Adaptive stop-layout optimization for demand-responsive transit (DRT) using variable-length NSGA-II algorithm

Abstract

Stop-based demand-responsive transit (DRT) complements conventional urban public transport and helps address first/last-mile connectivity problems. However, its stop layout must balance passenger-facing service quality and operator-facing operational efficiency. Existing DRT stop-layout studies often emphasize either operational efficiency or passenger service level. They rarely examine the trade-offs among spatial stop configuration, service quality, and operational efficiency. In addition, many approaches predefine stop numbers, limiting adaptive DRT stop planning. To address these gaps, this study develops a multi-objective optimization framework that jointly considers passenger service and operational efficiency. Specifically, demand-weighted accessibility and geographical coverage measure passenger service quality, whereas network compactness and deployment scale represent operational efficiency. A variable-length NSGA-II balances these objectives and determines stop numbers endogenously. The framework is applied to the automated-shuttle DRT demonstration area in Mingzhu Bay, Guangzhou. Results show that it generates compact stop configurations with a manageable deployment scale while maintaining favorable accessibility and coverage. Ablation analysis further shows that accessibility, coverage, and compactness are core and non-substitutable objectives. Removing any of them causes excessive spatial concentration or dispersion. Finally, cross-regime comparison shows that routine-week and holiday-week demand differences mainly adjust local stop selection. The corridor-scale stop backbone remains largely unchanged, indicating that the variable-length framework identifies stable service backbones and locally sensitive stops under changing demand regimes. Overall, this study advances human-centered DRT planning and supports adaptive stop-layout planning and coordinated multi-objective optimization of automated-shuttle services in new urban districts.

The authors' abstract, as published at the source. Journal of Urban Mobility, 2026 · DOI ↗