handling repetitive tasks, AI can free up humans for more creative and critical thinking. “With AI, the mechanical engineer will spend less time doing those mundane tasks that can be relegated to AI and focus more on high-level engineering,” said Dr. Shrikant Savant, Data Analysis & Science Director, Director and Engineering Manager with SOLIDWORKS. As an example, he cited how AI can automate finding parts in a catalog based on specified dimensions and other properties.
AI can help automate mundane tasks such as looking up parts in a catalog based on object dimensions and shapes. Image source: SOLIDWORKS.
The time savings achieved with AI can also lead to financial benefits. As tedious tasks are accomplished with AI and automation, projects can be completed faster and production can be increased. Human resources can be reallocated to other productive areas such as refining designs and exploring new production capabilities.
In addition to time and money benefits, AI can also aid design processes and help generate design options. This concept, known as generative design, is still evolving but may produce significant benefits in the near future, according to Savant. “Design cycles might get shortened significantly. For example, if you were taking two weeks to design something, now AI can quickly suggest multiple design options for you and then you can choose one of the designs,” said Savant.
Using the Design Assistant found in SOLIDWORKS xDesign, mechanical engineers can boost productivity with a whole family of AI-based tools such as the Design Assistant for Edge Selection, which predicts and suggests other edges in the models you may want to select. Similarly, Sketch Helper recognizes existing sketch geometry in relation to surrounding geometry and suggests additional locations to place sketch entities. Command Prediction can predict and suggest the next command to execute, increasing productivity, reducing clutter from the user interface, and streamlining the overall workflow specific to that user.
Looking ahead, AI may soon help engineers accelerate tasks such as fixing broken parts. By uploading a photo of a broken part along with relevant data, engineers could prompt AI to generate an initial 3D design, which the designer could then refine and validate.
The Sketch Helper in the Design Assistant recognizes existing sketch geometry in relation to surrounding geometry and suggests additional locations to place entities. Image source: SOLIDWORKS.
AI can also improve product quality with new tools that identify defects in manufactured products. AI-driven quality inspection uses ML to automate defect detection with greater precision than manual methods, providing real-time insights for proactive quality management.
Think of AI as your assistant — humans will not be replaced in the design process. Image source: BullRun/stock.adobe.com.
Addressing Concerns
While engineering organizations see tangible benefits with AI, some still remain apprehensive, and not because they’re technology-averse. In fact, many skeptics are well-versed in advanced tools and workflows; they’re simply not convinced that AI is the panacea some envision.
Job security. One of the most common concerns is job security. Savant sees engineering jobs changing, but not being eliminated. “I don't think mechanical engineers will get replaced, but they will have to learn to employ AI to solve some of the problems like mundane tasks that normally they would spend time doing,” he said. “The importance of humans is not going to get eliminated.”
Data security is another area of concern. To protect data from unauthorized use, organizations using AI should establish security policies on data use, data ownership, and password protection. If using a public cloud for data storage, the cloud provider should meet standard industry best practices for data security, including data encryption and other security measures. SOLIDWORKS has a policy of not using customer data for AI purposes unless they have explicit permission from the customer to do so, according to Savant.
Reliability. Regarding reliability of AI results, data management processes should include validation of data input and output. An AI model is only as good as the underlying data, so all data should be from known, reliable sources. Data validation is needed to assess the input data for possible potential biases or flaws, and again to assess the output for reasonableness and accuracy.
“There is a lot of biased information on the Internet that all these LLMs and AI agents are trained on,” said Savant. “When you get any results, you cannot just take it for granted. Engineers should always use their common sense and judgment to check the results.” In some cases, a full analysis may also be needed to verify results, he noted.
Liability issues also need to be considered. Savant also feels human involvement is critical here. For example, if AI is used in simulation of multiple designs, the final design still needs verification. “It is still up to the engineer to either do a full simulation or full analysis, or do actual prototype testing. So I think that liability still remains with the designer and the design firm.”
Final Thoughts
As with any new technology, users may encounter challenges in adapting to AI technology. Traditional roles may change, and legitimate concerns must be addressed. But for mechanical engineers, the potential is undeniable: smarter workflows, faster design cycles, and more innovative manufacturing processes. By increasing familiarity with AI, engineers can embrace the possibilities and put AI to work where it’s most effective.
“AI is definitely not a fad,” explains Savant. “It is going to change the way you do business. If your company is not embracing AI, you're going to be left behind, because your competition is going to embrace AI. It is better that you find ways to embrace AI in your workflows and stay on top of the technology.”
The bottom line? AI is here to stay. Engineers who embrace it now will be better positioned to lead the next wave of innovation.
In our next article, we’ll review practical ways AI enhances mechanical engineering workflows.
This article was sponsored by SOLIDWORKS.
###
Searching for more information about Product Design & Manufacturing?
Click here!
Share This Post