End-to-end IoT engineering
oscarcabreragonzalez IoT Studio What isn't measured slips away
TIDE · WIND
Blog · IoT cases › Ports and logistics
The Netherlands Smart Port 7 min

Port of Rotterdam: 44 sensors that decide when a ship berths

Europe's largest port put into production an IoT platform whose first critical application measures water and wind to plan each berthing.

In a port, an hour is not just an hour: it is fuel, crew, idle cranes and ships waiting outside. The Port of Rotterdam Authority (Havenbedrijf Rotterdam) decided that the berthing window could not depend on a generic forecast, and built an IoT platform to measure the port in real time.
44
hydrological and meteorological sensors in the port
1,2 M
data points processed every day
≈ 80.000 $
estimated saving for each hour of berthing time gained
4
technology partners: IBM, Cisco, Esri and Axians

The challenge: berthing with the real tide, wind and visibility

A deep-draught vessel can only enter or leave when the water gives it enough under-keel clearance and when wind and current allow it to manoeuvre safely. If those conditions are estimated imprecisely, the port covers itself with wide margins: the ship waits longer than necessary or loads less than it could.

In 2018 the Port Authority announced a digitalisation programme with IBM to connect the port with sensors and predictive models. According to the announcement itself, gaining up to one hour in berthing time can mean savings of around 80,000 dollars for shipping lines and the port.

From the tide to the berthing decision
44 hydro/meteo sensors
→
Port network
→
IoT platform + models
→
Berth planning

The solution: a generic IoT platform and a first critical application

In January 2019 the port put into operation an IoT platform developed with IBM, Cisco, Esri and Axians. The first application in production was the hydro-meteorological one: 44 sensors spread across the port, combined with forecasting models, Rijkswaterstaat data and astronomical tide calculations.

The system measures tide height, tidal current, salinity, wind speed and direction and visibility. From the outset it processed some 1.2 million data points a day for models, systems and users.

What matters about the architecture is that the platform is generic: the water and wind application was the first, but the same sensor, network and data infrastructure serves the next ones, from quay condition to traffic management.

What the system measures
WaterTide height, current and salinity
AirWind speed and direction, visibility
External dataRijkswaterstaat and astronomical tide calculation
Volume≈ 1.2 million points a day
In production sinceJanuary 2019

The calculation: what 1.2 million data points a day mean

1,200,000 points a day across 44 sensors is about 27,300 points per sensor per day, i.e. on the order of one reading every 3 seconds per sensor (86,400 s ÷ 27,300 ≈ 3.2 s). That resolution is enough to follow a gust of wind or the turn of a current, which is exactly what decides whether a manoeuvre is authorised or postponed.

At that rate, the value lies not in storing data but in turning it into a timely decision: if the model confirms the safe window sooner, the ship berths sooner, and that is where the hour gained in the 2018 announcement appears.

Estimated time resolution per sensor
Points per sensor per day≈ 27,300
Seconds between readings≈ 3.2 s
Our own calculation: 1,200,000 points ÷ 44 sensors; 86,400 s ÷ 27,300.

Lessons for your company

  • A generic IoT platform pays for itself with the first critical application and makes all the following ones cheaper.
  • Useful data are data that arrive in time to change an operational decision: here, the berthing window.
  • Combining your own sensors with public data (Rijkswaterstaat) and models improves the forecast more than adding sensors.

Sources and references

AJOT · Port of Rotterdam puts Internet of Things platform into operation (2019) ↗ Marine Link · Port of Rotterdam teams with IBM (2018) ↗ Óscar C. González · original post on LinkedIn ↗

The economic data and results cited come from the linked sources (cases published by the companies themselves, technology integrators or academic institutions). oscarcabreragonzalez IoT Studio compiles this information for informational purposes; it does not imply any commercial relationship with the companies mentioned.

Did you find this case useful?

Comments (0)

You need an account to comment. It's free and quick. Log in or sign up
Be the first to comment on this case.

Would you like an ecosystem like this for your organisation?

At oscarcabreragonzalez IoT Studio we design and implement end-to-end IoT solutions, from the sensor to the dashboard.

Let's talk about your project