The first evidence of the “wheel” dates back to 3,500 B.C. During that era, it served as a potter’s wheel. The idea gradually evolved, and in the next few hundred years, it found its use in chariots. Developments thereafter are well documented – from motor vehicles to aeroplanes to machines, and practically everywhere else. We never stopped innovating after inventing the wheel.

In 1950, English mathematician Alan Turing published a paper titled “Computing Machinery and Intelligence” which opened the doors to the field we know today as Artificial Intelligence. In 1956, John McCarthy, an American computer scientist and cognitive scientist, coined the term “Artificial Intelligence” or AI. After decades of research, computers are gradually moving closer to passing the Turing Test. The Turing Test is a method of inquiry in AI for determining whether a computer is capable of thinking like a human being. Here, we will explore the different building blocks of an AI system, the extent of AI adoption by businesses, and the next big leap. This document aims to help business leaders build a more focused strategy in their journey toward an Intelligent Factory.

Industry 4.0 and the AI Locomotive

According to McKinsey, 50% of companies that embrace AI over the next five to seven years have the potential to double their cash flow, with manufacturing leading all industries due to its heavy reliance on data (Source). Industry 4.0 makes factories and every entity on the shop-floor smarter, and AI plays a vital role in this journey. AI offers new ways to boost employee productivity and creativity, increase business agility, improve customer engagement, and jumpstart new product innovation. To infuse AI in the factory landscape, technology companies leverage the existing systems and also identify or create new avenues to make the systems more intelligent.

Integration of AI with recent emerging technologies such as Industrial Internet of Things (IIoT), big data analytics, cloud computing, and cyber-physical systems will enable operation of industries in a flexible, efficient, and sustainable way. If we consider the idea of an “AI locomotive” in the context of Manufacturing, we can break down the various components of this system as follows:

1. The fuel – Data: Today, technologies have enabled us to extract information from every bit of raw data, irrespective of whether they reside on premise, on the cloud, or on the edge. This helps manufacturers define substantial performance metrics across dimensions. The sources of data attribute to man, machine, material, and the interaction of these entities. Further, capture and storage of data has simplified over the years with cloud technologies complemented with a data security layer.
2. The engine – Analytics: A powerful analytics engine is capable of extracting hidden patterns, correlations, and other useful information from collected data (like sensory data). Big-data analytics complemented with Natural Language Processing has opened new doors by enabling more ways of effectively handling large volumes of information.
   
3. The compartment – Platform: The platform constitutes the underlying hardware architecture to support the overall mechanism. It helps improve agility and event processing of the existing industrial system. The platforms are primarily of three types – stand-alone, embedded, and cloud based. The cloud platforms are capable of rapid service deployment, high level of customization, knowledge integration, and effective visualization with high scalability.
   
4. The control mechanism – Operational Technology: The information collected through various sources and the insights generated with the analytical steps, assist in defining the appropriate set of activities for execution. The various entities of the factory communicate and collaborate with each other to arrive at the right set of operations. The decision-making capability lies with each of these entities because of the overall intelligent environment setup.
   

The Current state of adoption of AI/ML technology

Companies that have implemented Artificial Intelligence and Machine Learning have realized phenomenal gains in operational efficiency and revenue. Areas where they have demonstrated valuable outcome include:

• Predictive Maintenance: Analyze data captured from the machinery to ensure they are operating correctly and identify any ambiguous patterns for early detection of faulty behavior.
  
• Quality Management: AI/ML when combined with computer vision can enable real-time quality checks for products coming off the production line and reject those that do not meet the quality constraints.
  
• Workforce Transformation: Skill enhancement with developments like remote assistance and training. Mixed Reality, for instance, has redefined the training and skills transfer while increasing the efficiency on the shop floor.
  
• Automation of processes and operations: Industries are increasingly relying on embedded intelligence like learning from recorded data and hardware capabilities like collaborative robots (Cobots) on the shop floor, resulting in reduced production time.
  

The Future State: ‘Batch size 1’ production

‘Batch or Lot size 1’ refers to manufacturing a single product for an individual customer. This requires superior levels of customization of an item to match the buyer’s specifications. Achieving a production batch size of a single unit for any product with the current setup would require planned downtime of the production system with high probability of manual errors leading to high cost. Seamless customization is necessary for faster production at a reduced cost. With advancement in AI/ML technology, we are heading toward an environment of digital factories with networked machinery, equipment, and even workers, making it possible for these entities to interact and collaborate with each other. This setup enables an interconnected smart ecosystem capable of more accurate decision-making. As a result, individual products can be identified and necessary operations can be performed in an automated manner. The outcome will be distinct, tailored products with swift response to changes in demand in real-time.

There is a burning need for the systematic development and implementation of Artificial Intelligence to see its real impact in the next generation of industrial systems. As humans, we have always aimed for simplicity and a more productive lifestyle. With our intellect, we constantly brainstorm new ideas and innovate to simplify our lives, save time and resources, and in the end, define a transformed way of living. 

Mohit Prasad

Consultant, Manufacturing – Digital Practice

Mohit is an MBA graduate from IIM Indore and a Computer Science engineer with strong expertise in consulting, product messaging, and GTM for AI, IoT, and Cloud-based technologies. He also has prior experience in cloud computing and tools development.