Reading time ( words)
The smart factory is starting to become a reality, as part of the overarching Industry 4.0 paradigm. With the technology enablers, such as industrial IoT (IIoT) and cloud computing, electronics manufacturing operational technology (OT) are on a converging course with traditional information technology (IT). Beyond the challenges of data acquisition and transformation, the true “proof in the pudding” is in the quick ROI from advanced analytics. This is where domain knowledge application into data science is paramount. We will share examples of successful, profitable implementation of applied machine learning (ML) in the electronics manufacturing line, where measurement science meets data science.
Industry 4.0 is becoming a hot topic in the world today. The first industrial revolution saw mechanization through water and steam power. The second industrial revolution introduced the ability to mass produce products using electricity.
It was almost a century before the third industrial revolution where computers became cost-effective and powerful enough to be adopted in the factories. Through the unprecedented processing and computational speed available then, the factory could automate more processes and provide better feedback on their processes.
Data has started to become the currency of productivity. Supply chain, commodity, customer relationships, quality, and production management could be harmonized on an enterprise-wide platform that empowered companies to optimize costs of manufacturing, reduce inventory, ship quicker, and improve quality.
Fast forward to today and the internet of things (IoT), believed by many industry experts to be the key driver of the next industrial revolution. The concept simply implies that machines will be intelligent enough to communicate and automate processes between themselves with minimal human intervention, creating a self-monitoring and self-sustaining internet of machines, for machines.
The Hype of Smart Factory
Industry 4.0 is a very wide paradigm that consists of many seemingly technological miracles. A lot of excitement and marketing have been made on big data, artificial intelligence (AI), augmented reality, additive manufacturing, robotics, and autonomous machines. However, one needs to appreciate that many of the core fundamentals for these technologies were developed decades ago. For example, AI goes as far back as 1956  and has been refined over the years. Nevertheless, it is believed that the implementation phase of these technology enablers truly begins now, as we have achieved the beginning of the right performance and cost of electrical, electronics, and wireless components that makes it all feasible and practical.
To read this entire article, which appeared in the August 2020 issue of SMT007 Magazine, click here.