Manufacturing teams often face slow workflows caused by fragmented knowledge, manual processes, and reliance on physical documentation. Operators spend valuable time searching for manuals, relying on tribal knowledge, or walking to machines to verify basic data. These inefficiencies slow onboarding, increase downtime, and make reporting less reliable.
To solve this, a modular Manufacturing Execution System (MES) was built with a Gen AI-powered assistant at its core. The goal: give factory employees a conversational interface to access knowledge, manage maintenance tasks, and interact with operational data in real time.
The Challenge
Factories struggled with:
- Slow onboarding due to scattered knowledge
- Frequent equipment downtime
- Manual ticketing and maintenance workflows
- Inconsistent KPI visibility across teams
The client needed more than just an AI model—they needed a usable, intuitive interface that frontline workers would adopt in real factory conditions.
The Solution: A Chat-Based AI Assistant for MES
The team designed and built a chat-style AI assistant embedded directly into the MES platform. Users interact with it as a modern AI chatbot, but it is deeply integrated with factory systems, knowledge bases, and maintenance workflows.
Key capabilities include:
- Selecting specialized AI agents (manufacturing, finance, multilingual support)
- Choosing different LLMs per task
- Customizing response formats (bullets, summaries, structured outputs)
- Saving favorite prompts
- Accessing role-based FAQs
- Generating reports and even podcast-style summaries
Centralized Knowledge Access
Employees can now upload manuals, PDFs, and technical documents to a centralized knowledge base. Approved users contribute expertise, making institutional knowledge searchable and reusable.
Workers can:
- Ask questions instead of searching documents manually
- Upload files directly into chat for instant analysis
- Access approved organizational knowledge in seconds
Smarter Maintenance and Ticketing
The assistant streamlines maintenance workflows by:
- Automatically generating tickets from detected issues
- Allowing quick manual ticket creation and assignment
- Supporting AI-assisted troubleshooting with references from manuals and past tickets
Each recommendation includes supporting documentation, helping teams verify actions before execution.
Conversational Dashboards
Instead of navigating complex dashboards, managers can ask natural-language questions about machine performance and KPIs. The assistant returns structured insights instantly, enabling faster and clearer decision-making.
UX Design for Real Factory Use
The interface was designed in Figma with a focus on simplicity and speed:
- Familiar chat-based interaction model
- Clear hierarchy for fast scanning on the factory floor
- Responsive design for mobile, tablet, and desktop
- Accessibility and usability for high-tempo environments
Vuetify was used to maintain consistent UI components across the system.
Real-Time and Voice Capabilities
To support hands-free environments, the assistant includes:
- Voice input with live transcription
- AWS-based speech-to-text processing
- Daily audio summaries of factory operations
WebSocket communication ensures real-time responsiveness across chats, dashboards, and workflows.
Technology Stack
The system is built using:
- Django (backend)
- Amazon S3 (file storage)
- Amazon Cognito (authentication)
- MySQL and Elasticsearch (data storage)
- REST and GraphQL APIs
- AWS speech-to-text services
- Vuetify (frontend UI framework)
Early Impact
Even in its ongoing phase, the system is delivering measurable improvements:
- 50% faster onboarding (4 weeks → 2 weeks)
- Faster troubleshooting with AI-guided recommendations
- Reduced downtime through streamlined ticketing
- Better decision-making via conversational KPI access
- Improved knowledge retention across teams
Conclusion
By embedding Generative AI directly into manufacturing workflows, the platform turns complex factory systems into a simple conversational experience. The result is faster operations, smarter decision-making, and a more connected industrial workforce.