Demonstrating the End-to-End Capabilities of Manufacturing in SAP Supply Chain Management​

Introduction: Conversational AI introduces a new way for users to interact with manufacturing systems. Instead of navigating menus or searching documentation, users can access guidance and perform tasks through natural language.

The assistant is embedded directly within the application environment. It understands system context and provides relevant support based on the user's current activity.

How Users Interact with the Assistant: Users can ask questions in natural language rather than navigating through multiple screens. The assistant interprets intent and provides targeted responses.

• Answers how-to questions for system tasks
• Guides users to the correct application or function
• Provides contextual help based on the current screen
• Reduces reliance on external documentation

This simplifies how users access information and complete tasks.

System Awareness and Context: The assistant is aware of the user's environment and system configuration. Responses are tailored to the specific tenant and operational context.

• Connects to system data and configuration
• Aligns responses with available functionality
• Reflects the user's role and permissions
• Adapts guidance to real system conditions

This improves relevance compared to generic help systems.

Integration with AI and Knowledge Sources: The assistant combines system data with documentation and AI capabilities.

• Uses natural language processing to interpret requests
• Connects to documentation and knowledge bases
• Provides step-by-step guidance when needed
• Supports both learning and execution

This creates a unified access point for support and information.

• Reduces time spent searching for information
• Simplifies navigation across complex applications
• Improves task execution accuracy
• Lowers dependency on training for basic tasks

Users can focus more on execution and decision-making.

• Conversational AI enables natural language interaction with systems
• Users receive contextual, system-aware guidance
• Integration with documentation and data improves accuracy
• Navigation and task execution become more efficient
• User experience is simplified across manufacturing applications

Generative AI is changing how manufacturing applications are extended and customized. Instead of manually building components, developers and power users can generate functional elements quickly using prompts and contextual input.

A plugin for worker guidance can be created using a large language model rather than traditional development. The user provides context and instructions, and the system generates the required structure and interface.

From Prompt to Functional Plugin — The process begins with defining requirements in natural language. The system translates these inputs into working components.

• Provide context, structure, and desired functionality
• Generate UI elements such as tabs and guided steps
• Create logic for data capture and workflows
• Render initial screens for immediate use

This reduces the need for manual coding during early development.

Acceleration of Development Time — Traditional plugin development can take days depending on complexity. With LLM support, initial versions can be generated in minutes.

• Rapid creation of prototype interfaces
• Faster iteration on design and functionality
• Reduced effort for repetitive configuration tasks
• Immediate visualization of outcomes

This enables quicker experimentation and refinement.

Customization and Extension — Generated plugins can be refined and extended after creation. Users can adapt outputs to meet specific operational requirements.

• Modify generated components as needed
• Add additional logic or validation rules
• Tailor workflows to specific use cases
• Build on generated structures rather than starting from scratch

This balances speed with flexibility.

• Lowers barrier to creating custom functionality
• Encourages experimentation and rapid prototyping
• Speeds up delivery of user-facing improvements
• Enhances the adaptability of manufacturing applications

AI supports both development efficiency and user experience improvements.

• LLMs enable rapid creation of application plugins from prompts
• Development time is significantly reduced
• Generated components provide a starting point for customization
• Users can extend functionality without deep coding effort
• AI accelerates innovation in manufacturing application design

Introduction: Production issues often originate from changes across interconnected systems, especially between OT and IT layers. Identifying the root cause quickly is critical to maintaining stable operations.

Production failures can result from subtle changes in system behavior. These changes are often difficult to trace when multiple systems interact.

Understanding the Source of Issues: Changes in the OT layer can propagate into IT systems without clear visibility. When values or configurations shift, systems may encounter unexpected conditions.

• Tag values may change outside expected ranges
• Firmware updates can alter system behavior
• Configuration or mapping changes introduce inconsistencies
• Unknown status codes may trigger process failures

These issues are not always immediately visible in standard monitoring tools.

From Logs to Root Cause: Traditional troubleshooting requires reviewing logs across systems. This process can be time-consuming and dependent on expert knowledge. AI-assisted analysis accelerates this process by interpreting system data.

• Aggregates logs and process information
• Identifies anomalies and failure points
• Highlights the most likely root cause
• Provides contextual explanations for the issue

This reduces the effort required to diagnose complex problems.

Linking Upstream and Downstream Impact: A single issue can affect multiple stages of a process.

• A paused or incorrect order can disrupt downstream execution
• Upstream inconsistencies propagate through the system
• Dependencies between processes amplify the impact
• Identifying the origin prevents repeated failures

Understanding these relationships is key to effective resolution.

Faster Troubleshooting and Resolution: AI shortens the time required to identify and resolve issues.

• Reduces manual log analysis
• Provides immediate insights into failures
• Supports faster corrective actions
• Improves system stability and reliability

This enables teams to respond quickly to operational disruptions.

• Process issues often stem from changes across OT and IT layers
• Manual troubleshooting across logs is time-intensive
• AI assists in identifying root causes quickly
• Understanding upstream and downstream impact is critical
• Faster diagnosis improves operational stability and response time

Interactive value journeys provide a practical way to explore end-to-end manufacturing processes. They translate concepts into demonstrable scenarios that help users understand how solutions work in real situations.

These journeys present complete process scenarios rather than isolated features. They allow users to see how different capabilities connect across the manufacturing lifecycle.

Interactive Value Journeys (IVJ)

1. Installation and configuration: Manufacturing environments
2. Real-time monitoring: Operations and performance
3. AI-driven inspection: Quality processes
4. Handling of nonconformance: Issue resolution
5. Insights and analytics: Across manufacturing processes

Each journey focuses on a specific aspect while connecting to the broader process.

Value for Demonstration and Learning: Interactive journeys are designed to support both education and customer engagement. They provide structured, repeatable scenarios that are easy to present.

• Demonstrate capabilities in a clear, guided format
• Show how processes work across systems
• Support one-on-one discussions and larger sessions
• Help users understand business and technical value

This makes complex processes easier to communicate.

Continuous Updates and Relevance: The content is actively maintained and expanded. New scenarios and improvements are added over time.

• Reflects current capabilities and innovations
• Expands coverage across manufacturing use cases
• Incorporates feedback from field and customers
• Keeps demonstrations aligned with evolving solutions

This ensures ongoing usefulness and accuracy.

• Improves clarity during solution discussions
• Provides concrete examples of process execution
• Increases confidence in proposed solutions
• Enhances overall understanding of end-to-end value

Example: A customer can follow a full process from monitoring operations to resolving a quality issue, gaining a clear view of how systems interact.

• Interactive value journeys illustrate end-to-end manufacturing processes
• They connect multiple capabilities into coherent scenarios
• Useful for both learning and customer demonstrations
• Continuously updated to reflect current solutions
• Improve communication of business and technical value

Automation in production engineering often requires small but critical scripting tasks. AI-assisted script generation reduces the effort and time needed to create these components while improving accuracy.

Scripting is commonly used to handle logic within production processes. These tasks include data parsing, calculations, and updates that support execution.

Traditionally, scripts must be written and tested manually. This requires time, iteration, and familiarity with syntax. With AI assistance, scripts can be generated directly from a description.

1. Define the task in natural language: Describe what the script should accomplish.
2. Generate code for the required logic: AI produces functional code based on the description.
3. Insert parameters and variables automatically: Required inputs are identified and applied.
4. Test and refine within the same interface: Validation occurs in the same environment.

This reduces the need for manual coding effort.

Generating scripts manually can take several minutes or longer depending on complexity. AI reduces this to seconds while improving consistency.

• Immediate generation of functional code
• Fewer errors compared to manual entry
• Reduced trial-and-error during development
• Faster iteration and deployment

This accelerates development within production process design.

• Parsing and transforming data
• Updating timestamps or calculated values
• Managing workflow conditions
• Handling system-specific logic within processes

These tasks support automation and process consistency.

AI-assisted scripting changes how engineers build and maintain processes. Instead of focusing on syntax, they focus on intent and logic.

• Lower barrier to creating custom logic
• Faster development of process enhancements
• Improved reliability of generated scripts
• More time available for higher-level design

Example: A time calculation script that might take several iterations to write manually can be generated instantly and applied within the workflow.

• AI enables rapid generation of scripts for process tasks
• Development time is reduced from minutes to seconds
• Accuracy improves with fewer manual errors
• Common scripting tasks are simplified
• Engineers can focus on process design rather than coding details