---
title: "DART: Navigating Last-Mile Heterogeneity in Instant Delivery via Distribution-Adaptive Splines"
authors:
  - "Hao Xiong"
  - "Yang Gao"
  - "Haiyong Luo"
  - "Fang Zhao"
  - "Dan Luo"
doi: "10.24963/ijcai.2026/836"
url: "https://ijcai.org/proceedings/2026/836"
pdf: "https://ijcai.org/proceedings/2026/0836.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "7521-7529"
track: "AI and Social Good"
keywords:
  - "Humans and AI: Humans and AI"
  - "Machine Learning: Machine Learning"
  - "Data Mining: Data Mining"
---

# DART: Navigating Last-Mile Heterogeneity in Instant Delivery via Distribution-Adaptive Splines

Hao Xiong, Yang Gao, Haiyong Luo, Fang Zhao, Dan Luo

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 7521-7529. DOI: [10.24963/ijcai.2026/836](https://doi.org/10.24963/ijcai.2026/836)

## Abstract

On-demand delivery platforms rely on Travel Time Estimation (TTE) to balance courier earnings and overdue risks. In collaboration with one of China's largest platforms, we address a critical "Fairness Gap" in TTE: current systems fail to capture complex delivery patterns in GNSS-denied environments, subjecting couriers handling high concurrent order volumes to disproportionate pressure due to overdue deliveries. Analyzing 1.27 million real-world trajectories, we attribute this bias to unique challenges in GNSS-denied scenarios: distributional heterogeneity, structural heterogeneity, and contextual uncertainty. To bridge this gap, we propose DART (Distribution-Adaptive Robust Timing). DART incorporates a Learnable Adaptive Spline (LAS) encoder with a gradient-driven knot migration mechanism to enhance non-linear expressiveness for outliers, significantly improving long-tail accuracy. Furthermore, a Spatio-Temporal Transition Graph (STTG) reconstructs the latent topology by integrating sequence semantics, such as Wi-Fi-sensed arrival merchant timestamps. At the same time, a Distribution Gating Mechanism characterizes delivery time distributions under distinct contexts. Through extensive experiments and large-scale online A/B testing, DART not only reduces MAE by 14.0% in complex environments but also decreases the Order Overdue Rate by 1.7% (saving $24,000 daily), demonstrating how AI effectively reconciles operational efficiency with labor fairness.