Automated Document Generation via Hyper-scale AI

Overview

The maintenance of facilities managed by the Seoul Facilities Corporation (SFC) requires the repetitive creation of essential documents. This process has historically demanded significant resources, leading to operational inefficiencies. To minimize this resource drain, a Generative AI-based service was established to automatically draft documents in standardized formats.

Project description

The system learns from internal data and legal statutes to automatically draft documents and conduct legal reviews by leveraging a collection of standard and exemplary templates. Specifically, the Automated Standard Document Drafting Service organizes large volumes of collected data into a single, comprehensive document. Complementing this is the Exemplary Report Drafting Service, which generates optimal reports based on proven best practices even when only minimal input is provided. Additionally, the Automated Legal & Regulatory Review Service was developed to automatically reflect recent amendments in laws or regulations into existing documents. By integrating these features, the corporation has transformed its document-centric workflow into an intelligent, AI-assisted process.

Key technological features

• LLM-based Natural Language Processing: Automated generation of documents tailored to specific characteristics by collecting standard and exemplary documents, followed by legal/regulatory compliance reviews
• RAG-driven Search Functionality: Implementation of Advanced RAG and Hybrid Search techniques to ensure high-precision data retrieval and inquiry
• Summary Report Generation: A feature that segments document content into chunks for summarization, followed by the generation and visual validation of the final report

Expected outcomes

This service has improved operational efficiency by automating document creation, allowing staff to shift their focus from administrative paperwork to their core mission of facility safety. By streamlining the drafting and review process, the organization ensures higher consistency in its documentation while proactively managing legal risks through automated regulatory updates.

Launch year: 2024

Attribution

Case study courtesy of the National Information Society Agency, Republic of Korea, Apolitical's Content Partner. Sourced from Use Cases of Public AI Service, Vol.2, https://eng.nia.or.kr/site/nia_eng/ex/bbs/View.do?cbIdx=31975&bcIdx=29182&parentSeq=29237