Abstract:Origin-destination (OD) flow modeling is an extensively researched subject across multiple disciplines, such as the investigation of travel demand in transportation and spatial interaction modeling in geography. However, researchers from different fields tend to employ their own unique research paradigms and lack interdisciplinary communication, preventing the cross-fertilization of knowledge and the development of novel solutions to challenges. This article presents a systematic interdisciplinary survey that comprehensively and holistically scrutinizes OD flows from utilizing fundamental theory to studying the mechanism of population mobility and solving practical problems with engineering techniques, such as computational models. Specifically, regional economics, urban geography, and sociophysics are adept at employing theoretical research methods to explore the underlying mechanisms of OD flows. They have developed three influential theoretical models: the gravity model, the intervening opportunities model, and the radiation model. These models specifically focus on examining the fundamental influences of distance, opportunities, and population on OD flows, respectively. In the meantime, fields such as transportation, urban planning, and computer science primarily focus on addressing four practical problems: OD prediction, OD construction, OD estimation, and OD forecasting. Advanced computational models, such as deep learning models, have gradually been introduced to address these problems more effectively. We have constructed the benchmarks for these four problems at https://github.com/tsinghua-fib-lab/OD_benckmark. Finally, based on the existing research, this survey summarizes current challenges and outlines future directions for this topic. Through this survey, we aim to break down the barriers between disciplines in OD flow-related research, fostering interdisciplinary perspectives and modes of thinking.

Inferring Origin-Destination Flows from Population Distribution.

A Spatial Econometric Model for Travel Flow Analysis and Real-World Applications with Massive Mobile Phone Data

Learning to Generate Temporal Origin-destination Flow Based on Urban Regional Features and Traffic Information

GODDAG: Generating Origin-Destination Flow for New Cities Via Domain Adversarial Training

Understanding City-Wide Ride-Sourcing Travel Flow: A Geographically Weighted Regression Approach

Adaptively Exploring Population Mobility Patterns in Flow Visualization

City-wide Origin-Destination Matrix Generation via Graph Denoising Diffusion

Discovering spatiotemporal flow patterns: where the origin–destination map meets empirical orthogonal function decomposition

An Interdisciplinary Survey on Origin-destination Flows Modeling: Theory and Techniques

Origin-Destination Network Generation via Gravity-Guided GAN

Predicting Origin-Destination Flow Via Multi-Perspective Graph Convolutional Network

UMOD: A Novel and Effective Urban Metro Origin-Destination Flow Prediction Method

ODT Flow Explorer: Extract, Query, and Visualize Human Mobility

Mining The Semantics Of Origin-Destination Flows Using Taxi Traces

Trend Surface Analysis of Geographic Flows

Deep Learning Models for Population Flow Generation from Aggregated Mobility Data

Deep learning based origin-destination prediction via contextual information fusion

Estimating probabilistic dynamic origin-destination demands using multi-day traffic data on computational graphs

Origin-Destination Matrix Prediction Via Graph Convolution

Dynamic OD Expansion Method Based on Mobile Phone Location