As part of our miniseries on future trends in PCB assembly design, this blog post looks at the benefits and challenges of artificial intelligence (AI) in PCB design and assembly.
In this post we will cover:
- AI’s current role in PCB assembly, whilst outlining its limitations
- MPE’s integration of AI in PCB assembly, and how this positively impacts your electronic products
- The numerous ways in which AI has already boosted efficiency in PCB assembly
- Key challenges of adopting AI-driven models in the electronics manufacturing industry
| To learn more about MPE’s PCB Future Trends mini series click here |
How AI is already making an impact in PCB assembly
Artificial Intelligence plays a pivotal role in modern Printed Circuit Board Assembly (PCBA) design and assembly processes, revolutionising the electronics industry. AI is utilised for design, testing, inspection, routing and component placement.
But there are questions surrounding AI’s ability to perform autonomously, for instance can AI design an electronic circuit? While AI can significantly enhance various aspects of PCBA design, the answer to this question is complex. Current AI tools in electronics and electronic design can assist human designers to manufacture efficient circuits, but full autonomy in design is yet to be achieved.
Likewise, there is increased interest in what AI tools are available for electronic design. Several AI tools exist on the market for electronic design, including AI-enhanced electronic design automation (EDA) software, such as Synopsys, and browser-based, end-to-end electronic design tools, such as Flux AI. As with all tools, there are pros and cons. Pros include improved design efficiency, faster design validation and reduced time-to-market. However, potential drawbacks may include a learning curve for users, as well as the need for thorough validation of AI-generated designs to ensure they meet safety and reliability standards. The effectiveness of AI tools in the market depends on the specific requirements of the project and the expertise of the design team.
How MPE has integrated AI to improve our PCB assembly services
“MPE is dedicated to delivering high-quality electronic products swiftly and efficiently to the market. Through ongoing investments in cutting-edge technology and the development of our people, we prioritise achieving excellence. The profound impact of AI within our operations is evident in the remarkable gains of speed and accuracy it brings to PCB assembly, ultimately enhancing the experience for our valued customers.”
Nicola Evans, Managing Director, MPE Electronics
MPE uses AI for PCBA inspection purposes. An Automated Optical Inspection (AOI) machine is a piece of equipment used to inspect bare PCBs and PCB assemblies for defects using cameras, lighting and precision mechanics to obtain high-resolution images of the board.
By integrating AOI with AI, MPE is able to expand the machine’s inspection capabilities through machine learning. While conventional AOI relies on geometric algorithms guided by a set of predetermined rules, which are adept at identifying known defects, but less effective in finding unknown defects, this limitation often results in a higher number of products requiring manual human inspection.
AI is also used in MPE’s Surface Mount assembly, or Surface Mount Technology (SMT) service. This involves using a ‘pick and place machine’, an automated manufacturing tool that uses robotic arms and advanced vision systems to precisely pick up electronic components from a supply source and accurately place or mount them onto a printed circuit board. The newest machine in MPE’s equipment inventory, the Mycronic MY300DX, can place up to 40,000 components an hour. Integrating AI allows MPE’s engineers to plan the best routing on a single job, resulting in less travel time between picking up a new component, placing it and moving onto the next. AI can also plan the order in which to build jobs to minimise downtime and set-up, maximising machine run time.
| Learn more about our SMT service and its many benefits for your electronic products here. |
Numerous benefits of using AI in PCB design and assembly
The adoption of AI in PCB design and assembly brings about significant improvements in efficiency, accuracy, and innovation, ultimately contributing to the development of more reliable and sophisticated electronic products.
Efficiency improvement
AI automates time-consuming and repetitive tasks involved in PCB assembly. AI-driven routing algorithms optimise the placement of components on the PCB, considering factors such as signal integrity and thermal management. These algorithms use data-driven insights and learning from previous designs to automate and improve the process of determining the paths that electrical signals should take between various components on the PCB. Without AI, PCBA routing involves manually designing the pathways, resulting in longer design cycles and time-to-market for electronic products.
Improved testing and inspection reduces error
AI algorithms can analyse vast datasets, expediting the design validation process by quickly assessing the functionality and performance of the PCBA design, identifying design flaws and errors, and simultaneously suggesting improvements early in the design process. Automated testing and inspection processes leverage AI to identify defects and anomalies with greater accuracy and speed than traditional methods, such as visual inspection or manual probe testing. This decreases the number of defects and enhances the reliability of each circuit, reducing the need for manual inspection.
Cost reduction
Using AI to automate the PCBA design process reduces the costs associated with labour, likelihood of numerous design iterations and potential rework to correct errors. AI-driven optimisation can also lead to more energy-efficient designs by considering power distribution and consumption factors during the routing and placement phases. The gains in increased efficiency results in more cost-effective PCBA manufacturing.
Design innovation and adaptability
AI tools, such as generative design algorithms, can suggest innovative design alternatives based on specified constraints. This can lead to the discovery of novel and optimised solutions that might not be apparent through traditional design methods. AI systems can adapt to changes in design requirements more quickly than manual methods. This adaptability is crucial in dynamic design environments where modifications are common. Moreover, AI systems can learn from past designs, making continuous improvements over time. This learning capability contributes to a cycle of refinement and optimisation in PCBA design practices.
Predictive analysis for better performance
AI can predict potential issues and challenges in the manufacturing and operation of PCBAs, allowing designers to address problems proactively and optimise designs for better performance reliability.
Key challenges of adopting AI
Significant initial investment
Integrating AI systems into PCBA design requires investment in specialised hardware, software and training or new personnel. In terms of the manufacturing industry at large, some companies are hesitant to adopt AI owing to potential risks of mismanaging the new technology, leading to further costs.
Absence of internal expertise
Specialised skills and expertise are essential when it comes to developing and integrating AI into PCBA design and assembly. Small to medium-sized contract electronic manufacturers may struggle to fulfil this capability and may need to hire external AI consultants also fluent in electronics engineering. Machine learning technologies are constantly advancing and even electronics manufacturers with capable in-house data scientists must keep up-to-date with advancing AI. Moreover, without a clear understanding of the AI adoption strategy at a company-wide level, with committed leaders and accessible advocates, staff are less likely to embrace this new technology to its fullest potential.
Ensuring accurate and complete data
The quantity and quality of training data is critical to the success of AI. The electronic manufacturing industry produces a large amount of complex data. If the data should prove to be insufficient or inaccurate, the AI model is likely to underperform, resulting in suboptimal assembly of PCBs.
Regulations, Transparency and Security are a cause for concern
AI models, particularly deep learning algorithms, are often described as ‘black boxes’, where the decision-making process is not transparent. In industries where safety-critical applications require clear accountability, as well as regulatory compliance, such as aerospace or medical devices, this unambiguousness is a challenge. Verification and validation processes must be prioritised in order to comply.
AI enhances human capabilities in PCB design and assembly
Despite ongoing challenges and associated hesitations, the adoption of AI in electronics production has already contributed to the radical innovation of the industry.
According to global forecasts, the value of AI in the manufacturing industry is expected to reach US$20.8 billion by 2028 owing to the increased availability of big data and rising industrial automation.
Making huge contributions to the improvement of productivity, quality and reducing trial and error, AI is enhancing human capabilities in PCB assembly to provide an improved service and superior electronic products for the end-user.
MPE Electronics is an established and experienced contract electronics manufacturer specialising in PCB assemblies and full box build assembly for a wide range of commercial and industrial businesses.
To find out how MPE Electronics’ PCB manufacturing and assembly services can benefit your business, contact our expert and friendly team on +44 (0)1825 764822 or enquiries@mpe-electronics.co.uk
Artificial Intelligence in Manufacturing Market, Sept 2023, Markets and Markets
