Artificial Intelligence: What does it hold for the Manufacturing industry?

In the previous blog, we briefly discussed how AI is transforming the Banking, Financial Services, & Insurance (BFSI) industry. We talked about the use cases of artificial intelligence in BFSI sectors and how it is helping to mitigate the industry’s serious challenges. This blog discusses in detail the impact of Artificial Intelligence on another industry – manufacturing.

Artificial intelligence is one of the most significant drivers guiding the fourth industrial revolution or Industry 4.0. Per an Accenture report, AI can potentially boost profitability rates by an average of 38% and lead to an economic boost of $14 trillion across 16 industries in 12 economies by 2035. The report adds that manufacturing is one of the sectors that will experience the highest Gross Value Added (GVA) growth of 4.4% and add an additional $3.8 trillion in GVA by 2035. But what use cases are driving this substantial growth of AI in the industry?

The manufacturing industry depends on heavy machinery and has been benefiting from the roll-out of technologies like the Internet of Things (IoT) and connected devices. Moreover, manufacturing companies today rely on data and aim to become a data-driven enterprise. They collect data with the help of smart devices like sensors and RFID or from traditional maintenance survey-based processes. For instance, per a report by Fortune Business Inside, global big data in the manufacturing industry will exhibit a CAGR of 14% and spending for it is projected to reach USD 9.11 billion by 2026. Moreover, a Deloitte report suggests that the manufacturing industry is estimated to generate about 1,812 petabytes (PB) of data annually – which is significantly more than industries like communications, finance, and retail.

The increasing integration of IoT and adoption of data-driven metrics prime the manufacturing industry for AI application technologies. After all, harnessing the power of artificial intelligence makes sense as terabytes of data flow from almost every tool on the factory floor.

Like every other industry, the manufacturing industry is also dealing with unprecedented disruptions due to economic headwinds, supply chain disruption, labor shortages, and the consumerization of B2B sales. Thus, it is looking forward to capitalizing on solutions that help it cope with high revenue volatility, respond to real-time changes in demand across the supply chain, and increase productivity while mitigating expenses, enhancing quality, and decreasing downtime.

In such scenarios, AI has become a low-hanging fruit that can help propel the industry forward. A Deloitte survey found that 93% of companies believe that AI is a pivotal technology to drive growth and innovation in the manufacturing sector. The survey adds that 83% of companies think AI has made or will make a practical and visible impact.

Here are some use cases of AI in the manufacturing sector:

• Robotics

A decade or more ago, the thought of AI-enabled robots working in a factory only existed in the imagination or as a sci-fi movie scene. But thanks to artificial intelligence, it has become an important real-life scenario.

AI-powered robots and robotic process automation (RPA) are the best-use cases of how AI has transformed the manufacturing industry. Typically deployed in areas of monotonous or redundant work, robots have been used in manufacturing plants for decades to make operations faster, more efficient, and error-free. In fact, industrial robots began appearing in factories as early as the 1950s and 60s. However, it was only recently that manufacturing companies started deploying collaborative robots, also known as Cobots, to work alongside human workers and function as an extra set of hands.

Cobots are agile and spatially aware AI-powered robots that are capable of learning new tasks and augmenting human abilities with their own. Unlike the traditional industrial robots which are deployed in fenced premises, cobots are designed for safety and can operate in a shared workspace alongside humans. They have internal sensors that detect human contact and control motion to ensure safety.

While cobots may lift heavy car parts or handle assembly functions, they may also be programmed to perform tasks like material handling, packaging, labeling, quality inspection, and more.

One of the biggest multinational companies in the world, Amazon, has deployed more than 750,000 mobile robots with computer vision and machine learning to optimize operations worldwide. For instance, Robin, one of Amazon’s robotic handling systems, delivered nearly 1 billion packages (one-eighth of all the orders to customers worldwide). The company recently launched its first fully autonomous mobile robot, Proteus, and robotic work cell, Cardinal, aimed at improving safety and reducing the risk of injury in the warehouse.

• Supply Chain

A well-managed supply chain is the lifeline of every manufacturing company, and AI enables effective supply chain optimization. AI-powered tools help manufacturers forecast and respond to real-time changes in demand, optimize inventory levels, and enhance logistics. For instance, they can process a much broader spectrum of information than traditional models and utilize real-time data to develop more accurate and timely planning scenarios. They can even extract meaningful data and insights with deeper analysis of inventory data to transform inventory management and generate profits.

In industries like transportation, AI can evaluate variables like real-time traffic patterns and weather conditions to provide an expansive view of the supply chain. This goes a long way in helping managers optimize transportation and delivery options to reduce vehicle downtime and lower costs.

BMW, one of the largest manufacturers of premium automobiles and mobility services, has fully embraced artificial intelligence and uses it in critical areas like supply chain, production, R&D, and customer service. The company’s Steyr plant is a modern-day marvel and presents a state-of-the-art example of utilizing AI applications to enable logistics and optimize the supply chain. For instance, BMW uses AI applications in the facility to eliminate empty containers on conveyor belts and direct boxes to the removal station by the shortest route. The facility is also equipped with robotic applications that identify objects accurately to strategize the correct routes for the objects on the production line.

• Quality Control

Imagine working in a pharmaceutical or a fast-moving consumer goods manufacturing business, and a shipment of defective products from your factory reaches the market. Unsafe and life-threatening consumables can easily cost tens of millions of dollars due to lawsuits, fines, and the damage to your reputation. Even with non-consumable products, the cost can be huge. Quality control is a crucial component of the manufacturing process, and AI-powered tools make it faster, more effective, and efficient. Technologies like computer vision and image recognition systems can help detect defects throughout the production process and enhance quality control.

Samsung Electro-Mechanics, a subsidiary of Samsung group which manufactures smartphone parts, adopted an AI-enabled quality control system at its multi-layer ceramic capacitor (MLCC) production plants in South Korea and other countries. It uses an AI-enabled smart factory system to manage the production quality of tiny electric parts – which are sometimes as thin as a strand of hair. The company found that the AI-enabled quality management system increased the production yield of smart factories and helped save about $83 million annually.

• Predictive Maintenance

The equipment used in the manufacturing processes is typically expensive and capital-intensive.  Businesses want to avoid large, replacement capital expenditures, minimize repair costs, and prevent major financial impact due to unplanned downtime. To avoid reactive, unplanned equipment repairs after breakdowns, many businesses do regular, scheduled, preventative maintenance to fix equipment before it breaks down and cut expensive, unplanned downtime. However, since it relies on average fixed time intervals; it may result in maintenance costs even when the actual health of the equipment may not require maintenance. On occasions, it can even miss equipment that prematurely breaks down.

Predictive maintenance takes the volumes of sensor data available today in most modern factories to schedule maintenance based on the equipment’s condition and various other indirect measures.  AI and machine learning take this model even further by learning and refining models for when predictive maintenance should occur based on actual experience.  Furthermore, AI-enabled predictive maintenance helps to balance “the premature and costly repair of a system and ensures a timely repair before a failure.” In some discrete manufacturing cases, AI-enabled predictive maintenance can assist with the machines used to manufacture products and with the actual products produced to make them more attractive to buyers.

PepsiCo, one of the top consumer packaged goods companies in the world, uses a robust set of predictive maintenance solutions at its Fayetteville, Tennessee Frito-Lay plant. The plant produces around 150 million pounds of product per year, including Lays, Ruffles, Cheetos, and Doritos. AI-enabled predictive maintenance helped the company reduce equipment downtime by 0.75% in 2021 and avoid serious outages and mishaps. For instance, data from vibration monitors kept a combustion control motor from failing, thus avoiding an outage of the entire potato chip manufacturing process.  Similarly, data from acid monitors measuring oil from a baked extruder gearbox preidentified issues that otherwise would have resulted in the shutdown of Cheetos Puffs manufacturing.

Conclusion

While the above-mentioned are the most prevalent use cases of AI in manufacturing, many companies also deploy artificial intelligence to enhance customer experience, improve employee satisfaction, offer customized products at scale, strengthen cybersecurity, and enable continuous improvement.

In a nutshell, artificial intelligence is a force revolutionizing the manufacturing industry due to the availability of big data, increasing industrial automation, and improving computing power. It unlocks abilities to succeed in Industry 4.0 with enhanced efficiency, quality, and flexibility while minimizing costs and environmental impact.

Author:

Shobhit Kulesh
Celsior