United StatesEnergy and utilities2024 – ongoingHawaiian Electric Company
RAG chatbot for regulatory document queries
Hawaiian Electric's regulatory team cut a document query from five minutes to 5 seconds in a chatbot proof of concept
Problem
Hawaiian Electric's regulatory team had to search large sets of regulatory documents one query at a time. Each lookup took about five minutes.
Approach
The utility built a retrieval augmented generation proof of concept on Databricks. It works as a conversational chatbot, and staff ask questions about regulatory documents and get answers with page references.
Results
Takeaway
A document-search chatbot that cites page numbers lets compliance staff check an answer against the source. The utility reported the result from a proof of concept, not from a full rollout.
Categories
- Industry
- Energy and utilities
- Business function
- Internal work and knowledge management
- Innovation type
- Process innovation
- AI technology
- Generative AI
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Vendor case study
Sources
Source summaries
Databricks describes a chatbot proof of concept that lets Hawaiian Electric's regulatory team query regulatory documents and receive answers with page references. It reports that an individual query fell from five minutes to 5 seconds.
Full details are in the original sources.