Data-driven demand and supply management for online-to-offline logistic services

Over the last decade, many new online-to-offline logistics services have emerged that have generated significant interest in the research community and the public.These services include, for example, attended home delivery, bike sharing, crowd shipping, same-day delivery, and ride hailing.Today, services like Instacart, Amazon Flex / PrimeNow, Share Now, and Uber are commonplace and represent an essential part of the modern on-demand lifestyle.The services under consideration are often characterized by the term "online-tooffline", as online platforms provide the necessary interface to book a service such as delivery, acquire a delivery person, or match a passenger and a driver.In contrast, the actual service provision takes place offline, usually by operating a number of vehicles.They can only be profitable and sustainable if service providers effectively manage supply and demand.From a methodological point of view, managing such services is both exciting and challenging, as it usually requires taking into account the behavior of actors such as customers and drivers within sequential and stochastic decision-making processes.Consider, for example, the case of attended home delivery.Here, providers have to