Dr.-Ing. Andreas Balster

Senior Researcher

Dr.-Ing. Andreas Balster

Senior Researcher

Andreas Balster joined Kühne Logistics University in August 2016 as a Research Associate in the fields of Freight Transportation Modeling and Computer Science in Logistics. His research topics are freight transport demand modeling, risk management in transport and logistics, critical infrastructures, food logistics and Machine Learning.

Currently, he works on the SMECS Project (SMart Event foreCast for Seaports), which is funded by the Federal Ministry of Transport and Digital Infrastructure. The project focus is on the regulation of multimodal transport networks at seaports so that existing capacities can be utilized in a resource-efficient manner. Its aim is to design transport networks in such a way that reactions to disruptions can be prepared proactively across all stakeholders.

Before joining KLU, he worked as a Research Associate at TU Darmstadt. There he played a leading role in the BMBF-project SEAK (Decision support for food supply shortfalls), where he developed a dynamic commodity flow model for the German food supply system. He also organized the 1st International Conference on Production, Logistics and Traffic. Invited by the Massachusetts Institute of Technology to be Guest Researcher in 2016, he applied his knowledge of modeling food supply systems to identify outbreak origins.

He studied Industrial Engineering at Karlsruhe Institute of Technology with focus on applied computer sciences, logistics, production technology, and finance. In his diploma thesis, which he wrote as part of the BMBF-project POWer.net (Planning and optimization of agile global production networks) at Schott AG, he developed an evaluation system for agile global production networks.



Tel: +49 40 328707-308
Fax: +49 40 328707-109


Academic Positions

since 2019Senior Researcher at Kühne Logistics University, Hamburg, Germany
2016 - 2018Research Associate SMECS Project (SMart Event foreCast for Seaports) at Kühne Logistics University, Hamburg, Germany
2016Guest Researcher at Massachusetts Institute of Technology, USA
2012 - 2016Research Associate at Technische Universität Darmstadt, Germany
2007 - 2008

Student Assistant at Karlsruhe Institute of Technology (KIT), Germany

Working Experience

2011Graduand at SCHOTT, Germany
2008 - 2011Working Student at Majolika Manufaktur, Karlsruhe, Germany


2018Doctor of Engineering, Technische Universität Darmstadt, Germany

Diploma in Industrial Engineering (Msc equivalent), Karlsruher Institut für Technologie (KIT), Germany


DOI: 10.1007/s12599-020-00653-0 

Abstract: Transparency in transport processes is becoming increasingly important for transport companies to improve internal processes and to be able to compete for customers. One important element to increase transparency is reliable, up-to-date and accurate arrival time prediction, commonly referred to as estimated time of arrival (ETA). ETAs are not easy to determine, especially for intermodal freight transports, in which freight is transported in an intermodal container, using multiple modes of transportation. This computational study describes the structure of an ETA prediction model for intermodal freight transport networks (IFTN), in which schedule-based and non-schedule-based transports are combined, based on machine learning (ML). For each leg of the intermodal freight transport, an individual ML prediction model is developed and trained using the corresponding historical transport data and external data. The research presented in this study shows that the ML approach produces reliable ETA predictions for intermodal freight transport. These predictions comprise processing times at logistics nodes such as inland terminals and transport times on road and rail. Consequently, the outcome of this research allows decision makers to proactively communicate disruption effects to actors along the intermodal transportation chain. These actors can then initiate measures to counteract potential critical delays at subsequent stages of transport. This approach leads to increased process efficiency for all actors in the realization of complex transport operations and thus has a positive effect on the resilience and profitability of IFTNs.

Export record:CitaviEndnoteRISISIBibTeXWordXML

Open reference in new window "An ETA Prediction Model for Intermodal Transport Networks Based on Machine Learning"

DOI: 10.1016/j.tre.2018.03.002 

Abstract: This paper presents a calibrated dynamic multi-scale multi-regional input–output (MSMRIO) model of the German food supply system based on real data. The model comprises 51 commodity groups from farm to fork differentiating three different temperature ranges as well as living animals. Spatially, it works on an aggregate level of 402 regions within Germany as well as its 50 most important trading nations. It determines the commodity flows and the additionally needed transport capacity in case of disruptions. Showing how changes in production, inventories, sourcing, and consumption affect commodity flows, the model uncovers vulnerabilities and makes risk evaluation possible.

Export record:CitaviEndnoteRISISIBibTeXWordXML

Open reference in new window "Dynamic freight flow modelling for risk evaluation in food supply"

Abstract: A reliable food supply is an essential part of the populations well-being. In Germany, this challenging task is being handled through the cooperation of a large number of private sector actors. The basic requirements for reliable supply are the operability of infrastructures and the availability of goods and services from other sectors. Extreme events, such as power outages, heat waves or pandemics, pose risks whose occurrence can significantly compromise this complex and dynamic food supply system, resulting in severe consequences for the population. The existing emergency plans of state and private sector actors are currently not designed to mitigate the effects of extreme events on food supply through early intervention, for example in order to cover major labour shortages or large-scale, prolonged disruptions of technical infrastructures. Better planning requires quantitative analyses that take into consideration the economic interdependence, spatial structure, and temporal dynamics of the food supply system. Purely static analyses of existing statistics are not adequate for such an integrated approach. In order to uncover possible risks, meaningful indicators such as stock developments and necessary transport capacities must be determined. For this purpose, a quantitative model is necessary that simulates the food supply and realistically maps the dependencies resulting from the economic, spatial and temporal context of production, storage, transport, trade, and consumption on the basis of real data. Existing research in this area focuses on qualitative risk analyses or limits the scope to delimited subsystems, such as individual companies, specific supply chains, or selected areas. Risks arising from the dynamics and complexity of the entire food supply system cannot yet be analysed in detail. This thesis addresses this research gap by developing a macroscopic and at the same time detailed, dynamic freight transport demand model of the German food supply system. The model called FOODFLOW differentiates 51 commodity groups in three temperature ranges. The economic interdependencies of these commodity groups are identified with an input-output analysis. The resulting sectoral, physical input-output model is extended to a multi-scale multi-regional input-output (MSMRIO) model by incorporating spatial interactions. To do so, commodity productions and demands are first allocated to 402 regions within Germany and the 50 most important international trading nations. Thereby the entire supply system including agriculture, food processing, food retailing, wholesale, and the end consumer is covered. Realistic results are ensured by calibration on official data of the Federal Transport Infrastructure Plan. The calibration method developed for this purpose combines gravitational and optimisation models. On this basis, the course of inventories and required transport capacities of all groups of goods, groups of actors, and regions are simulated. The base year for the model application is 2012 with a day-accurate resolution. Possible applications are illustrated by four examples: Initially, FOODFLOW visualises the vulnerability of the German regions using various indicators. Afterwards, the example of the EHEC outbreak of 2011 will show how the calibrated spatio-economic interactions can be used to track foodborne disease outbreaks. In addition to these static applications, FOODFLOW makes it possible to track the spatial and temporal propagation of disruption impacts in the food supply system and thus to estimate the effects for end consumers. This functionality is demonstrated by the scenarios of the closure of the Port of Hamburg and an extreme increase in demand for beverages in the greater Berlin area. The implementation of FOODFLOW shows that it is possible to develop a dynamic commodity flow model of the German food supply system by combining and interpreting available data. The high spatial resolution as well as the actor group and day-accurate mapping of system dynamics make it possible to identify developments and dependencies even during the year. For the first time FOODFLOW enables a comprehensive analysis of the vulnerabilities of the German food supply system as well as the impact on the affected population. In addition, the extent of certain responsive measures, like the resulting additional freight transport demand, and the influence of certain preventive measures can be forecasted. These results increase the transparency of the German food supply system and enable improved crisis prevention at a national level. The present work thus contributes both to research and to securing the food supply system. http://tuprints.ulb.tu-darmstadt.de/8336/

Export record:CitaviEndnoteRISISIBibTeXWordXML

Open reference in new window "Modellierung dynamischer Güterflüsse zur Analyse von Risiken in der Lebensmittelversorgung (Dissertation)"

DOI: 10.1007/978-3-030-13535-5_12 

Abstract: Intermodal logistics networks such as the maritime transport chain require a precise interaction of numerous actors. However, due to their complexity, the closely interlinked processes are highly susceptible to disruptions. Companies are constantly faced with the challenge of dealing effectively and efficiently with disruptions and resultant delays. At the same time, they are confronted with increasing logistical requirements related to higher quality and flexibility demands of customers (Straube et al. 2013). Supply chains are becoming increasingly vulnerable, due to the associated necessity to cope with increasing volatility while simultaneously reducing risk buffers in processes as a result of rising cost pressure. Combined with ongoing changes due to digitization, this situation contributes significantly to an increasing need for improved information transparency among companies and their customers.

Export record:CitaviEndnoteRISISIBibTeXWordXML

Open reference in new window "Realization of ETA Predictions for Intermodal Logistics Networks using Artificial Intelligence"

  • Balster, Andreas and Hanno Friedrich (2016): Dynamic Freight Flow Modelling for Risk Evaluation in Food supply. 14th World Conference on Transport Research, 10 – 15 July 2016, Shanghai, China.
  • Balster, Andreas and Hanno Friedrich (2015): Modelling Dynamic Commodity Flows – Using the Example of the German Food Supply Sector. 12th International Multidisciplinary Modelling & Simulation Multiconference, 21 – 23 September 2015, Bergeggi, Italy.
  • Balster, Andreas (2014): Modelling Dynamic Commodity Flows – Using the Example of the German Food Supply Sector. International Doctoral Seminar, 07 – 12 December 2014, Nagoya University, Japan.
  • Balster, Andreas and Hanno Friedrich (2014): Modelling Dynamic Commodity Flows – Using the Example of the German Food Supply Sector. 26th NOFOMA Annual Conference, 11 – 13 June 2014, Copenhagen, Denmark.
  • Motzke, A., Andres Balster, Ole Hansen, Maja Herrmannsdörfer, Frank Schätter, Hanno Friedrich, Wolfgang Raskob, Marcus Wiens and Frank Schultmann (2014): The SEAK Project: Decision Support for Managing Disruptions in Food Supply Chains. Future Security Conference 2014, 16 – 18 December 2014, Berlin, Germany.
  • Friedrich, Hanno, Andreas Balster and Ole Ottemöller (2012): Supply Chain Risk Analysis with Extended Freight Transportation Models. International Colloquium on Recent Developments in Freight Transport Modelling, 4 – 6 October 2012, Antwerp, Belgium.
  • Balster, Andreas, Hanno Friedrich, Ulrich Berbner, Li Zhang, Steffen Despotov, Patrick Kroner (2012): Measures of Supply Chain Risk Management. IDRC Davos, 26 – 30 August 2012 Davos, Switzerland.