# Logistics RAG System (Logistics Industry Knowledge Base Q&A)

A local, offline question-answering system built on retrieval-augmented generation, demonstrated on logistics-industry documents but built to be repointed at any organization's own knowledge base.

This project is a local knowledge-base question-answering system built on retrieval-augmented generation. It performs semantic search over a local vector database in response to natural-language questions, matches relevant document fragments, and returns answers grounded in those fragments rather than the model's own unaided output.

It is built and demonstrated using logistics-industry documents, but the system is described as general enough to be pointed at knowledge documents from other industries, such as finance, healthcare or law, to build a private, local question-answering deployment. It is built on the LangChain framework, uses FAISS for the vector store, and uses Ollama to run open large language models such as ChatGLM, Qwen or Llama locally, so no data processing or model inference leaves the machine.

It requires Python 3.8 to 3.11 and the faiss-cpu, langchain and ollama packages.

Website: https://github.com/phoenix-zhou/logistics-industry-RAG