Selected work

Representative Papers

When better traffic forecasts fail to improve signal control: a layered diagnostic study of forecast-to-decision value

Oct 04, 2026

This study investigates why improved traffic prediction fails to enhance signal control performance, focusing on the "value gap" between prediction and decision-making. We propose a hierarchical diagnostic framework and a practical auditing protocol, leveraging real-world data from Xuancheng within a closed-loop controller setting. By integrating conformal intervals, dependence-aware scenario generation, causal prediction, and exhaustive joint action search, we systematically evaluate how prediction accuracy, uncertainty quantification, and interface alignment affect control outcomes. Our findings reveal a counterintuitive phenomenon wherein reduced prediction errors paradoxically increase queue lengths. Furthermore, this work establishes the critical roles of temporal observability, action identifiability, and objective alignment, offering a precise diagnostic methodology for identifying cross-stage failures in prediction-driven traffic control systems.

0 citationsRead paper

Joint upper-bound coverage and route-choice utility: an empirical evaluation on two urban proxy tasks

Oct 04, 2026

This study investigates whether high uncertainty coverage can outperform precise prediction in improving path decision utility. Leveraging trajectory data from Beijing and Chengdu, it proposes a decoupled evaluation framework that employs a freezing protocol to disentangle speed prediction errors from actual path losses, combined with offline surrogate task construction, minimum-boundary path selection, and joint calibration techniques for empirical analysis. The work reveals that joint upper-bound coverage is not a reliable proxy for downstream path utility. Experiments demonstrate that although enhancing joint coverage improves uncertainty quantification, it paradoxically increases late arrival rates by 0.16 to 0.92 percentage points and elevates average travel time. These findings provide critical counterintuitive evidence for research on prediction-decision consistency.

0 citationsRead paper

Strip Convolution and Direction-Aware Exclusion Loss for Oriented Ship Detection

Sep 22, 2026

为解决高分辨率遥感图像中船只检测问题，提出使用条带卷积和方向感知排除损失方法，有效提升检测精度。

0 citationsRead paper

A QUBO-Inspired Computational Framework for Airport Landside Bottleneck Diagnosis and Dynamic Dispatch Optimization

Aug 09, 2026

This study addresses the multi-stage coordination bottlenecks arising from the coupling of passenger and vehicle flows during peak periods in airport landside operations. The authors develop a five-minute-resolution state evolution model that integrates passenger arrivals, vehicle supply, shuttle services, storage capacity, and roadway capacity. They introduce novel diagnostic dimensions—including a composite congestion severity index and shadow price-based levers—and, for the first time, apply QUBO (Quadratic Unconstrained Binary Optimization) modeling to landside scheduling. A hybrid approach combining finite-action model predictive control with a QUBO-inspired simulated annealing algorithm enables differentiated, bottleneck-targeted dynamic scheduling strategies. Evaluated under intense peak scenarios at Shanghai Pudong and Hangzhou Xiaoshan airports, the method reduces passenger queue lengths from 3,445 to 2,477 and from 2,053 to 1,482, respectively, while maintaining robust congestion mitigation under various disturbances.

0 citationsRead paper

Quantum-Inspired Evolutionary Neighborhood Search for Arrival-Departure Track Utilization Adjustment under Short-Term Disturbances

Jul 27, 2026

This study addresses the scheduling disruptions in large passenger railway stations caused by short-term perturbations that alter train arrival/departure times and the sequence of station resource releases. To tackle this challenge, the authors propose an optimization framework integrating a quantum-inspired evolutionary algorithm (QEA) with neighborhood search (NS). The approach models station resources as zone-level occupancy intervals and formulates a track reassignment model that enforces resource compatibility constraints while jointly minimizing total train delay and resource reallocation costs. Experimental results on perturbation scenarios derived from GTFS timetables demonstrate that, within a unified feasible solution space, the proposed method consistently outperforms the CP-SAT solver across ten test instances, reducing average total delay from 673.8 to 390.5 minutes (a 42% improvement) and decreasing per-train average delay from 4.99 to 3.73 minutes, thereby significantly enhancing disruption recovery performance.

0 citationsRead paper

Recent publications

Latest Papers

When better traffic forecasts fail to improve signal control: a layered diagnostic study of forecast-to-decision value

Oct 04, 2026

This study investigates why improved traffic prediction fails to enhance signal control performance, focusing on the "value gap" between prediction and decision-making. We propose a hierarchical diagnostic framework and a practical auditing protocol, leveraging real-world data from Xuancheng within a closed-loop controller setting. By integrating conformal intervals, dependence-aware scenario generation, causal prediction, and exhaustive joint action search, we systematically evaluate how prediction accuracy, uncertainty quantification, and interface alignment affect control outcomes. Our findings reveal a counterintuitive phenomenon wherein reduced prediction errors paradoxically increase queue lengths. Furthermore, this work establishes the critical roles of temporal observability, action identifiability, and objective alignment, offering a precise diagnostic methodology for identifying cross-stage failures in prediction-driven traffic control systems.

0 citationsRead paper

Joint upper-bound coverage and route-choice utility: an empirical evaluation on two urban proxy tasks

Oct 04, 2026

This study investigates whether high uncertainty coverage can outperform precise prediction in improving path decision utility. Leveraging trajectory data from Beijing and Chengdu, it proposes a decoupled evaluation framework that employs a freezing protocol to disentangle speed prediction errors from actual path losses, combined with offline surrogate task construction, minimum-boundary path selection, and joint calibration techniques for empirical analysis. The work reveals that joint upper-bound coverage is not a reliable proxy for downstream path utility. Experiments demonstrate that although enhancing joint coverage improves uncertainty quantification, it paradoxically increases late arrival rates by 0.16 to 0.92 percentage points and elevates average travel time. These findings provide critical counterintuitive evidence for research on prediction-decision consistency.

0 citationsRead paper

Strip Convolution and Direction-Aware Exclusion Loss for Oriented Ship Detection

Sep 22, 2026

为解决高分辨率遥感图像中船只检测问题，提出使用条带卷积和方向感知排除损失方法，有效提升检测精度。

0 citationsRead paper

A QUBO-Inspired Computational Framework for Airport Landside Bottleneck Diagnosis and Dynamic Dispatch Optimization

Aug 09, 2026

This study addresses the multi-stage coordination bottlenecks arising from the coupling of passenger and vehicle flows during peak periods in airport landside operations. The authors develop a five-minute-resolution state evolution model that integrates passenger arrivals, vehicle supply, shuttle services, storage capacity, and roadway capacity. They introduce novel diagnostic dimensions—including a composite congestion severity index and shadow price-based levers—and, for the first time, apply QUBO (Quadratic Unconstrained Binary Optimization) modeling to landside scheduling. A hybrid approach combining finite-action model predictive control with a QUBO-inspired simulated annealing algorithm enables differentiated, bottleneck-targeted dynamic scheduling strategies. Evaluated under intense peak scenarios at Shanghai Pudong and Hangzhou Xiaoshan airports, the method reduces passenger queue lengths from 3,445 to 2,477 and from 2,053 to 1,482, respectively, while maintaining robust congestion mitigation under various disturbances.

0 citationsRead paper

Quantum-Inspired Evolutionary Neighborhood Search for Arrival-Departure Track Utilization Adjustment under Short-Term Disturbances

Jul 27, 2026

This study addresses the scheduling disruptions in large passenger railway stations caused by short-term perturbations that alter train arrival/departure times and the sequence of station resource releases. To tackle this challenge, the authors propose an optimization framework integrating a quantum-inspired evolutionary algorithm (QEA) with neighborhood search (NS). The approach models station resources as zone-level occupancy intervals and formulates a track reassignment model that enforces resource compatibility constraints while jointly minimizing total train delay and resource reallocation costs. Experimental results on perturbation scenarios derived from GTFS timetables demonstrate that, within a unified feasible solution space, the proposed method consistently outperforms the CP-SAT solver across ten test instances, reducing average total delay from 673.8 to 390.5 minutes (a 42% improvement) and decreasing per-train average delay from 4.99 to 3.73 minutes, thereby significantly enhancing disruption recovery performance.

0 citationsRead paper