Upskilling Today’s Workforce to Use AI Effectively

We’ve all seen the hype and excitement surrounding advancements in artificial intelligence (AI). But are today’s traditional digital engineering tools, business processes, and personnel actually ready for AI? Often enough, the answer seems to be “not yet.” Many companies face challenges as they attempt to implement AI in their product lifecycle management (PLM) tech stack in a way that provides real value. Many of these challenges might relate to how today’s workforce leverages AI.

There is no doubt that AI has the potential to significantly impact the workforce, saving time, generating new concepts to pursue, and increasing the number of actionable insights available to inform decisions. We shouldn’t underestimate AI’s impact as a partner in engineering for product design.

AI is also prone to mistakes. As a result, it’s just not realistic to remove the human element from the equation. To gain true value from emerging technologies, companies must position AI as an enabler of the engineering process rather than a replacement for engineers.

AI use cases for product-driven companies

To leverage AI to bolster the engineering process, you must invest in AI processes and tools that eliminate repetitive, non-value-added work. This approach ensures that workers invest their time in more creative tasks that humans prefer to do while letting AI do the repetitive work as an advisor in the background.

Over the next few years, we’ll see more and more use cases reinforcing the idea of an AI-supported human-centric engineering process. Here are three examples we see today where organizations are leveraging AI to shape engineering practices for the better:

AI-driven design

Today, we see human-driven activities assisted by human-prompted AI. However, the key to successfully setting this up is arming the AI with the correct training data. Product data organized strategically, such as within a proper digital thread or digital twin, can help the AI tool provide real value to the humans doing creative work. Then, we can ask the tool well-formed questions and get reliable answers based entirely on the data living within the organization. Plus, when plugged into powerful connectivity solutions like digital thread and digital twins, AI can also monitor these platforms for changes that impact design and provide suggestions for optimizing engineering work.

We also see emerging use cases that will enable AI-led interaction with systems — all guided by human-prompted suggestions. Once these technologies become more sophisticated, engineers can prompt chatbots to perform complex tasks, analyze the results, and choose the best designs for additional refinement. While humans will still have control over the solution space, AI will do the tedious work of testing out possible combinations and optimizing for valuable factors such as cost, sustainability, and reliability.

Workforce augmentation

Many organizations face challenges in setting up feedback cycles to improve product quality. Often, quality problems get lost in translation and never get reported back to engineering. When integrated with product lifecycle management and quality reporting, AI tools can parse natural language descriptions of problems and automatically direct issues to the appropriate development teams. It can then be further leveraged to detect anomalies and recommend solutions.

Virtual Assistant for change control

In addition, AI can be trained as a virtual assistant to support meeting planning, task organization, and approval workflows, including updating customer-facing, manufacturing, and service documentation. In these use cases, the virtual assistant could even detect patterns pointing to a larger issue and then use generative AI to schedule a meeting with the appropriate agenda. It would invite all relevant team members, create a report that documents the impact of change, and generate most of the recommended deliverables and tasks to correct the issue. Once a problem is reported, AI can create a proposed solution, auto-schedule a meeting, and provide supporting documentation.

It’s important to note that all these use cases require human interaction to evaluate and finalize decisions. To make AI a successful investment for your organization, everyone must know how to participate in the process and feel comfortable letting the tools take over at certain points. They must also be trained on the fact that AI hallucinations are real, in which the tool creates an answer to a question that sounds good but is actually incorrect. (AI will never say I don’t know – it will always provide an answer, correct or not.) Organizations also need to monitor and ensure their AI implementations have proper guardrails implemented. Documenting process steps is critical, ensuring full transparency into how the AI solution works.

Best practices for integrating AI into PLM processes

As you look to integrate AI tool(s) into your PLM processes, here are a few considerations to keep in mind:

Start with the data

AI-infused engineering requires the right kind of data. The more data, the better—especially well-integrated and connected data from across the product lifecycle. The greater the quantity and quality of data AI has to work from, the better results it will generate. We’re seeing new ways of tracking and visualizing thread-connected information in the industry, contributing to stronger AI-driven strategies and helping us see problems from new angles.

Embrace the idea that rote work can be done by a machine

Sometimes, it’s hard to conceptualize the future, especially when it involves complex ideas that only the experts in a particular discipline truly understand. While some may have valid concerns about AI, it’s important to make a mental shift and trust that it can handle certain kinds of repetitive work, such as synthesizing information from complex databases, making updates to corresponding documents when a change occurs somewhere in the pipeline, and more. We can move forward confidently by trusting but verifying.

Verify the data and assumptions proposed by AI

It is essential to build guardrails for AI tools and authenticate their findings. AI may only introduce partial solutions based on the models it’s using, so it takes human intellect to interpret the information, direct additional analysis, and conclude what is best for the situation. Experienced, knowledgeable engineers and their counterparts must be the decision-makers in the process.

How Aras Innovator plays a role in AI innovation

As companies consider using emerging AI technology to extend their product data and offer engineers more value, Aras Innovator® can help by enabling an end-to-end digital thread of data. Aras Innovator connects digital assets throughout the product lifecycle and builds a solid foundation for data analytics and AI initiatives.

To dive deeper into the world of AI in PLM, tune into an on-demand Aras webinar, Bring the Power of Data and AI to Managing Compliance. It features PLM experts from across the industry covering how AI can keep tabs on shifting compliance requirements and bring value to the engineering teams that must pivot with these changes.