Artificial intelligence (AI) has been hotly debated in headlines and across industries. Is it safe? Good for business? The way of the future? While the debate rages on, one thing has become abundantly clear: AI is finding its place on the factory floor and will only continue to scale.
The reason? The rise of the fourth industrial revolution (Industry 4.0) has brought about significant opportunities for manufacturers to optimize processes, costs and streamline throughput. And manufacturers are tapping into the power of advanced technologies like AI and machine learning (ML) to reap these benefits.
The applications of AI and ML in electronics manufacturing in particular are numerous, from advanced predictions and quality assurance to waste reduction and more. There are several challenges AI and ML can help solve on the factory floor.
Challenges Faced by Electronics Manufacturers
As electronic devices and systems have become ubiquitous, and their production increasingly complex, manufacturers face several challenges, including:
- Inefficient Maintenance and Downtime: Manufacturing equipment requires frequent maintenance. Inefficient monitoring and updates can result in costly repairs and replacements. Additionally, inadequate visibility into operations makes it difficult to detect abnormalities and predict required maintenance, which can lead to unplanned downtime.
- Ensuring World Class Quality: Manufacturers must detect and address product defects and quality issues early in the product lifecycle without the lengthy process of human inspection to meet volume demands and avoid scrap.
- High Operational Costs: For products requiring vision inspection, a person cannot catch every imperfection in every product that goes through the production line. Unplanned interruptions from quality issues can cost significant amounts of money, cause delays in output and be disruptive to customer relationships.
Benefits of AI and ML in Electronics Manufacturing
The good news? AI and ML can play a role in tackling the challenges faced by electronics manufacturers. These advanced manufacturing technologies have quickly become essential tools for optimizing the manufacturing process. Some of the benefits of using these technologies include:
- Predictive Maintenance: Intelligent systems can be used to predict, detect and alert manufacturers to potential failures in their factory equipment. Predictive maintenance helps reduce downtime, improve reliability and increase efficiency.
- Quality Control: By using AI and ML algorithms, manufacturers can identify defects and quality issues in real time. This allows them to adjust the manufacturing process to reduce waste and increase quality assurance across production volumes.
- Process Optimization: By collecting and analyzing data from various sources and parts of the process, like production lines, sensors and equipment, manufacturers can optimize the manufacturing process faster and more efficiently.
How AI and ML is Being Used on the Factory Floor
To understand the full potential of using AI and ML, it can be beneficial to see how it is being employed in a real-world manufacturing operation.
Flex, a global electronics manufacturer, produces thousands of printed circuit boards (PCBs). They are crucial components in all electronic devices but have often remained reliant on human inspection for quality control. For Flex, product inspections play a vital role in producing flawless solutions.
Products that require visual inspection are traditionally inspected by human workers as they travel through the manufacturing line. Yet, as product demands and timeline speeds have increased, it becomes more difficult for the human eye to detect anomalies. To combat this challenge, Flex implemented an AI/ML-based defect detection system using deep neural networks to detect defects that are difficult to see with conventional vision systems or by human inspectors.
Flex’s new tool not only streamlined inspection processes, resulting in greater efficiency performance of over 30% and improved product yield by 97%, but also helped the company reduce scrap by identifying issues before sending a part to another step in the production line. It also helped optimize factory floor space, making room for other lines and solutions by eliminating legacy inspection stations.
Perhaps the most significant impact was on the inspection staff, who were given training in managing these innovative technologies. This increased morale and provided employees with advanced career opportunities, rather than focusing on the tedious process of inspecting products.
Lessons Learned from AI and ML Implementation
AI and ML technologies hold sheer unlimited promise and will continue to fundamentally change the manufacturing industry. But there is still much to learn about these powerful tools.
In production, Flex encountered countless lessons, such as:
- Data readiness: AI/ML models use algorithms to recognize patterns in data and learn from them, so the quality of an AI/ML model is only as good as the training data. This requires due diligence in making sure that the datasets conform to your requirements and that you’re applying the right analytical methods against that data.
- Ambiguous ROI: Quantifying the return on investment from AI and ML solutions is not easy. Organizations need to identify the right use cases for the business, find relevant data, process it and then develop, fine-tune, and eventually deploy models – all of which are vital and take time.
- Real-Time Decision Making: Time-crucial applications are related to safety issues, monitoring quality and more. But due to a common lack of system interoperability, the continuous cycle of “closed loop” updates and improvements can be a challenge. To overcome this challenge, companies need to implement digitization.
Optimizing Electronics Manufacturing with AI and ML
Manufacturers today are successfully leveraging AI and ML in their manufacturing processes, such as Flex’s AI/ML-powered vision inspection tool implementation. This enabled Flex to make the historically onerous and time-intensive PCB testing process more efficient and reliable.
While hurdles remain, advanced technologies like AI and ML are at the forefront of Industry 4.0 and have the power to transform production and operations on every level.
This article was written and contributed by Murad Kurwa and Rahul Katkar of Flex.