Real-Time Forecasting of Metro Origin-Destination Matrices with High-Order Weighted Dynamic Mode Decomposition

Forecasting short-term ridership of different origin-destination pairs (i.e., OD matrix) is crucial to the real-time operation of a metro system. However, this problem is notoriously difficult due to the large-scale, high-dimensional, noisy, and highly skewed nature of OD matrices. In this paper, we address the short-term OD matrix forecasting problem by estimating a low-rank high-order vector autoregression (VAR) model. We reconstruct this problem as a data-driven reduced-order regression model and estimate it using dynamic mode decomposition (DMD). The VAR coefficients estimated by DMD are the best-fit (in terms of Frobenius norm) linear operator for the rank-reduced full-size data. To address the practical issue that metro OD matrices cannot be observed in real time, we use the boarding demand to replace the unavailable OD matrices. Moreover, we consider the time-evolving feature of metro systems and improve the forecast by exponentially reducing the weights for historical data. A tailored online update algorithm is then developed for the high-order weighted DMD model (HW-DMD) to update the model coefficients at a daily level, without storing historical data or retraining. Experiments on data from two large-scale metro systems show that the proposed HW-DMD is robust to noisy and sparse data, and significantly outperforms baseline models in forecasting both OD matrices and boarding flow. The online update algorithm also shows consistent accuracy over a long time, allowing us to maintain an HW-DMD model at much low costs.

• Publication year

  2021

• Venue

  Transportation Science

• Publication date

  2021-01-02

• Fields of study

  Mathematics, Computer Science, Engineering

• Identifiers
  DOI 10.1287/trsc.2022.1128arXiv 2101.00466
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Dynamic Origin-Destination Prediction in Urban Rail Systems: A Multi-Resolution Spatio-Temporal Deep Learning Approach

  2021cited by this paper
• A Low Rank Dynamic Mode Decomposition Model for Short-Term Traffic Flow Prediction

  2021cited by this paper
• Short-term origin-destination demand prediction in urban rail transit systems: A channel-wise attentive split-convolutional neural network method

  2021influential reference
• DNEAT: A novel dynamic node-edge attention network for origin-destination demand prediction

  2021influential reference
• Short-term prediction of urban rail transit origin-destination flow: A channel-wise attentive split-convolutional neural network method

  2020cited by this paper
• A dynamic OD prediction approach for urban networks based on automatic number plate recognition data

  2020cited by this paper
• Stochastic Origin-Destination Matrix Forecasting Using Dual-Stage Graph Convolutional, Recurrent Neural Networks

  2020cited by this paper
• Data-driven analysis and forecasting of highway traffic dynamics

  2020influential reference
• Online Spatio-Temporal Crowd Flow Distribution Prediction for Complex Metro System

  2020influential reference
• Incorporating travel behavior regularity into passenger flow forecasting

  2020influential reference
• Subway Passenger Flow Prediction for Special Events Using Smart Card Data

  2020cited by this paper
• Deep Multi-Scale Convolutional LSTM Network for Travel Demand and Origin-Destination Predictions

  2020influential reference
• Hybrid Approach Combining Modified Gravity Model and Deep Learning for Short-Term Forecasting of Metro Transit Passenger Flows

  2020cited by this paper
• Multi-task Adversarial Spatial-Temporal Networks for Crowd Flow Prediction

  2020cited by this paper
• Travel Time Prediction for Congested Freeways With a Dynamic Linear Model

  2020cited by this paper
• Short-Term Origin-Destination Forecasting in Urban Rail Transit Based on Attraction Degree

  2019influential reference
• DeepPF: A deep learning based architecture for metro passenger flow prediction

  2019cited by this paper
• Contextualized Spatial–Temporal Network for Taxi Origin-Destination Demand Prediction

  2019cited by this paper
• Deep Factors for Forecasting

  2019cited by this paper
• Origin-Destination Matrix Prediction via Graph Convolution: a New Perspective of Passenger Demand Modeling

  2019cited by this paper
• An incremental approach to online dynamic mode decomposition for time-varying systems with applications to EEG data modeling

  2019influential reference
• Dynamic Origin–Destination Matrix Prediction with Line Graph Neural Networks and Kalman Filter

  2019cited by this paper
• Predicting origin-destination ride-sourcing demand with a spatio-temporal encoder-decoder residual multi-graph convolutional network

  2019cited by this paper
• Robust principal component analysis for modal decomposition of corrupt fluid flows

  2019cited by this paper
• Deep Learning Architecture for Short-Term Passenger Flow Forecasting in Urban Rail Transit

  2019cited by this paper
• Short-Term Origin-Destination Based Metro Flow Prediction with Probabilistic Model Selection Approach

  2018cited by this paper
• Autoregressive models for matrix-valued time series

  2018cited by this paper
• Network-wide Crowd Flow Prediction of Sydney Trains via Customized Online Non-negative Matrix Factorization

  2018influential reference
• Higher Order Dynamic Mode Decomposition

  2017cited by this paper
• Online dynamic mode decomposition for time-varying systems

  2017influential reference
• Efficient OD Trip Matrix Prediction Based on Tensor Decomposition

  2017cited by this paper
• Forecasting short-term subway passenger flow under special events scenarios using multiscale radial basis function networks ☆

  2017cited by this paper
• Forecasting dynamic public transport Origin-Destination matrices with long-Short term Memory recurrent neural networks

  2016cited by this paper
• Chaos as an intermittently forced linear system

  2016cited by this paper
• Ergodic Theory, Dynamic Mode Decomposition, and Computation of Spectral Properties of the Koopman Operator

  2016influential reference
• Temporal Regularized Matrix Factorization for High-dimensional Time Series Prediction

  2016cited by this paper
• A novel wavelet-SVM short-time passenger flow prediction in Beijing subway system

  2015cited by this paper
• Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting

  2015influential reference
• Dynamic Mode Decomposition with Control

  2014cited by this paper
• Dynamic mode decomposition for large and streaming datasets

  2014cited by this paper
• On dynamic mode decomposition: Theory and applications

  2013influential reference
• Variants of Dynamic Mode Decomposition: Boundary Condition, Koopman, and Fourier Analyses

  2012cited by this paper
• An error analysis of the dynamic mode decomposition

  2012cited by this paper
• Forecasting the short-term metro passenger flow with empirical mode decomposition and neural networks

  2012cited by this paper
• Estimation of latent factors for high-dimensional time series

  2011cited by this paper
• Spectral analysis of nonlinear flows

  2009cited by this paper
• Dynamic mode decomposition of numerical and experimental data

  2008influential reference
• The Generalized Dynamic-Factor Model: Identification and Estimation

  2000cited by this paper
• The approximation of one matrix by another of lower rank

  1936influential reference

• Short-Term Area Trip Attraction Prediction Based on Real-Time Human Mobility Data

  2026cites this paper
• Data-Driven Capacity Management of Ride-Hailing Service with Prescriptive Analytics and Deep Learning

  2026cites this paper
• Multi-operator free-floating GBFS trip destination prediction in public mobility sharing systems

  2025cites this paper
• Low-rank tensor learning for spatiotemporal traffic and mobility data: A survey

  2025cites this paper
• Joint Short-Term Origin-Destination Demand Prediction for Multimodal Transport Systems

  2025cites this paper
• Dynamic and Probabilistic Forecasting of Multimode Metro Passenger Flows: A Functional Data Analysis Approach

  2025cites this paper
• Detecting shifts in urban rail passenger behavior during emergencies: A data-driven comparative approach

  2025cites this paper
• Passenger flow prediction between subway stations based on congestion state parameters and composite neural network

  2025cites this paper
• Railway booking demand forecasting for revenue management: a deep probabilistic approach

  2025cites this paper
• Optimising Multimodal Integration: Global Best Practices for Seamless Connectivity and Their Applicability in Sri Lanka

  2025cites this paper
• Macroeconomic and Price of Oil and Gold on Liabilities of Hajj Fund Management Agency

  2025cites this paper
• LLM4PT: A large language model-based system for flexible and explainable public transit demand prediction

  2025cites this paper
• Graph neural networks with flow conservation constraints for real-time origin-destination matrix completion

  2025cites this paper
• Adaptive Passenger Flow Prediction in Metro Systems: An Incremental Learning Framework for Evolving Patterns

  2025cites this paper
• Collaborative optimization of vehicle formation and timetable for modular autonomous bus considering demand uncertainty:A data-driven continuous approximate method

  2025cites this paper
• Simultaneous Multimodal Demand Imputation and Forecasting via Graph-Guided Generative and Adversarial Network

  2025cites this paper
• Hybrid Learning Model of Global-Local Graph Attention Network and XGBoost for Inferring Origin-Destination Flows

  2025cites this paper
• Ride-Hailing Service Pattern Recognition and Demand Prediction: A Reinforcement Ensemble Learning With Fuzzy C-Means Clustering Approach

  2025cites this paper
• Virtual car following based cooperative control of connected automated vehicles in complex scenarios: A roundabout example

  2025cites this paper
• Modelling time series with temporal and spatial correlations in transport planning using hierarchical ARIMA-copula Model: A Bayesian approach

  2025cites this paper
• Bus Passenger Origin-Destination Flow Estimation Using Entry-Only Smartcard Data: A Self-Supervised Learning Method Without Alighting Data

  2025influential citation
• Origin-destination prediction from road average speed data using GraphResLSTM model

  2025cites this paper
• Heterogeneous multi-view graph gated neural networks for real-time origin-destination matrix prediction in metro systems

  2025cites this paper
• Multi-Graph Inductive Representation Learning for large-scale Urban Rail demand prediction under disruptions

  2024cites this paper
• Two-Stage OD Flow Prediction for Emergency in Urban Rail Transit

  2024influential citation
• MTLMetro: A Deep Multi-Task Learning Model for Metro Passenger Demands Prediction

  2024cites this paper
• Traffic Origin-Destination Demand Prediction via Multichannel Hypergraph Convolutional Networks

  2024cites this paper
• Real-time prediction of transit origin–destination flows during underground incidents

  2024cites this paper
• Spatiotemporal Implicit Neural Representation as a Generalized Traffic Data Learner

  2024cites this paper
• Spatio-Temporal Attention Hybrid Network for Short-Term Origin-Destination Flow Prediction in Urban Subways

  2024cites this paper
• A method of time-varying demand distribution estimation for high-speed railway networks with user equilibrium model

  2024cites this paper
• Space-Time adaptive network for origin-destination passenger demand prediction

  2024cites this paper
• Urban traffic analysis and forecasting through shared Koopman eigenmodes

  2024cites this paper
• Stochastic route optimization under dynamic ground risk uncertainties for safe drone delivery operations

  2024cites this paper
• Leveraging Intra-Period and Inter-Period Features for Enhanced Passenger Flow Prediction of Subway Stations

  2024cites this paper
• Enhancing origin-destination flow prediction via bi-directional spatio-temporal inference and interconnected feature evolution

  2024cites this paper
• Large-Scale Origin–Destination Prediction for Urban Rail Transit Network Based on Graph Convolutional Neural Network

  2024cites this paper
• Predicting Subway Passenger Flows under Incident Situation with Causality

  2024cites this paper
• Fast Koopman Surrogate Falsification Using Linear Relaxations and Weights

  2024cites this paper
• Passenger Flow Path Prediction Based on Urban Rail Transit AFC Data: An Example of Chengdu, China

  2023cites this paper
• Short‐term origin‐destination demand forecasting in bus rapid transit based on dual attentive multi‐scale convolutional network

  2023cites this paper
• A DeepLearning Framework for Dynamic Estimation of Origin-Destination Sequence

  2023cites this paper
• Dynamic mode decomposition type algorithms for modeling and predicting queue lengths at signalized intersections with short lookback

  2023cites this paper
• Deep Learning for Metro Short-Term Origin-Destination Passenger Flow Forecasting Considering Section Capacity Utilization Ratio

  2023cites this paper
• Designing a Novel Two-Stage Fusion Framework to Predict Short-Term Origin–Destination Flow

  2023cites this paper
• Anti-circulant dynamic mode decomposition with sparsity-promoting for highway traffic dynamics analysis

  2023cites this paper
• Analyzing and forecasting service demands using human mobility data: A two-stage predictive framework with decomposition and multivariate analysis

  2023cites this paper
• Model Generation and Origin - Destination(OD) Matrix Forecasting for metro systems

  2023cites this paper
• Metro OD Matrix Prediction Based on Multi-View Passenger Flow Evolution Trend Modeling

  2023cites this paper
• Completion and augmentation-based spatiotemporal deep learning approach for short-term metro origin-destination matrix prediction under limited observable data

  2022cites this paper
• Unsupervised Knowledge Adaptation for Passenger Demand Forecasting

  2022cites this paper
• Forecasting Sparse Movement Speed of Urban Road Networks with Nonstationary Temporal Matrix Factorization

  2022cites this paper
• Point-process modeling and divergence measures applied to the characterization of passenger flow patterns of a metro system

  2022cites this paper
• Multi-Scale STATCN: Self-Attention based Spatiotemporal Model for Short-Term Metro Origin-Destination Matrix Prediction

  2022influential citation
• Demand Estimation for Non-Working Trips: Forecasting Passengers regarding the New Open of Railway Line Maros-Barru

  2022cites this paper
• Individual Mobility Prediction in Mass Transit Systems Using Smart Card Data: An Interpretable Activity-Based Hidden Markov Approach

  2022cites this paper
• A case study of spatiotemporal forecasting techniques for weather forecasting

  2022cites this paper
• Machine learning til realtidsforudsigelser af oprindelse-til-destination efterspørgsel for jernbaner med smart card og udbudsdata

  2022cites this paper
• Discovering low-rank representations of large-scale power-grid models using Koopman theory

  2022cites this paper
• Comparing spatial interaction models and flow interpolation techniques for predicting “cold start” bike‐share trip demand

  2022influential citation
• An effective spatiotemporal deep learning framework model for short-term passenger flow prediction

  2022cites this paper
• A Heterogeneous Graph Convolution Based Method for Short-Term OD Flow Completion and Prediction in a Metro System

  2021influential citation
• Multi-View TRGRU: Transformer based Spatiotemporal Model for Short-Term Metro Origin-Destination Matrix Prediction

  2021influential citation
• Online Metro Origin-Destination Prediction via Heterogeneous Information Aggregation

  2021cites this paper
• Journal of Cycling and Micromobility Research

  year unknowncites this paper