All projects

Open source /2026 /Geospatial · Full-stack · AI

Depo Yönetim Sistemi

See your warehouses on a map, analyze your supply network, track delivery trucks live, and manage every rack in 3D. An inventory system for businesses with more than one warehouse, built around location.

How it works

  1. 1

    See it on the map, analyze an area

    Warehouses sit on the map. Draw an area and its stock and occupancy appear at once.

  2. 2

    Solve the supply network

    See where demand is heavy, assign customers to warehouses, measure coverage, and find where a new warehouse should go.

  3. 3

    Track vehicles live

    Plan delivery tours; trucks move on the map in real time and show an arrival time for each stop.

  4. 4

    Manage in 3D and ask

    See racks in 3D, catch low stock from an alert pin, get the shortest picking route, and ask the AI in Turkish.

Live logistics: fleet tracking, scenario, forecast, wave picking

Real-time vehicle tracking on the map (with per-stop ETAs), a close-a-warehouse what-if scenario, Holt demand forecasting + reorder, and order wave picking — all recorded from the real app.

Problem

Enterprise WMS tools are heavy and costly for small businesses. Excel cannot show which rack a product physically sits on. Companies with several warehouses want to see stock on a map and read rack occupancy at a glance.

Solution

Warehouses sit on a map; draw an area and its stock and occupancy appear at once. On the supply side it works out where demand is heavy, the nearest warehouse for each customer, coverage gaps, and where a new warehouse should go. Delivery trucks move on the map in real time, with an arrival time for each stop. Inside, a 3D view shows how full each bin is by color, and a realistic mode brings it to life with pallets and cartons. Alert pins appear over low-stock bins, and the shortest picking route is drawn on the floor. Ask the AI a question in Turkish and it becomes a safe, read-only query that never writes SQL. FastAPI + PostGIS backend, React 19 frontend.

Results

An open-source, production-ready product. It opens with demo data via `docker compose up`, and the video and images here are recorded from the real app. The live demo runs with no server — the whole API runs in the browser, and everything, including live vehicle tracking, behaves like the real thing.