Michael Hyland

Taxonomy of shared autonomous vehicle fleet management problems to inform future transportation mobility

This paper presents a taxonomy for classifying vehicle fleet management problems, across several dimensions, to inform future research on autonomous vehicle (AV) fleets. Modeling the AV fleet management problem will bring about new classes of vehicle routing, scheduling, and fleet management problems; nevertheless, the existing literature related to vehicle routing, scheduling, and fleet management is a valuable foundation for future research on the AV fleet management problem. This paper classifies the broadly defined AV fleet management problem by using existing taxonomic categories in the literature; adds additional, or more nuanced, dimensions to existing taxonomic categories; and presents new taxonomic categories to classify specific AV fleet management problems. The broadly defined AV fleet management problem can be classified as a dynamic multivehicle pickup and delivery problem with explicit or implicit time window constraints. Existing studies that fit into this class of fleet management problems are reviewed. New taxonomy categories presented in this paper include fleet size elasticity, reservation structure, accept–reject decision maker, reservation time frame, ridesharing, vehicle repositioning, underlying network structure, and network congestion. Two goals of the taxonomy presented in this study are to provide researchers with a valuable reference as they begin to model AV fleet management problems and to present novel AV fleet management problems to spur interest from researchers.

Dynamic Assignment-Simulation Methodology for Multimodal Urban Transit Networks

This paper presents an integrated transit assignment-simulation tool. Finding least cost hyperpaths in a large-scale network and assigning travelers onto these paths are computationally challenging problems. Moreover, modeling the spatial and temporal complexities in a transit network that result from the discontinuities in transit events, such as missing a connection and not receiving a seat, exacerbates the issue of capturing realism. These challenges are overcome by (a) using a least cost hyperpath algorithm that captures the multimodal, multipattern, time-, and approach-dependent features of a transit network to provide realistic optimal strategies; (b) using a gap-based assignment approach to reach fast convergence; and (c) developing a multiagent particle simulation platform that is able to capture the heterogeneities and the discontinuities in travel. The platform was tested on the Chicago Transit Authority network of 14,000 nodes and 64,000 links; 1.25 million travelers were assigned and simulated, along with 21,000 transit vehicles. The assignment-simulation framework can be used as a network evaluation tool to assist decision making at the strategic and operational levels.

Characterization of Taxi Fleet Operational Networks and Vehicle Efficiency: Chicago Case Study

Taxi fleets serve a significant and important subset of travel demand in major cities around the world. This paper characterizes the Chicago taxi fleet operational network using complex network metrics and analyzes the operational efficiency of individual taxis over the past four years using an extensive taxi-trip dataset. The dataset, recently released by the city of Chicago, includes the pickup and drop-off census tracts and time stamps for over 100 million taxi trips. The paper explores year-over-year changes in the spatial distribution of Chicago taxi travel demand. The taxi pickup and drop-off census tract locations are modeled as nodes, and links are generated between unique pickup and drop-off node pairs. The analysis shows that high-demand pickup and drop-off location pairs in 2013 generated similar trip volumes in 2016; however, the low-demand pairs in 2013 generated significantly fewer trips in 2016. Additionally, this paper presents temporal efficiency and spatial efficiency metrics. The temporal efficiency metric determines the percentage of in-service time taxis are productive (i.e., transporting travelers), rather than empty. The spatial efficiency metric measures the percentage of taxi miles that are productive (i.e., loaded), rather than empty. The efficiency analysis of the Chicago taxi fleet shows that, for most taxis, around 50% of their in-service time and travel distance are unproductive. This inefficiency negatively affects the profitability of individual drivers and the fleet, traffic congestion, vehicle emissions, the service quality provided to customers, and the ability of taxi services to compete with emerging mobility services.

Schedule Consistency for Daily Activity Chains in Integrated Activity-Based Dynamic Multimodal Network Assignment

The dynamic multimodal network assignment problem at the daily schedule level is addressed by integrating an activity-based model and a dynamic traffic assignment tool through a unified framework. The framework achieves this integration while retaining disaggregated individualized information. The problem is formulated as a fixed-point problem, and equilibrium is achieved by minimizing the gap between the expected travel time, which is used by the activity-based model to generate the travelers’ individual and household activity schedules, and their experienced travel times, simulated by the dynamic traffic assignment tool. The schedule adjustment problem for individuals and households is formulated as a linear optimization problem. Two measures—inconsistent-schedule penalty and number of households with unrealistic schedules—are defined to monitor the status of the equilibrium and convergence gap of the integrated system. To ensure convergence of the applied integration, heuristic strategies for selecting individuals for schedule adjustment and path swap are tested in a subarea network of Chicago, Illinois. Selecting individuals for schedule adjustment based on their inconsistent-schedule penalty reduces both defined measures significantly and leads to the convergence of the planned schedule and the experienced (i.e., simulated) schedule.

Integrated Mode Choice and Dynamic Traveler Assignment in Multimodal Transit Networks: Mathematical Formulation, Solution Procedure, and Large-Scale Application

This paper introduces an integrated mode choice–multimodal transit assignment model and solution procedure intended for large-scale urban applications. The cross-nested logit mode choice model assigns travelers to car, transit, or park-and-ride. The dynamic multimodal transit assignment–simulation model determines minimum hyperpaths and assigns and simulates transit and park-and-ride travelers iteratively until the network approaches a state of equilibrium. After a given number of iterations, the updated transit network travel times are fed into the mode choice model and the model reassigns travelers to transit, car, or park-and-ride. The outer feedback loop between the mode choice model and the transit assignment model continues until the mode probabilities for each traveler do not change between iterations. A unique contribution of the method presented in this paper is that it reaches mode choice convergence with the use of disaggregate agents (travelers) instead of aggregate modal flows at the origin–destination level. The integrated model is successfully implemented on the Chicago Transit Agency’s bus and train network in Illinois. Different procedures for reaching convergence are tested; the results suggest that a gap-based formulation is more efficient than the method of successive averages.

Analytical Five-Phase Bus Rapid Transit Traffic Flow Model

Bus rapid transit (BRT) systems are becoming increasingly popular in cities worldwide because of their (a) efficiency and reliability advantages over traditional bus service and (b) cost advantages over rail transit systems. As transportation decision makers consider the implementation and planning of BRT systems, it is important that they be able to analyze different operational components of these systems. This paper describes an analytical five-phase BRT traffic flow model that is able to model the movement of a bus throughout an entire BRT corridor and network. The five-phase model includes (a) a queuing model to determine the time a bus spends waiting for access to the loading area, (b) an access time model to determine the time that it takes a bus to access a loading area position from the queue when a loading position becomes available, (c) a nonlinear dwell time model to determine the time that a bus spends at a loading area position, and (d and e) a two-part model of the following behavior of buses between bus stations, dependent on whether there is a bus between the following bus and the approaching station. The five-phase BRT traffic flow model provides a comprehensive modeling framework for a networkwide simulation of a separate right-of-way BRT system. The model builds on research in the areas of car-following (and more recently bus-following) models, dwell time models, and bus station queuing models.