PhD candidate in Data Driven Synchromodality

Research / Academic

The Amsterdam Business School (ABS), part of the Faculty of Economics and Business, has a vacancy for a PhD candidate in Data Driven Synchromodality.

Synchromodality is a highly powerful and promising concept for boosting the efficiency of freight transportation, based on combining multiple transportation modes (barges, trucks, trains) in a smart way. This makes a transition possible from the delivery of plain logistic services to integrated services by exploiting the complementary nature of available transportation modes.

In this data driven research, we first identify the particular business problems that require modelling. The aim behind the modelling is to create a purely data driven predictive model (without developing a mathematical model upfront) that provides reasonably accurate predictions in the complex domain and then to use machine learning approaches to find the optimum synchromodal assignment. The challenge being, that the more complex a domain is the harder it is to make good predictions, as more implicit domain knowledge is required that is not always available. This is especially true in a complex supply chain domain where there are soft factors like the interaction of the conflicting and cooperating objectives of the stakeholders and system dynamics of the uncertainty of the transport network play a significant role.

This research would be embedded within the NWO project titled COMET-PS and would be jointly supervised by researchers in UvA and VU. The goal of this research is to design and develop predictive and prescriptive models, that could eventually be incorporated into a business intelligence dashboard and then use these predictions to optimize the synchromodal allocations. As a result, one would (i) understand the nature and origin of data that allows the system user to determine the quality of the data to perform the data cleaning; (ii) understand the factors in the domain that influence the predicted variable, leading the developer to determine which variables need to be included in the predictive model; (iii) develop predictive models that are usable and interesting within the domain in terms of predictive power, integrating with existing infrastructure, and integrating with business rules & processes; and finally (iv) use the predicted data to find the optimize synchromodal assignment.


We expect the candidate to have:

  • a very good MSc degree (or is about to complete it) in applied mathematics, optimization, operations research, artificial intelligence, machine learning, computer science, systems & control or a related field;
  • strong analytical skills and must be able to work at the intersection of several research domains;
  • having exhibited their ability to perform research in data analytics and/or combinatorial optimization, are especially encouraged to apply;
  • the drive to publish in top-level academic journals in the field of logistics and data analytics;
  • mastery of both written and spoken English.

Salary Benefits:

You will be appointed for an initial period of 18 months with a possibility to extend it for another 2,5 years, pending positive evaluation. As part of your contract, you will spend 20% of your time on teaching. You will be classified as PhD candidate (promovendus) in the Dutch University job-ranking system (UFO), providing a gross monthly starting salary of €2,325 to €2,972 gross per month with an additional end-of-year bonus (8,3%) and holiday allowance (8%), in accordance with the Collective Labour Agreement of Dutch Universities

Work Hours:

38 hours per week


Spui 21