The unit selection problem aims to identify a set of individuals who are most likely to exhibit a desired mode of behavior, which is defined in counterfactual terms. A typical example is that of selecting individuals who would respond one way if encouraged and a different way if not encouraged. Unlike previous works on this problem, which rely on ad-hoc heuristics, we approach this problem formally, using counterfactual logic, to properly capture the nature of the desired behavior. This formalism enables us to derive an informative selection criterion which integrates experimental and observational data. We show that a more accurate selection criterion can be achieved when structural information is available in the form of a causal diagram. We further discuss data availability issue regarding the derivation of the selection criterion without the observational or experimental data. We demonstrate the superiority of this criterion over A/B-test-based approaches.

This dissertation studies the traffic management and resource allocation problems under the context of multimodality in aviation systems including Unmanned Aircraft Systems (UAS) and the legacy aviation system. The goal is to contribute toward a safer, more efficient, and sustainable aviation system with emerging vehicles involved. We cover three parts in this thesis: 1) UAS Traffic management (UTM) and resource allocation; 2) Multi-modal strategies for congestion reduction integrating UAVs and ground vehicles; 3) Multi-modal regional Air Traffic Management (ATM) for commercial aviation operations. In part 1, we address high density UAV delivery operations in low-altitude urban airspace. We propose a UTM framework including UAV path planning algorithms, traffic management models with Conflict Detection and Resolution (CD&R), and mechanism design for airspace resource allocation. We find that strategic UTM can enable a sizable share of UAV deliveries with relatively low congestion. In Part 2, we explore the potential benefit of operating multiple delivery modes synergistically integrating traditional trucks, electric cargo bikes and UAVs. Multimodal systems are increasingly perceived as the future solution to help reduce traffic congestion, air pollution, noise, and safety concerns. We conceptualize and optimize several different multimodal strategies considering the congestion effects. We work on zone-based multimodal delivery strategies in multi-echelon networks using optimization models and Continuous Approximations (CAs). Then, a simulation-based congestion model based on Macroscopic Fundamental Diagrams (MFD) is proposed to capture the congestion impact from both background road traffic and delivery traffic on system delay. This work will eventually determine whether and under what conditions UAV delivery can mitigate road congestion in a cost-effective manner. In part 3, we study the multimodal systems in commercial aviation, integrating air with ground transportation to enable flight diversion in metropolitan regions with multiple airports, to improve the efficiency of ATM. We analyze the interdependent capacity profiles of Multi-Airport Region (MAR) airports and redistribute airport traffic to make more efficient use of the available capacity. Both the deterministic and stochastic versions of a flight diversion model are proposed, where the deterministic model is for short-term flight rescheduling and shifting, and the stochastic model is for flight shifting in the original schedule. Results show that by reassigning the landing airport and time of flights, the total flight delay in the New York MAR could be significantly reduced, even when a high airport reassignment cost is assumed.