Industries
Logistics & supply chain
TMS/WMS integrations, tracking, route and load planning, driver apps and forecasting — systems that keep moving when a carrier API times out.
What are the typical challenges in logistics software?
Logistics software is an integration problem before it is a software problem. A typical operator talks to dozens of carrier, customs, warehouse and ERP systems, each with its own failure modes, and still has no single view of an order from warehouse to doorstep. Planning lives in spreadsheets maintained by a few experts, and CMR notes, invoices and customs declarations are keyed in by hand. The EU eFTI Regulation (Regulation (EU) 2020/1056) obliges authorities to accept freight transport information electronically, which raises the bar for paper-based flows, and the World Bank Logistics Performance Index (2023) shows that timeliness and tracking are what separate leaders from the rest.
- Dozens of carrier, customs and ERP integrations, each with its own failures
- No single view of an order from warehouse to doorstep
- Planning done in spreadsheets by a few experts
- Documents (CMR, invoices, customs) processed by hand
What do we build for logistics and supply chain?
We build order and shipment tracking platforms with customer portals, WMS/TMS integrations over EDI and carrier or customs APIs, offline-first driver and warehouse mobile apps with barcode scanning, and event-driven back ends on queues that absorb high volumes of tracking events. Every third-party API is treated as unreliable by default: timeouts, retries, idempotent writes, fallbacks and monitoring are part of the design, not the incident review. Digital consignment notes follow the e-CMR Additional Protocol to the CMR Convention (UNECE, 2008), so paperwork can go electronic where your partners accept it.
- Order and shipment tracking platforms with customer portals
- WMS/TMS integrations, EDI, carrier and customs APIs
- Driver and warehouse mobile apps, offline-first with scanning
- Event-driven back ends on queues for high-volume tracking events
Where does AI move the numbers in logistics?
AI moves logistics numbers in forecasting, optimisation and paperwork. McKinsey (2021) reported that early adopters of AI-enabled supply-chain management improved logistics costs by 15%, inventory levels by 35% and service levels by 65% relative to slower competitors. The lever is also environmental: the IEA (2023) attributes a large share of transport emissions to road freight, so better load factors and routing cut fuel as well as cost. We build demand and ETA forecasting, route, load and slot optimisation, Document AI for shipping paperwork, and anomaly detection on delays and costs, always on top of clean event data.
- Demand and ETA forecasting
- Route, load and slot optimisation
- Document AI for shipping paperwork
- Anomaly detection on delays and costs
Heuristics vs ML for routing and planning: which do you need?
Classical heuristics and solvers (OR-Tools, VRP heuristics, MILP) remain the right tool for routing itself: they are deterministic, explainable and fast. Machine learning earns its place in the inputs, predicting demand, travel time and dwell time that the solver then optimises against. Replacing a solver with a model is rarely the answer; feeding the solver with better predictions usually is.
| Criterion | Heuristics and solvers | ML models |
|---|---|---|
| Best used for | Route, load and slot assignment | Demand, ETA and dwell-time prediction |
| Data needed | Constraints and distances | Months of historical events |
| Explainability | Full: constraints are explicit | Partial; needs monitoring |
| Adaptation to change | Edit constraints | Retrain on new data |
| Typical combination | Solver consumes predicted inputs | Model feeds the solver |
Why Glanit for logistics?
Because resilience to unreliable partners is our default posture, learned across 16 years and 500+ projects in retail, logistics and event-heavy platforms. Gartner (2024) continues to rank supply-chain visibility and AI among the top investment priorities of chief supply-chain officers, and visibility is only as good as the integration layer beneath it. We build that layer with queues, idempotency and monitoring, then add forecasting and optimisation on top. Security testing for customer portals and APIs is covered under security services.