When studying for a doctoral degree (PhD), candidates submit a thesis that provides a critical review of the current state of knowledge of the thesis subject as well as the student’s own contributions to the subject. The distinguishing criterion of doctoral graduate research is a significant and original contribution to knowledge.
Once accepted, the candidate presents the thesis orally. This oral exam is open to the public.
Crowd-sourced delivery services represent an innovative urban logistics solution that has garnered considerable attention in recent years due to its superior economic, environmental, and social benefits. Distinct from conventional freight transportation, the crowd-sourced delivery model employs ordinary people as either a supplement or an alternative to professional delivery personnel, integrating them into the delivery process. Due to the inherent uncertainties associated with these non-professional participants, manifested in their strategic refusal of orders, aiming for obtaining better matching outcomes. Furthermore, the prevalent issue of frequent order refusals leads to challenges such as repeated matching and delivery failures, posing significant challenges for the current state of crowd-sourced delivery services.
Addressing these challenges, this dissertation introduces sophisticated matching mechanisms and compensation schemes for crowd-sourced delivery systems, aiming to optimize outcomes in complex scenarios. The research begins with an empirical investigation into the determinants of driver decisions via a series of stated preference surveys. This foundational work enables the development of an accurate predictive model for driver behavior. Integrating this model into an advanced optimization framework, the study then assesses various matching and compensation strategies, considering factors like acceptance probability and decision-making processes. Further innovation is demonstrated through the proposal of an order-postponement mechanism, informed by the urgency value of deliveries. This approach aims to increase the efficiency of crowd-sourced delivery, accommodating more orders within given time windows. A key contribution of this dissertation is the introduction of the concept of reinforced matching stability. Building on this notion, a novel algorithm is proposed, demonstrably reducing order refusal rates to as low as 1% and achieving operational cost savings of up to 18%. This research not only addresses the immediate challenges of crowd-sourced delivery services but also contributes significantly to the broader discourse in urban logistics and transportation planning.