Programme

Here, you can find the general conference scheduling, including the different parallel sessions and presentations as well as the complete conference booklet in PDF.

• Prof. dr. Margaretha Gansterer - ​University of Klagenfurt (AT)
       "Last-mile delivery with shared resources: modeling and recent developments" 
  
  The Sharing Economy is on the rise. Traditional business models have to be adapted and players have to succeed in a world of shared idle capacities on digital platforms. Innovative concepts related to shared transportation resources include collaborative vehicle routing or crowd delivery systems. A collaboration can be described as a partnership between two or more companies to optimize operations by making joint decisions and sharing information, resources, or profits. Crowdsourced deliveries are either conducted by freelance drivers or by occasional drivers, where the former conduct several tasks over a longer time horizon typically being active on different platforms, while the latter offer their service on a more irregular basis. Both of these concepts have their advantages but also specific challenges. In case of horizontal transport collaborations we distinguish between centralized and decentralized settings. In the former, it is assumed that one fully informed decision maker exists, while in the latter, decisions under incomplete information have to be orchestrated. Crowd delivery concepts, however, highly rely on the willingness of external drivers to offer their service. Hence, attractive offers have to made. We focus on the generation of bundles of requests such that the distribution of orders assigned to external drivers and to a company-owned fleet is optimized. The talk will cover modeling approaches, recent developments, and promising future directions.  
  
  
• Prof. dr. Dolores Romero Morales - Copenhagen Business School (DK) 
       "A tour through Explainable and Fair Machine Learning with an OR lens" 
  
  The use of Artificial Intelligence and Machine Learning algorithms is ubiquitous in Data Driven Decision Making.  Despite their excellent accuracy,  these algorithms are often criticised for their lack of transparency. Algorithms such as Random Forest, XGBoost and Deep Learning are often seen as black boxes, for which it is difficult to explain their predictions. In addition, when applied to sensitive situations with consequential impact on citizens’ lives, including access to social services, lending decisions or parole applications, this opaqueness may hide unfair outcomes for risk groups. Therefore, there is an urgent need to strike a balance between three goals, namely, accuracy, explainability and fairness. In this presentation, and with an Operations Research lens, we will navigate through some novel Machine Learning models that embed explainability and fairness in their training.