Subscribe to Email Updates

Recent Stories

Unlocking Sustainability with Circularity
Unlocking Sustainability with Circularity Cyient
Unlocking Sustainability with Circularity
Accelerating Digital Transformation in Industry: Cyient's Proven Approach
Accelerating Digital Transformation in Industry: Cyient's Proven Approach Cyient
Accelerating Digital Transformation in Industry: Cyient's Proven Approach
Building an Efficient Face Recognition System
Building an Efficient Face Recognition System Cyient
Building an Efficient Face Recognition System
The Symphony of Automation: RPA and GenAI across Industries
The Symphony of Automation: RPA and GenAI across Industries Cyient
The Symphony of Automation: RPA and GenAI across Industries
GeoAI in Pipeline Monitoring and Management
GeoAI in Pipeline Monitoring and Management Cyient
GeoAI in Pipeline Monitoring and Management
Rajaneesh Kini Rajaneesh Kini Written by Rajaneesh Kini, President & Chief Technology Officer
on 04 Jan 2022

Manufacturing plants are fast gearing up for the future as they transform into autonomous smart factory systems. While organizations have already embarked on this journey over the last two or three years, the route to intelligent automation is still being mapped, one success story at a time.

AI-driven technology is improving every day and achieving better efficiency for manufacturing processes through predictive and prescriptive analytics. As in the case of any transformation, this journey too involves multiple stages and organization change management. To transition to a smart factory, it is important to take a step-by-step approach that will ensure true industrial automation with edge AI-driven distributed decision-making.

Step-by-step evolution

We could look at human social evolution as a model for the evolution of smart manufacturing. Over time, humans evolved from isolated tribes to small communities, to societies, cities, states, and countries. Leadership and decision-making power concomitantly progressed from tribal chieftains, to monarchs, to democratically elected governments. As social evolution became more complex, several layers of intermediate decision-making at lower levels—or the edge—evolved to sustain the system and ensure it was responsive and efficient. Smart manufacturing is charting a similar path.

The manufacturing industry is well along the road that leads to Industry 4.0, with several organizations having taken the first step—plugging individual products and machines into a smart factory ecosystem. With this, individual machines are starting to communicate data with each other and work in collaboration. This is called connected products or a connected factory ecosystem—one that promises to improve dramatically with the advent of 5G and improved infrastructure.

Manufacturing units are now approaching the next step—an integrated cloud-based AI platform to monitor, decide, instruct, and manage products and their health. With these two steps in place, we have a monolithic system of improved decision-making in place. The advantage of this system is that data is available in one place to make integrated autonomous decisions, largely based on predictive analytics and monitoring. Cloud AI platforms and solutions are being used for various purposes including analytics, predictive solutions, and monitoring.

With the advent of 5G and advanced computing capabilities in cloud improving every day, connected products and cloud AI will continue to evolve swiftly. For evolution of smart automation at the factory level, however, intermediate steps are necessary to complete the transition. Tomorrow’s engineering will therefore focus significantly on three intermediate areas: Embedded AI; Edge Compute and AI; Fog Computing and AI.

Embedded AI or increased device intelligence: Engineering companies will invest in improving product intelligence. Machines and products will not just capture information but start making sense of it, take first-level decisions, and find ways, for instance, to defend themselves against security attacks.

Edge Compute and AI or decision-making at the edge: Typically, democracies have a tiered system for socio-economic decision-making—Village councils in rural areas and municipal administration in cities—to handle immediate, community-focused projects such as healthcare, water supply, local energy network, waste management, education, and family welfare.

Edge computing and edge AI replicate the tiered decision-making of villages and municipalities to industrial environments such as a smart factory or smart energy setup. Areas that require quick and contextual decisions may be managed at edge level such as monitoring a patient’s vital signs, safety and security, and energy and waste management. Studies show that tiered decision-making using edge AI reduces the operational costs by 30-50% as compared to monolithic cloud AI-based decision- making, due to reduced data transfer to cloud.

Fog Computing and AI: Fog computing acts an intermediate layer between edge and cloud. While edge AI is applied directly on the devices or another intelligent node device that is physically very “close,” Fog AI is about decision-making a layer above, in more sophisticated processors that are connected to the LAN or into the LAN hardware itself. Fog computing allows organizations to aggregate data from multiple devices and nodes into regional stores.

Decisions that are relevant to a small unit of the factory may be taken within the local data center using fog computing and AI. This reduces data transferred to cloud and ensures security of local information management. Fog computing mainly supports decision--making that needs slower response times compared to edge, but faster than cloud.


Emerging technology solutions

Edge AI technologies are fast evolving to support such transformation. In 2021, Edge Computing and AI related strategic funding touched $3.3 billion compared to $500 million in 2018. This investment is spread across edge landscape technology areas such as edge AI processors, edge platforms, edge orchestration, edge servers, edge security, and edge databases.

  • In 2019, Google launched Coral—a platform of hardware and software components that help you build devices with local AI—providing AI acceleration right on the edge device without connecting to the cloud.
  • In 2020, hybrid edge cloud company Mimik Technology collaborated with IBM to create AI-enabled integrated workflow technologies to make edge computing more attainable for the manufacturing, retail, IoT, and healthcare industries.
  • In 2021, Microsoft launched Azure Percept, a comprehensive, easy-to-use platform with added security for creating edge AI solutions.
  • The AWS Snow Family comprised of AWS Snowcone, AWS Snowball, and AWS Snowmobile, offers a number of physical devices and capacity points, most with built-in computing capabilities.

Decisive edge

Industry 4.0 is about smart operations using analytics and AI at various stages for decision-making to improve operational efficiency. Edge AI is assuming critical importance in delivering last mile intelligent automation in industrial environments from smart factories and healthcare to autonomous vehicles. While deployment of edge AI is still picking up, the march of 5G and IoT will accelerate its reach to millions of connected devices and products enabling real-time decision-making and action at the edge, driving never-seen-before efficiencies and cost savings. It will also mean that IT-OT convergence rather than business unit personnel will increasingly manage decisions at the edge, fog, and the cloud.

Let Us Know What You Thought about this Post.

Put your Comment Below.

You may also like:

Talk to Us

Find out more about how you can maximize impact through our services and solutions.*

*Suppliers, job seekers, or alumni, please use the appropriate form.