AI Agents Automate Order Intake and Shipment Updates for a Logistics Company

Industry

Logistics & Transportation

Services

AI Agent DevelopmentProcess AutomationSystem Integration

Technologies

LangGraphOpenAI GPT-4oAzure Document IntelligencePythonMicrosoft 365 APIREST APIsAzure

Challenge

A freight and logistics company received thousands of booking requests and status queries every week by email, often with PDFs, spreadsheets and scanned documents attached.

Operations teams re-keyed this information into the transport management system by hand:

Manual re-keying of orders from emails and attachments.
Errors in addresses, references and quantities causing delivery issues.
Customer service teams flooded with 'where is my shipment?' emails.
Peaks in volume creating backlogs and overtime.

The company wanted to scale volume without hiring proportionally more operations staff.

Solution

Winklix designed a set of cooperating AI agents that sit on the company's shared mailboxes and connect to its transport management system.

Email triage agent

Classifies incoming emails as new bookings, amendments, status queries or other requests, and routes them to the right agent or team.

Order extraction agent

Extracts order details from email bodies, PDFs and spreadsheets, validates them against customer master data and creates draft orders in the TMS.

Status agent

Looks up live shipment status and replies to customers with accurate, friendly updates in seconds.

Human review queue

Low-confidence extractions and exceptions go to an operations review screen, where staff approve or correct them with one click.

Freight & Logistics Company solution delivered by Winklix

The Approach

Workflow first — We mapped the real workflow with the people doing the work before choosing models or tools.
Human in the loop — Every AI output had a clear review point, confidence thresholds and an escalation path to a person.
Prototype, then scale — A working pilot on real data proved the value before the solution was rolled out more widely.
Measured from day one — Baseline metrics were captured up front so improvements could be tracked objectively.