AI-Powered Supply Chain Control Tower with BigQuery and GPT-4o

This n8n workflow demonstrates how to build an AI-powered control tower for supply chain management. It leverages Google BigQuery for data retrieval and an OpenAI GPT-4o agent for intelligent analysis and decision-making, triggered by chat messages.

What it does

This workflow orchestrates the following steps:

1. Listens for Chat Messages: The workflow is activated when a new chat message is received, acting as the primary input for user queries or commands.
2. Initializes an AI Agent: An OpenAI GPT-4o powered AI agent is set up to process the incoming chat messages.
3. Maintains Conversation History: A simple memory buffer is used to keep track of the conversation context, allowing the AI agent to understand follow-up questions and maintain continuity.
4. Provides a BigQuery Tool: The AI agent is equipped with a custom tool that allows it to execute queries against a Google BigQuery dataset. This tool is wrapped as an n8n workflow call, enabling the AI to retrieve relevant supply chain data.
5. Executes BigQuery Queries: When the AI agent determines that data from BigQuery is needed to answer a query, it uses the "Call n8n Workflow Tool" to trigger a sub-workflow.
6. Retrieves Data from BigQuery: The sub-workflow, triggered by the AI agent, connects to Google BigQuery and executes a predefined query to fetch supply chain data.
7. Processes and Responds: The AI agent receives the BigQuery results, processes them in the context of the user's chat message, and formulates an intelligent response.
8. Outputs AI Response: The final response from the AI agent is then outputted, presumably to the chat interface from which the initial message was received.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Instance: A running n8n instance.
• OpenAI API Key: An API key for OpenAI to use the GPT-4o chat model.
• Google BigQuery Account: Access to a Google BigQuery project with relevant supply chain data.
• Google Cloud Credentials: Credentials (e.g., Service Account Key or OAuth 2.0) configured in n8n for Google BigQuery.
• A Chat Application: An application integrated with n8n's Chat Trigger (e.g., Slack, Telegram, Discord, or a custom chat interface) to send and receive messages.

Setup/Usage

1. Import the Workflow:
   • Download the provided JSON file for this workflow.
   • In your n8n instance, go to "Workflows" and click "New".
   • Click the "Import from JSON" button and paste the workflow JSON.
2. Configure Credentials:
   • OpenAI Chat Model: Configure your OpenAI API Key credential in the "OpenAI Chat Model" node.
   • Google BigQuery: Configure your Google Cloud credential (e.g., Google OAuth2 API or Google Service Account) in the "Google BigQuery" node. Ensure it has the necessary permissions to query your BigQuery datasets.
3. Configure the "Call n8n Workflow Tool":
   • The "Call n8n Workflow Tool" node needs to be linked to the sub-workflow that performs the BigQuery query. You will need to create a separate n8n workflow for the BigQuery operation and then select it in this tool node.
4. Configure the "Google BigQuery" Node:
   • In the "Google BigQuery" node, specify the Project ID, Dataset ID, and the SQL Query to retrieve your supply chain data. This query should be designed to fetch data relevant to typical supply chain inquiries.
5. Activate the Workflow:
   • Once all credentials and configurations are set, activate the workflow.
6. Interact via Chat:
   • Send a message to your n8n Chat Trigger. The AI agent will process your request, potentially query BigQuery for data, and respond with insights.

This workflow provides a powerful foundation for building interactive, AI-driven supply chain analytics and control capabilities directly from a chat interface.