Eight ways to optimize operations with industrial AI

A digital business platform can industrial customers accelerate through challenges that cross many industries. The Siemens Industrial Copilot for operations helps resolve production challenges eight ways, including with code generation, as explained at the 2024 ARC Industry Forum.

Industrial AI help for optimization insights

• A digital business platform helps industrial customers accelerate from where they’re at, in eight ways.
• The Siemens Industrial Copilot for operations helps resolve production challenges.
• Broad AI offerings are available in a platform designed to accelerate industrial production.

Use of industrial artificial intelligence (AI) tools integrated into a digital business platform can help resolve current and address future challenges across industries, as explained by Siemens software experts at 2024 ARC Industry Forum. Mathias Oppelt, Ph.D, Siemens vice president customer-drive innovation, and Timur Yaprak, Siemens industrial AI innovation and partner engagement (Figure 1), discussed “The future of industrial engineering and operations: Will industrial copilots help us accelerate?”

Oppelt said using the right tools can with market challenges industries face: globalization and localization, individualization, digital talent, resilience, digital disruption and sustainability.

Localization is defined as the need to produce where it’s used, he said. Customers also might require more individualized products. Most industries need talent. If things happen to change the rules during the game, industries need more speed and adaptability.

Siemens Xcelerator is an open, digital business platform with digital and internet of things-enabled offering from Siemens and certified partners, Oppelt said. The platform is used in more than 200 applications among more than 130 customers, including automotive, intralogistics, electrics, chemical, original equipment manufacturers, pharmaceutical and food and beverage industries, as well as original equipment manufacturers and system integrators and others. It’s a portfolio, ecosystem and marketplace to exchange and learn with customers, partners and experts, he explained.

The future of industrial operations will be changing. For example, bio-produced products have more variability so need to be more flexible when using bio-based materials. Automotive differs significantly from 10 years ago requiring additional adaptability. Pharmaceuticals are using human digital twins for specific medicine with integrated cybersecurity, he said.

Digital business platform helps industrial customers accelerate eight ways

Each customer is in a different place, and cross-industry parameters are more important. Industrial software can help in eight ways (Figure 2): Localization, production network, volume and variability, degree of automation, business model, operations flow, openness of automation and shopfloor design.

Industrial control can adapt in the speed of software, Oppelt said. Imagine control, devices, security, data and integration all in software (Figure 3). Programmable automation controllers (PLCs) provide useful functionality, but controllers don’t need to be in a gray box. PLCs can be deployed from the cloud to dedicated industrial devices with applications scaled quickly as needed. Support can be provided to manage device status, inventory, regulation and carbon footprint. Cybersecurity can be provided for all industrial operations devices with automated detection and mitigation of vulnerabilities. The power of generative artificial intelligence (AI) can be used as a co-pilot for manufacturing engineering.

Integration adds value to businesses by connecting companies across the value chain.

Future challenges make traditional operational technology (OT) and information technology (IT) methods difficult. Moving from traditional control to software-driven control improves data integration and flow across ecosystem partners enabling more flexible decisions for continuous optimization. Current state of industrial engineering most often is sequential: design, simulation, engineering, commissioning, production, service and recycling.

With new tools, industries will more quickly iterate and leverage the full value of data from OT sensors in the field to controllers and edge computing to IT cloud capabilities. Cybersecurity is enhanced with over-the-air updates, automated vulnerability mapping, security management with vulnerability management, configuration management and anomaly detection for all OT assets.

Siemens Industrial Copilot for operations helps resolve production challenges

Yaprak continued by explaining how Siemens Industrial Copilot for Engineering can help.

In cooperation Schaeffler, it creates automation code with generative AI for faster development, higher efficiency and cost reduction (Figure 4). It helps operations and can result in faster maintenance. Ask the machine what’s wrong, Yaprak said, and it’s like a backseat driver that actually provides help. Generative AI operated on the low-code Siemens development platform Mendix in partnership with AWS.

It democratizes knowledge to a wider base, Yaprak said.

A process flow diagram shows the general flow of process and equipment, capturing the main devices, systems and workflow. The P&ID Engineering Copilot software then generates a piping and instrumentation diagram (P&ID) used to control the process.

Extensive AI offerings, platform to accelerate industrial production

Three decades of AI efforts at Siemens includes more than 1400 AI experts, 3700 AI patents, 36 university and research partnerships, five cloud provider collaborations, more than six industrial AI services, 20 factories with AI applications in use, 15 startup collaborations and more than 15 industrial AI applications. More collaborators are welcomed.

As a service, Yaprak said the Siemens Xcelerator platform is flexible, interoperable, open and cybersecure, designed for easier and faster digital and sustainable transformation.

Tools like these, said Oppelt and Yaprak, are helping automated production move to adaptive production then more to future production of an automated factory with an industrial metaverse.

Mark T. Hoske is content manager, Control Engineering, CFE Media and Technology, [email protected].

KEYWORDS: Digital industrial platform, industrial IA, industrial optimization

CONSIDER THIS

With better industrial software tools, how quickly could you accelerate past competitors?

ONLINE

https://www.controleng.com/ai-machine-learning/

See related industrial AI stories linked below.