Interpretable Machine Learning for Traffic Congestion Prediction: Unveiling the Impact of Different COVID-19 Periods

2026-08-02

AI Summary

The COVID-19 pandemic substantially altered travel behavior, complicating traffic congestion forecasting. This study systematically evaluates the mechanisms influencing congestion across three distinct phases—pre-pandemic, lockdown, and post-reopening—in Alameda County, California, integrating weather, seasonal, and pandemic-related variables. An adaptive hyperparameter selection strategy is proposed, combining multiple models including SVR, MLR, RNN, and bidirectional LSTM, with recursive feature elimination and cross-validation for robust feature selection. Results demonstrate that the bidirectional LSTM model achieves overall superior performance. Notably, rising new case counts and hospitalization rates significantly suppress congestion, whereas high fuel prices exacerbate it by encouraging private vehicle use. Employing Integrated Gradients and SHAP for interpretability, this work provides the first dynamic analysis of key factor contributions across different pandemic stages.

Abstract

Traffic congestion prediction is essential for congestion mitigation, but the COVID-19 pandemic and related control measures altered travel behavior and increased prediction complexity. This study predicts congestion in Alameda County, California, during pre-lockdown, lockdown, and post-lockdown periods. Weather, seasonality, and COVID-19 variables are incorporated, and Recursive Feature Elimination with Cross-Validation is used to select important features and reduce overfitting. Support vector regression, multiple linear regression, recurrent neural networks, and long short-term memory networks are trained and optimized. Because LSTM is more sensitive to hyperparameter settings, an adaptive parameter selection approach is used, while SVR and RNN are manually tuned. Performance is evaluated using Normalized Root Mean Square Error. Bidirectional LSTM consistently performs best across all periods because it captures temporal dependence in both directions. Integrated Gradients is used to interpret Bi-LSTM predictions, and SHapley Additive exPlanations is applied to SVR. New COVID-19 cases have a mainly negative effect on congestion during lockdown and post-lockdown, likely due to greater risk awareness, voluntary travel reduction, and compliance with mobility restrictions. In the post-pandemic period, higher hospitalization reduces travel and congestion, while higher fuel prices do not prevent a shift toward private vehicles and therefore increase congestion.

Problem

• traffic congestion prediction
• COVID-19
• travel behavior
• pandemic periods
• interpretability

Innovation

• Interpretable Machine Learning
• Bidirectional LSTM
• Traffic Congestion Prediction
• Feature Selection
• SHAP