Private AI for Logistics and Freight Forwarding
Private AI for logistics means models for forecasting, dispatch and document processing run on servers assigned to you and connect directly to your TMS. Shipment and customer data stay in Germany, and costs are a fixed monthly rate instead of per request.
What slows logistics companies down today
Driver and skills shortage
The German haulage association BGL expects a shortage of 120,000 truck drivers. Dispatchers are scarce too, and they spend their day on phone calls, email and status requests instead of planning and customer care.
Margin pressure and empty runs
CO₂ toll, diesel and staff costs are rising faster than freight rates, while margins in transport stay thin. At the same time, a large share of truck trips still runs empty.
Manual processes and media breaks
Orders arrive by email and PDF, delivery notes and CMR consignment notes on paper. TMS, telematics and accounting are barely connected, so data is typed in several times: slow, expensive and error prone.
What we do for your logistics business
Your own language model, behind your firewall
Self-hosted LLMs on your own servers or in a private cloud. As easy to use as ChatGPT, while customer data, rates and tenders stay in your company.
- Knowledge assistant for tariffs, SOPs, dangerous goods and customs rules (RAG)
- Auditable usage, ready for GDPR, the EU AI Act and customer audits
Orders from email and PDF straight into your TMS
AI reads transport orders, delivery notes, CMR consignment notes and invoices, structures the data and creates the records in your TMS. Your team reviews instead of retyping.
- Freight invoice checks against agreed rates
- Customs and dangerous goods documents processed in a structured way
Relieve dispatch, keep customers informed automatically
Status requests, ETA updates, pallet accounts, damage reports: recurring communication runs automatically, so your dispatchers plan instead of making phone calls.
Forecast utilization, volumes and demand
Forecasts based on your order and telematics data: shipment volumes, staff and capacity needs, ETAs and fewer empty kilometers, run on our forecasting platform Orpheon.
Integration: AI in the systems you already use
Your business runs on a TMS, telematics and an ERP, and we do not change that. We connect AI systems through interfaces, so results arrive where your team already works.
TMS integration
Captured orders, status updates and checked documents land directly in your transport management system, through an interface instead of copy and paste. No system change needed.
Telematics and fleet data
Position, driving and vehicle data flow automatically into ETA forecasts, capacity planning and customer communication.
Email, EDI and customer portals
Orders and status requests from every channel are captured and answered automatically, including customers without their own EDI connection.
ERP and accounting
Checked freight invoices and receipts flow into ERP and accounting in a structured way, including e-invoicing (XRechnung, ZUGFeRD) and DATEV export.
How a project with us runs
Audit
Analyze processes, system landscape, data quality and savings potential
Design
Architecture, model selection, interfaces and access concept
Build
Implement AI systems, workflows and the TMS and telematics integration
Deploy
On-premise or private cloud, security and CI/CD included
Operations
Monitoring, updates, governance and continuous optimization
Is this right for your company?
Best fit
Freight forwarders and logistics providers with about 20 or more employees, their own TMS, high order and document volumes and the goal of handling more shipments with the team they have.
Your advantage
AI as easy as ChatGPT, without data leaving your company: customer data, rates and tenders stay in house. Predictable costs instead of per-token billing, no dependence on US cloud providers.
Measurable impact
Significantly less time spent on order entry, relieved dispatchers, fewer empty kilometers through better forecasts, more shipments per employee.
What private AI for logistics runs in practice
Typical first use cases are volume forecasts per depot and weekday, automatic reading of delivery notes and freight documents, and suggestions for dispatch when orders change at short notice. Each use case starts as a pilot with your own data and moves into operations only when it delivers. We connect to the TMS through its existing interfaces and run the system after go-live.
Frequently asked questions
Free AI audit for your logistics business
In 60 minutes we show you which processes in your company can be automated, how the connection to your TMS works and what that saves in concrete terms.