Mitsubishi Heavy Industries accelerates energy digital transformation with domain-specific generative AI
Mitsubishi Heavy Industries Ltd (MHI), a global leader in engineering and manufacturing, set an ambitious goal to reach carbon neutrality by 2040. As part of their Mission Net Zero, MHI sought to digitally transform their energy operations by harnessing advanced AI to support smarter infrastructure and accelerate energy transition. The company established TOMONI, an intelligent solution supporting power plant operations, and, in 2023, launched TOMONI TALK with ChatGPT—a generative AI chat app based on Azure OpenAI Service, Azure AI Search, and Cosmos DB. With the expert support of Zen Architects, MHI implemented Retrieval Augmented Generation (RAG) architecture to connect generative AI with internal data for document summarization, translation, and workflow enhancements. The system, rapidly prototyped in a joint workshop, is now used daily by 200-300 employees, dramatically boosting productivity and knowledge sharing. The project showcases the power of cloud-native AI and cross-functional collaboration to drive digital innovation and lays the foundation for further AI-enabled transformation across MHI's business domains.
- Organization
- Mitsubishi Heavy Industries
- Industry
- Energy & Utilities
- Location
- Japan
- Published
- May 2024
Reported outcomes
Strategic outcomes
Primary read
Use case focus
Showing 3 of 4
- 1Domain-specific Knowledge Search using Generative AI
- 2Employee Support Chatbot for Power Plant Operations
- 3Automated Document Summarization and Translation
- Difficulty supporting energy transition and carbon neutrality with traditional infrastructure
- Lagging digitalization in energy sector operations
- Need to efficiently harness vast amounts of proprietary know-how and documents
- Challenges with information search, summarization, and knowledge sharing across business units
- Concerns about data security and privacy when using generative AI services
- Deployed TOMONI intelligent solution to digitize power plant operation and maintenance
- Developed TOMONI TALK with ChatGPT, an internal AI chat app leveraging Azure OpenAI Service with Retrieval Augmented Generation (RAG)
- Integrated Azure AI Search and Cosmos DB for secure and accurate internal data retrieval
- Partnered with Zen Architects for custom RAG architecture development and rapid prototyping
- Focused on business-driven AI applications: document summarization, translation, troubleshooting, and workflow automation
- 200-300 daily active users of TOMONI TALK with ChatGPT
- Significantly improved operational efficiency and internal knowledge sharing
- Streamlined document summarization, translation, and troubleshooting for employees
- Established best practices for cloud-native, secure generative AI implementation
- Enabled scalable AI-driven transformation with potential expansion group-wide
Architecture
TOMONI TALK with ChatGPT runs on Azure OpenAI Service connected to internal business data via Azure AI Search and Cosmos DB, using a Retrieval Augmented Generation (RAG) architecture. User queries are interpreted by the LLM, which then triggers domain-specific data searches; search results are returned, and the LLM generates responses based on this context. Zen Architects supported customized RAG implementation, leveraging Azure PaaS services for secure and scalable deployment.
Implementation partners1
Sources & evidence1
The case's original source is still reachable.
- Cited source last checked Jun 1, 2026 — ok (0/1 broken).
Measures whether this deployment's public evidence persists — not whether the system is still in production.
AI-generated summary. Verify important details with the linked sources before relying on this case.
Explore related AI use cases
Was this useful?