Shipment data
Bookings from emails and PDFs
Transport orders and booking requests arrive in every format. Shipment, routing and scheduling data is pulled out and handed to your TMS, and a missing field is flagged for your team.
Industries · Logistics
AI for logistics companies in shipment data, operations and warehousing
theBlue.ai builds custom AI systems that pull shipment data out of emails and PDFs, answer status requests, read transport documents and watch the sensor data of vehicles and equipment. Each system connects to the TMS, WMS and ERP you already run, such as SAP or your own applications, on-premise, in the cloud or hybrid.
AI in logistics: shipment data from emails and freight documents
AI in logistics reads information out of emails, PDFs and other documents automatically. It picks up the relevant details, such as shipment data, dates, addresses or changes, and makes them usable for the next step.
In freight forwarding this covers work like order processing and data entry. The structured information can then be handed to systems such as TMS, WMS or ERP.
The gain: less manual data entry, faster processing and less effort on recurring cases. Unclear or incomplete documents still need validation and a person who checks them.
Project example · Fr. Meyer’s Sohn
Thousands of shipment emails, and no two looked alike
Fr. Meyer’s Sohn, an international logistics and shipping company, receives operational emails with shipment, routing and scheduling data in German and English, in no fixed format. Rule-based extraction could not keep up with the variety, so the data was pulled out by hand. A GPT-powered extraction pipeline now takes it out of every email automatically.
2
Languages, German and English, processed automatically
PoC → prod
From proof of concept to production deployment
Flagged
A missing field is marked for the team rather than guessed
Some of the organisations we have built for
What AI takes over
Nine processes AI takes over in logistics
Logistics runs on emails, documents and status updates. AI takes over in order intake, document checks, customer service, planning, the fleet and the warehouse.
Documents
Transport documents matched
Delivery notes, bills of lading, proofs of delivery and freight invoices are read and matched against the order, and differences in quantities, dates or prices are flagged.
Customer service
Status requests answered
Where is my shipment? Customers ask by email or chat and get an answer from the current data in your TMS, and exceptions go to the responsible dispatcher.
Sales
Quote requests prepared
Origin, destination, weight, dimensions and dates are pulled out of quote requests, and a draft quote based on your rates is ready for the sales team to check.
Planning
Capacity and volume figures on demand
Planners and branch managers ask for volumes, capacity and delivery performance in plain language, by text or voice, straight from SAP, the data warehouse or your own reporting.
Fleet and equipment
Anomalies in vehicle and equipment data
Sensor data from vehicles, reefer units, conveyors and sorters is monitored for anomalies and failing sensors, and maintenance follows the actual condition of each machine.
Assets
Location and state of assets
Trailers, containers and equipment report where they are and what state they are in from a minimal set of sensors, and the models keep working where the connection drops.
Warehouse
Counting at gates and docks
Compact devices count people, vehicles or movements at entrances, gates and loading docks from sensor data and camera images, processed on the device.
Knowledge
Answers from handling instructions
Dispatchers and warehouse staff ask about customer-specific handling instructions, procedures and contracts in plain language and get the answer with the source.
How it works
From a booking email to structured shipment data
From: customer, export department
Subject: Booking 2 x 40ft HC to Shanghai New email
Subject: Booking 2 x 40ft HC to Shanghai New email
Please book 2 x 40ft high cube, machine parts, 18,400 kg each, for pickup at our Hamburg warehouse on 14 March at 8:00.
Destination Shanghai. Our reference: PO-55821.
Booking · draft
Reference
PO-55821✓
Pickup
Hamburg, 14 March, 08:00✓
Cargo
2 x 40ft HC, 18,400 kg each✓
Consignee
Not in the email, flagged
Sent to the TMS · 1 field to add
Illustrative example with invented data, based on the Fr. Meyer’s Sohn case study.
Built for logistics
What AI in a logistics company has to handle
Every sender with its own format, several countries and languages, and data that has to be right before it moves on.
No fixed format
Formats that vary by sender, country and language, with different conventions for dates, addresses and reference numbers.
Missing fields flagged
Where a value is not in the email or document, the system marks it for your team and keeps the data in the TMS trustworthy.
Connected to your systems
TMS, WMS, ERP such as SAP or your own applications, behind an API, on-premise, in the cloud or hybrid.
A person where it matters
The system extracts, matches and prepares. Exceptions, claims and every decision stay with your dispatchers and teams.
Case studies
Shipment emails, order intake and sensor data
Our logistics project, and two projects from related fields: orders pulled out of customer emails, and the location and state of equipment from sensor data.
–80%
Manual effort
On-prem
Deployed
Logistics · Fr. Meyer’s Sohn
Routing data pulled from thousands of shipping emails a day
Operations staff read German and English logistics email every day to pull out routing and scheduling data. A pipeline now extracts it, structures it and delivers it automatically.
Read the case study
–90%
Manual intervention
95%+
Match accuracy
Manufacturing · Radaway
Order data read straight out of customer email
Orders arrive as free-form email in whatever wording the customer chose. We took an existing LLM system the last stretch to production: semantic product matching, attachment processing and 90 percent less manual work.
Read the case study
Minimal
Sensors needed
Offline
Capable
Manufacturing & IoT · under NDA
Elevator status read from minimal sensors, even offline
Maintenance teams inspected elevators by hand to determine their status. A model now reads position and operating state from few sensors, which lays the groundwork for fleet-wide predictive maintenance.
Read the case studyHow to start
Start with one process and the data behind it
A booking inbox, a document check or a fleet of sensors: the work your team does by hand today, and the data it runs on. We check early whether the data carries the use case and where it may be processed.
Describe your process
Tell us where the manual work sits and which data it involves. We come back within one business day with an initial assessment and a proposal for a 30-minute scoping call.
Describe your processStart with the process analysis
Workflows mapped, your data sources and requirements checked, and an architecture proposed with scope, timeline and cost. A standalone engagement with no commitment to proceed.
See the AI Discovery WorkshopFAQ
Questions about AI in logistics
In logistics, AI takes over work that dispatchers and back-office teams otherwise do by hand: booking requests and transport orders from emails and PDFs, delivery notes and freight invoices, status requests from customers, quote requests, capacity figures for planning, sensor data from vehicles and equipment, and counts at gates and docks. The AI system prepares the result, and dispatchers and teams decide.
Yes. For Fr. Meyer’s Sohn, an international logistics and shipping company, theBlue.ai built a pipeline with GPT-3.5 and GPT-4 that reads operational emails in German and English and returns shipment, routing and scheduling data in a structured format. Manual extraction effort fell by 80 percent.
A missing field is flagged rather than guessed. That keeps the data that flows into your TMS or ERP trustworthy, and your team only has to add what the email or document did not contain.
Yes. theBlue.ai built models that read the location and state of elevators from a deliberately minimal sensor set and keep working where the connection does not, and machine learning pipelines that detect anomalies and failing sensors in the operational data of more than 100 devices with 50 or more sensors each.
Yes. At Fr. Meyer’s Sohn the extraction runs behind an API in a Docker container on the company’s own servers, inside the existing workflow. theBlue.ai also deploys in the cloud or hybrid and connects to TMS, WMS, ERP such as SAP or your own applications.
With one process, such as booking entry from customer emails, and the data behind it. The process analysis has a fixed price from €3k and ends with an architecture proposal that states scope, timeline and cost, including where the data may be processed. The build is priced in milestones, and first working components typically arrive six to eight weeks in.
Tell us which process is costing you the most
Describe the process and we’ll come back within one business day with an initial assessment and a proposal for a 30-minute scoping call.
- Phase 1 is a standalone engagement, with no commitment to proceed further.
- Senior engineers from day one. No account managers, no handoffs.
- We integrate, never replace. Your systems stay exactly as they are.