Whether it’s a Tesla sedan or the latest iPhone, the application of Artificial Intelligence (AI) is pervasive across the products of our time. However, it’s not just the products but also the manufacturing process that creates them that has evolved thanks to AI. ‘Industry 4.0’ – a blend of AI algorithms, Internet of Things (IoT) devices, and advanced engineering – holds tremendous potential for manufacturers in diverse industries. In this post, we explore 3 ways that AI in manufacturing is driving transformation in the manufacturing industry.
Three Ways that AI in Manufacturing is Transforming the Industry
1. Quality Inspection: Finding the needle in a haystack
One of the major challenges of manufacturing is detecting defective raw materials or components in the production process. The further defective components move along the production line, the greater the risk to manufacturers and customers. In 2015, defective airbags in passenger cars led to the recall of over 60 million vehicles and is estimated to have cost the automotive industry $25 billion.
In most industries, ensuring a high degree of quality control measures is an expensive and time- consuming operation. When companies neglect quality control measures, however, it can lead to recalls, safety concerns, and dent the reputation of the firm with customers. According to the American Society for Quality, businesses lose 10-15% of annual revenue due to defective products.
In recent years, AI algorithms like deep neural networks have been successfully deployed to streamline quality control in manufacturing. For example, IT firm Capgemini shares the case of a large food processing company that needed to ensure the quality of eggs in its production line. Between 30,000 to 270,000 eggs were processed per hour, making any manual and human-intensive approach expensive and prone to errors. However, by implementing AI in manufacturing of the egg processing, each egg could be inspected in just 40 milliseconds. This reduced the cost of quality control while maintaining the speed of production.
2. Higher Asset Utilization Through Predictive Maintenance
For manufacturing to be efficient and profitable, firms must make the optimum use of machinery and production equipment over their lifetime. Machines also require downtime for repair and maintenance. According to the International Society of Automation, such downtime results in a revenue loss of $647 billion annually. Despite the downtime for preventive maintenance, 89% of machine failures occur at random, according to IBM.
By integrating AI in manufacturing with IoT sensors in machines, firms can now improve the efficiency of their production equipment. Using predictive AI algorithms, firms can detect which machines need maintenance and when they will need it. This reduces idle time, helps in planning resources, and streamlines the production schedule. It also reduces wastage and lowers the probability of defective products that could have come from an unmaintained machine. For example, General Motors leveraged AI and cloud technology to detect component failures in 7000 robots at its factory. The AI-powered system detected 72 instances in these robots and helped reduce unplanned downtime.
3. Helping Workers Stay Safe in the Manufacturing Industry
According to the International Labour Organization (ILO), there are 340 million occupational accidents every year. Further, the ILO estimates that 4 percent of global GDP is forgone due to lost workdays and accidents. Apart from the physical and economic costs, accidents take a toll on the emotional well-being of workers and their families.
AI in manufacturing helps manufacturers take care of their most important asset- their workers. This has become even more critical with recurring COVID waves, the need for social distancing, and restrictions on the number of workers available at a given time. Using wearable technology, companies can ensure that workers are healthy and safe on the factory floor. For example, a wearable device can monitor the temperature and oxygen levels of mineworkers while IoT sensors detect the level of gases in the environment. This helps supervisors to assess risks in real-time and the best possible course of action in different scenarios.
Another way that AI in manufacturing helps increase the safety of workers is by ensuring personal protective equipment (PPE) compliance. In the United Kingdom alone, non-compliance with PPE requirements cost businesses 79 billion pounds in 2018. AI-enabled systems using machine-vision and video-capturing equipment can detect if workers comply with PPE requirements at the factory or construction site. For example, if a worker forgets to wear a hard hat, the AI system provides alerts to the worker and their supervisors. It can also restrict access to certain sites based on PPE compliance. This reduces the risk of worker accidents.
Conclusion
AI is proving to be a game-changer in the manufacturing industry. McKinsey estimates that AI can potentially unlock $1 trillion to $2 trillion in supply-chain and manufacturing across global businesses. In many industries, integrating AI into the manufacturing process will require large capital investment at the front end. However, over the long-run, such investments increase the life of equipment, reduce wastage, create better quality products for customers, and help keep workers safe.
If you would like to explore how Machine Learning and AI deliver strong results for your business, please connect with us AnyML.
