Digital Identity

Industry 4.0 at Scale Can Make Smart Manufacturing Pervasive

Doug Bellin on IT-OT Convergence and AI in Smart Manufacturing
Industry 4.0 at Scale Can Make Smart Manufacturing Pervasive
Douglas Bellin, business development executive and worldwide head of smart manufacturing, AWS

The past decade has seen Industry 4.0 take strides with advancements in technologies such as IoT, cloud and 3D printing. However, compatibility and scalability issues often stall wider implementation. AWS addresses this data challenge with its Industrial Data Fabric and Manufacturing and Industrial Competency. These initiatives aim to streamline, contextualize and manage vast amounts of manufacturing data using AI and ML. This approach structures data for insightful analysis, addressing fragmentation and complexity.

See Also: The Forrester Wave™: Governance, Risk, And Compliance Platforms, Q4 2023

At the AWS re:Invent 2023, ISMG spoke with Douglas Bellin, business development executive and worldwide head of smart manufacturing for AWS, on how AWS powers smart factories, the convergence of information technology (IT) and operational technology (OT) and AI's potential in smart manufacturing.

Edited excerpts follow:

How has Industry 4.0 evolved in the past decade, particularly in terms of scalability, and what challenges persist in its implementation?

Industry 4.0 has seen significant advancements in technological domains like IoT, cloud computing and 3D printing. However, implementing these innovations across diverse machinery setups remains a challenge. While isolated implementations like predictive maintenance have shown success, scaling these solutions across multiple machines with varying configurations poses challenges due to compatibility issues and scalability concerns. IDC research suggests that 65% of Industry 4.0 projects get stuck at pilot stage.

Can you provide more details on AWS' proposed strategies to address the data challenges in manufacturing?

There is an explosive increase in the volume of data in manufacturing organizations. Some of these companies produce more data that any financial or telecom company combined. This leads to the need for effective data management and governance. AWS introduced the concept of an Industrial Data Fabric, and that aim to streamline, contextualize and manage vast amounts of manufacturing data. Leveraging AI and ML techniques, this approach intends to structure data for more insightful analysis, addressing the issues of data fragmentation and complexity.

How did AWS extend its competency program and tackle the integration challenges between OT and IT in smart manufacturing?

Integrating OT and IT in smart manufacturing has historically been challenging due to differing operational approaches. There is a need for collaboration and mutual understanding between these domains to drive successful transformations and streamline technology adoption within manufacturing to make it smarter. AWS expanded its competency program to include hardware and system integrators, recognizing the importance of partners understanding industry-specific requirements.

Can you share recent instances of customer deployments in the smart factory domain, particularly highlighting the AWS services utilized?

Some of the recent instances involve a couple of prominent paper companies in the U.S., which happen to supply to Amazon as well. Enhancing the paper production or ensuring no stock outs here has a direct impact on Amazon's operations. One of our clients, operating across 130 locations in the U.S. with plants ranging from 40 to 50 years old to more recent ones, faced challenges in accessing diverse data across these sites. To address this, they aimed to establish a centralized infrastructure to manage all locations, providing near-real-time data feeds. This facilitated maintenance programs and enabled remote monitoring across their field operations. As a result, the company decided to utilize various AWS services such as IoT Greengrass, Amazon Outposts, Lambda functions, multiple databases, AI/ML and SageMaker extensively. Additionally, they incorporated Amazon Bedrock to facilitate this process. Recently, they've focused on implementing knowledge management systems, enabling field workers with limited experience to access accumulated tribal knowledge swiftly. This initiative aims to empower newer workers, who lack extensive plant experience, to leverage the existing tribal knowledge effectively, thereby facilitating faster decision-making and driving business transformation.

How do you envision the evolving role of generative AI in the context of smart manufacturing?

We're barely scratching the surface to realize the potential of generative AI. Consider a worker's typical task: locating and utilizing manuals to address issues at a workstation. This process usually takes hours, sifting through pages and troubleshooting cryptic error codes. Now, picture digitizing these manuals, integrating them into a bedrock system and overlaying an anthropic cloud interface. With this setup, a worker standing before a machine encountering an error can simply input the error code and receive a clear explanation of the problem. This saves an immense amount of time, directly translating to cost savings and increased uptime.

But the real potential lies in integrating this knowledge into workflows seamlessly. Imagine moving from receiving an error message to being provided with repair instructions while simultaneously updating maintenance records and checking inventory for required spare parts - all automatically. This transformative capability eliminates the need for manual intervention, which currently involves hours of inventory checks, finding personnel for repairs, and managing associated logistics.

Do you foresee an increase in the development of industry-specific language model (LLMs) due to the impact of these advancements?

We're already witnessing it. For instance, with projects like Code Whisperer, we're collaborating with several companies to craft LLMs dedicated to programmable logic controllers. These specialized models enable swift coding or on-the-fly alterations based on specific product needs or production line changes. They essentially offer a pre-built library tailored to these specialized requirements. Generative design has been around for a while and gained popularity. However, the critical evolution it needs involves setting constraints within generative design. For example, requesting the creation of the finest ergonomic chair might produce an outstanding design, but it might be unfeasible for current manufacturing capabilities, even with advanced technologies like 3D printing. The future of generative design lies in setting these guardrails to align with the capabilities of the intended manufacturing equipment, ensuring practical and feasible outputs.

Bellin leads the strategy and execution of manufacturing and supply chain solution areas across industrial customers at the intersection between OT and IT. Prior to AWS, he ran the marketing, go-to-market and business development teams for the industrial markets within Cisco Systems.


About the Author

Rahul Neel Mani

Rahul Neel Mani

Founding Director of Grey Head Media and Vice President of Community Engagement and Editorial, ISMG

Neel Mani is responsible for building and nurturing communities in both technology and security domains for various ISMG brands. He has more than 25 years of experience in B2B technology and telecom journalism and has worked in various leadership editorial roles in the past, including incubating and successfully running Grey Head Media for 11 years. Prior to starting Grey Head Media, he worked with 9.9 Media, IDG India and Indian Express.




Around the Network

Our website uses cookies. Cookies enable us to provide the best experience possible and help us understand how visitors use our website. By browsing cio.inc, you agree to our use of cookies.