Learning Minimally-Congested Drive Times from Sparse Open Networks: A Lightweight RF-Based Estimator for Urban Roadway Operations

TL;DR

This paper addresses the need for accurate travel-time estimates without relying on data-intensive congestion models. It introduces a lightweight random-forest estimator that corrects a Dijkstra-based baseline using sparse open data and engineered features for traffic controls and turning movements, formalized as $t_{actual}=f(t_{naive}, \boldsymbol{X}_{controls}, \boldsymbol{X}_{turns}, \epsilon)$. Trained on limited high-quality references (Google Maps BEST_GUESS at 3:00 AM) in Los Angeles, the approach achieves MAPE around 8.4% and $R^2$ about 0.93, substantially reducing bias compared with the naive baseline. The method offers a practical, scalable middle-ground for metropolitan travel-time estimation when proprietary data or heavy computation is unavailable, with applications in planning, accessibility analysis, and network performance under minimally congested conditions.

Abstract

Accurate roadway travel-time prediction is foundational to transportation systems analysis, yet widespread reliance on either data-intensive congestion models or overly naïve heuristics limits scalability and practical adoption in engineering workflows. This paper develops a lightweight estimator for minimally-congested car travel times that integrates open road-network data, speed constraints, and sparse control/turn features within a random forest framework to correct bias from shortest-path traversal-time baselines. Using an urban testbed, the pipeline: (i) constructs drivable networks from volunteered geographic data; (ii) solves Dijkstra routes minimizing edge traversal time; (iii) derives sparse operational features (signals, stops, crossings, yield, roundabouts; left/right/slight/U-turn counts); and (iv) trains a regression ensemble on limited high-quality reference times to generalize predictions beyond the training set. Out-of-sample evaluation demonstrates marked improvements over traversal-time baselines across mean absolute error, mean absolute percentage error, mean squared error, relative bias, and explained variance, with no significant mean bias under minimally congested conditions and consistent k-fold stability indicating negligible overfitting. The resulting approach offers a practical middle ground for transportation engineering: it preserves point-to-point fidelity at metropolitan scale, reduces resource requirements, and supplies defensible performance estimates where congestion feeds are inaccessible or cost-prohibitive, supporting planning, accessibility, and network performance applications under low-traffic operating regimes.

Figures (2)

• Figure 1: Workflow integrating open network data, feature extraction, and reference times for random forest training.
• Figure 2: Comparison of error metrics between naïve baseline and random forest model, showing substantial improvement.