Weather-Based Delivery Holds in n8n: Open-Meteo + Slack

Automatically place delivery holds when severe weather hits—using Open-Meteo, Postgres, Gemini, ServiceNow, and Slack

This n8n workflow polls Open-Meteo every 15 minutes and, when conditions like heavy snow, flood warnings, or low visibility are detected, it identifies impacted stores, creates a ServiceNow incident, logs the alert in Postgres, and notifies your team in Slack.

What this workflow does

• Runs on a 15-minute schedule, loading a configured region, zip code, and coordinates.
• Calls the Open-Meteo API for the current conditions and daily forecast, then evaluates whether the region meets thresholds for heavy snow, flood warning, or low-visibility risk.
• Prevents duplicate holds by querying Postgres for an existing active alert for the same region and event type within a cooldown window.
• If no active duplicate exists, it inserts a new processing record in Postgres and fetches today’s delivery manifest rows for the region’s zip code.
• Uses Google Gemini via an agent with structured output to determine which stores are impacted and to generate an incident summary, priority, and estimated delay.
• Creates a ServiceNow incident using the AI-generated details, then updates the Postgres alert record as active (or no_impact if no deliveries are affected).
• Sends an operations summary to Slack.
• If the workflow fails at any point, it logs execution errors to Postgres and sends a Slack alert including the execution link.

Use cases

• Proactively managing weather-based delivery holds for specific regions and zip codes.
• Reducing operator workload by automating ServiceNow incident creation with AI-generated summaries.
• Ensuring consistent alerting through cooldown-based duplicate suppression in Postgres.

Technical details

• Open-Meteo (weather data via API)
• Postgres (alert log, cooldown checks, delivery manifest lookups, error logging)
• Google Gemini (agent with structured output for store impact, summary, priority, estimated delay)
• ServiceNow (incident creation)
• Slack (operations and failure notifications)
• n8n logic nodes including if, set, and other supporting flow controls