• Reduce unproductive movement in a large port maritime yard without deterioring the performance of the terminal
  • Increase ressource utilization rate (cranes, forklift automatic guided vehicles, and other container handling vehicles)

Data coming from TOS

  • upstream data: port of departure, shipping company, shipping line
  • downstream data: receiving company, logistics companies involved
  • inside of a container
  • external data: weather and seasonality data


  • Machine learning and statistical modelling algorithms
  • Big Data technologies for creating analytical boards and models
  • Python

Business KPIs

  • Unproductive moves ratio
  • Container dwell time
  • Vessel dwell time
  • Congestion at gate entrance
  • Ressources useful utilization rate


  • Improved container stacking on terminal site alongside the found customer categories
  • Reduction of the unproductive moves ratio by 25%
  • Our model predicted unproductive moves with 71% accuracy
  • Main cost drivers were savings in machine utilization and respective maintenance leading to a yearly savings potential of multi million EUR