TU Berlin

Bereich LogistikSELECT

Logoschriftzug des Bereichs Logistik

Page Content

to Navigation

SELECT - AI-based ETA predictions for inland navigation logistics chains


As part of a cooperative effort between the TU Berlin and various shipping companies and operators and coordinators of inland and seaport terminals, the SELECT project implements the potential of digitisation for the logistics chains of inland navigation by using innovative data technologies

Project content

Initial situation

In order to increase the attractiveness of inland waterway for freight transport, the economic efficiency and reliability of its logistics chains must be improved. At present, there are high inefficiencies due to the lack of possibilities to synchronise the individual processes, especially those of shipping companies, ports and adjacent transports. Despite the introduction of the Inland AIS infrastructure for tracking and tracing of barges some years ago, a high degree of uncertainty still exists in the planning and control of waterside logistics chains. This is due to the many dynamic influencing factors, e.g. water levels and lock utilisation rates, which affect the transport processes of barges and which so far do not allow reliable prediction of arrival times at actor interfaces.


Goals of the project and solution approach

The SELECT project combines the expertise of public research and business practice to solve this problem by providing a novel technology for the port industry. Within the scope of a cooperative effort of different actors (shipping companies as well as operators and coordinators of inland and seaport terminals) the potentials of digitisation for the logistics chain of inland waterway are implemented by using innovative data technologies.With the help of artificial intelligence (AI) methods from the field of machine learning, an IT system is being developed which is able to determine a reliable arrival time of barges at inland and seaport terminals as well as other important reference points dynamically and automatically (ETA - Estimated Time of Arrival). For this purpose, data from various actors, including transport routes, waterways, vehicles, and transhipment processes, will be used in the project and transferred into intelligent prediction models. On the basis an alignment of the prognosticated travel times with additional process and environmental information the IT-system of SELECT will supervise permanently the continuation of the further transport and will evaluate its effects on the logistic total process. When inefficiencies and disruptions are detected, situation-specific measures for process planning and control will be suggested to the actors, e.g. a ship-related allocation of suitable loading and unloading times and optimal travel speeds.This digital decision assistant is intended to enable in particular the operators of sea and inland terminals as well as shipping companies or skippers to select optimal actions with regard to the expected arrival time, taking into account the further logistic process flow. Furthermore, the technology acts as a digital interface for the transmission of ETAs, related measures and transport-related additional information between the actors involved.The SELECT project contributes to the long-term improvement of efficiency, reliability, sustainability and IT networking of the port industry players. At the same time, the project creates an important basis for the implementation of future data-based projects by evaluating the potential and restrictions of inland waterway data.


Results & Dates

Project started in March 2020

Under the direction of the Chair of Logistics of the TU Berlin, the project SELECT ("AI-based ETA predictions for inland navigation logistics chains") was launched on 1.3.2020, which is funded for 3 years as part of the "Innovative Port Technologies" (IHATEC) initiative of the Federal Ministry of Transport and Digital Infrastructure (BMVI): https://www.innovativehafentechnologien.de/projekt-select-gestartet/.

Within the framework of a cooperation with various actors of the German port industry (including inland shipping companies and operators of inland and seaport terminals), the SELECT project will tap significant potentials for the logistics chains of inland navigation from the increasing data stock, including Inland AIS. With the help of artificial intelligence (machine learning), SELECT will develop a digital decision assistant that enables inland navigation players to optimise transport processes by providing arrival time predictions (ETA) and related recommendations for action.

By increasing reliability, efficiency and environmental sustainability, the SELECT project contributes to increasing the competitiveness and attractiveness of inland navigation compared to alternative transport options.

The kick-off of the project will take place on 31.3.2020 in Duisburg.



Project Organization



Federal Ministry of Transport and Digital Infrastructure (BMVI) as part of the initiaive Innovative Harbour Technologies (IHATEC)


Project Executing organisation: 

TÜV Rheinland Consulting GmbH


Affiliated partners:

Technical University of Berlin (TU Berlin), Chair of Logistics (consortium leader)

BEHALA Berliner Hafen- und Lagerhausgesellschaft mbH

Deutsche Binnenreederei AG

Duisburger Hafen AG

Imperial Shipping Services GmbH

modal 3 Logistik GmbH


Associate partners:

Contargo GmbH & Co. KG

HVCC Hamburg Vessel Coordination Center GmbH

Rhenus PartnerShip GmbH & Co. KG


Project duration: 

01.03.2020 - 28.02.2023


Quick Access

Schnellnavigation zur Seite über Nummerneingabe