Sophia Barnett, Content and Technical Writer

Artificial intelligence (AI) is rapidly transforming the hardware industry, from streamlining component procurement to revolutionizing how we design and build the gadgets that shape our lives. The responsibility of repetitive chores such as circuit layout, material selection, and endless testing can be shifted to AI, freeing engineers for more creative pursuits and significantly accelerating the development cycle. 

Duro recently hosted a webinar to discuss how AI can benefit hardware engineers, today and in the near future. We chatted with Cofactr’s Co-founder, Phillip Gulley to get his perspective on how AI is changing component procurement, as they’re at the forefront of its transformation. Cofactr’s platform handles everything from sourcing and purchasing electronic components to managing inventory, kitting parts, and shipping them to manufacturing. The company is incorporating AI to clean and standardize data and generate actionable insights on component procurement. 

This article summarizes some of the key discussion points in the webinar, such as how AI is making sourcing more predictable as well as some future predictions. If you missed the webinar, watch the recording on YouTube. 

The current state of hardware procurement

The global hardware supply chain, scarred by the early days of the Covid-19 pandemic, has undergone a significant transformation. Initial shortages and scrambling for essential components exposed critical vulnerabilities: limited visibility, outdated bills of materials, and inflexible sourcing strategies.

Additionally, traditional hardware procurement has been fixated solely on component quantity and cannot survive under the weight of today’s market stressors. Legacy systems, blind to intricate supplier networks, negotiated contracts, and compatibility nuances offer false hope in easy component substitutions. Available alternatives often harbor hidden costs, performance compromises, or integration nightmares. Moreover, sudden part discontinuations and supply chain disruptions can wreak havoc on projects, rendering reactive sourcing scrambles both costly and ineffective. 

In response to these challenges, manufacturers are taking a look at their sourcing processes, focusing on improving the following areas:

  1. Ensuring BOMs are procurable in product development stages. New product introduction (NPI) processes prioritize readily available parts, diversified sourcing, and shortened supply chains. This shift aims to build resilience and mitigate future disruptions.
  2. Addressing ongoing challenges with long-term supply forecasting Manufacturers are looking for ways to predict longer-term forecasting.

There’s an opportunity for AI to enhance both of these processes. The right tools can provide hardware companies with data-driven insights, adaptable predictions, and intelligent sourcing recommendations.

How AI can help navigate the procurement landscape

AI revolutionizes modern supply chains by utilizing diverse data sources to offer comprehensive insights into component availability and lead times. Predictive analytics enable companies to proactively adjust production schedules and identify alternative components, ensuring product continuity and minimizing delays. Responsible implementation, emphasizing data quality and human oversight, is crucial to address biases and maintain transparency.

AI-powered predictions and substitutions 

Proactive foresight in the dynamic realm of lead times is essential, and AI, powered by real-time data, excels in predicting fluctuations and recommending reliable alternatives. However, responsible data management, human oversight, and transparent communication are vital to navigate the complexities of AI-driven procurement successfully. The shift from quantity-focused procurement to data-driven agility requires engineers to anticipate disruptions intelligently and source alternatives with precision and foresight.

Using AI to make supply chain predictions

AI models are starting to revolutionize supply chain forecasting by predicting future days of supply based on real-time market data and historical insights. Here’s how it works:

  1. Data Gathering: Real-time information on inventory levels, supplier lead times, demand fluctuations, and market trends is continuously collected from diverse sources like websites, reports, and even sensor data. Historical data on past orders, production schedules, and disruptions adds context and reveals patterns.
  2. Model Training: Sophisticated AI models can then analyze this data to identify relationships and learn how past factors like demand changes or unexpected events impacted previous supply levels. This training equips them to recognize leading indicators and develop predictive capabilities.
  3. Forecasting Future Supply: The trained models analyze real-time data to predict future supply for specific components. They consider current inventory, recent orders, supplier lead times, and potential disruptions based on market trends and news events. The output includes not just a single prediction, but a range of possibilities with associated confidence levels, providing a realistic picture of potential supply scenarios.
  4. Actionable Insights: These AI-powered predictions are presented clearly through dashboards and visualizations, enabling businesses to quickly understand the predicted supply situation for critical components. Based on these insights, proactive decisions can be made to mitigate potential shortages, optimize inventory levels, or identify alternative suppliers.
  5. Continuous Improvement: As new data emerges, AI models continuously learn and adapt, refining their forecasting accuracy over time. They can be retrained on new situations and unexpected events, enhancing their capabilities and resilience in a dynamic market.

AI-powered supply forecasting using real-time data empowers businesses to gain valuable insights, make informed decisions, and build more resilient supply chains in an ever-changing market. It’s not magic, it’s data-driven intelligence making the future of supply chains clearer, one prediction at a time.

How AI will enhance PLM solutions 

On the other hand, generative AI’s impact on PLM extends far beyond suggesting spare parts. Over the coming months expect to see AI used to optimize designs, predict failures, automate tasks, and personalize sourcing recommendations. This AI infusion will transform PLM from a source of truth into a catalyst for innovation, efficiency, and data-driven decision-making. While challenges like data quality and human-AI collaboration need addressing, embracing this technology promises to revolutionize how we design, develop, and manage products.

Addressing the risks of AI in procurement

Predictions may not always be correct

Although AI offers a myriad of benefits with improving procurement processes, AI is still largely making projections on future scenarios. Therefore, it should be viewed as a tool to help, not as a human replacement. Anyone utilizing AI in a deterministic environment, whether in manufacturing or assembly management, can use it as a tool to strategize, but not as a replacement for research or testing. 

Hallucinations can occur

Additionally, AI systems occasionally produce erroneous outcomes. The phenomenon of hallucinations, where AI algorithms generate incorrect or misleading information is common. Hallucinations can manifest in several ways, including incorrect predictions, false positives, false negatives, sentence contradictions, prompt contradictions, factual contradictions, and irrelevant or random hallucinations.

Don’t assume that AI is always right

When training your AI model, ensure that regularization, accurate task-relevant data, templates, and specific prompts are used. This approach guarantees that your AI model is optimally configured for successful–and factually correct–outcomes.

Although AI is a convenient avenue for expediting processes, due diligence is still required. Tracking and researching the market–whether in the role of an engineer or a procurement lead–requires ongoing attention and effort. Engineers must conduct hands-on, physical testing of components to verify authenticity and uphold quality standards. We are far from reaching a point where these responsibilities can be entirely delegated to AI.

The key takeaways

In summary, AI’s integration into component procurement is reshaping the hardware industry, streamlining processes, accelerating development cycles, and fortifying supply chains. Our webinar featuring Cofactr highlighted real-world applications, showcasing AI’s capacity to navigate modern supply chain complexities effectively. While it won’t replace humans, we expect to see AI playing a pivotal role in:

  • Shaping long-term supply availability forecasting
  • Offering data-driven predictions about stock
  • Recommending sourcing options as alternates
  • Ensuring data quality
  • Compensating for human oversight

Watch the webinar to gain deeper insights into AI’s transformative role and start using data now to propel the industry into a future defined by efficiency, creativity, and adaptability.