Bridging the gap between academic research and real-world solutions

The integration of generative artificial intelligence (AI) is rapidly transforming various aspects of the industrial sector, offering innovative solutions for maintenance strategies and the complexities of programmable logic controller (PLC) programming. This post explores how these advancements, particularly those highlighted by Siemens' new offerings and the application of retrieval-augmented generation (RAG) in PLC coding, are poised to redefine industrial operations.

Intelligent Maintenance with Generative AI

Siemens AG has launched a new generative AI-powered maintenance offering designed to support every stage of the maintenance cycle. This solution aims to help industries move beyond traditional, often reactive, maintenance practices toward an intelligent, data-driven approach. The offering extends the Senseye Predictive Maintenance solution, powered by Microsoft Azure, with two new packages:

• Entry Package: Provides an accessible introduction to predictive maintenance, combining AI-powered repair guidance with basic predictive capabilities. It facilitates the transition from reactive to condition-based maintenance through limited sensor data collection and real-time condition monitoring. This package offers AI-assisted troubleshooting and minimal infrastructure requirements, leading to reduced downtime and improved maintenance efficiency.

• Scale Package: Designed for enterprises aiming for a full transformation of their maintenance strategy. It integrates Senseye Predictive Maintenance with the full Maintenance Copilot functionality, enabling customers to predict failures before they happen, maximize up

  

  time, and reduce costs using AI-driven insights. This package offers enterprise-wide scalability, automated diagnostics, and supports long-term efficiency and resilience across multiple sites.

This new offering provides comprehensive coverage across the entire maintenance cycle, from reactive repair to predictive and preventive strategies, leveraging generative AI-driven insights to enhance decision-making and efficiency. Early pilot use cases have shown that the Industrial Copilot for maintenance can help save an average of 25% in reactive maintenance time. Siemens emphasizes that their Industrial Copilot empowers customers to leverage generative AI across the entire value chain, from design and planning to engineering, operations, and services.

Simplifying PLC Programming with Generative AI

Generative AI is also making significant strides in simplifying complex PLC programming. While caution is advised due to the potential for AI to generate incorrect information (hallucinate), techniques like retrieval-augmented generation (RAG) are being employed to mitigate this risk. RAG systems enhance the accuracy and relevance of large language models (LLMs) by searching external data sources and pre-processing information before integrating it into the LLM. This allows for training AI with specific knowledge, such as approved libraries and best practices, ensuring that AI-generated code aligns with established standards. RAG can also learn machine specifications and coding practices to assist in generating PLC, HMI, and SCADA code, potentially reducing tedious, error-prone, and repetitive programming tasks.

Furthermore, breaking down PLC programs into smaller program organization units (POUs) is considered a best practice, making code easier to build, troubleshoot, and maintain. Generative AI can act as a valuable partner in this process, helping programmers explore the function and purpose of individual POUs within a larger program and optimize their scope, structure, and variable names. AI can also estimate program complexity using known programming metrics and suggest methods for reducing it. While current AI tools work better with structured text (ST), the future may see advancements allowing for more seamless integration with ladder logic diagrams (LD). Ultimately, AI tools are expected to give engineers more time to focus on higher-level projects.

References:

• Siemens AG. (2025, March 26). Copilot with New Generative AI-Powered Maintenance Offering. IEN Europe.

• Townshend, Anna. (2025, March 19). How can generative AI simplify complex PLC programming? Control Design.