Comparisons of Mandatory and Discretionary Lane Changing Behavior on Freeways

Vechione, Matthew; Balal, Esmaeil; Cheu, Ruey Long · 2018 · ROSA P / Tongji University

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This study addresses the lack of quantitative evidence comparing driver behavior during mandatory lane changes (MLCs) and discretionary lane changes (DLCs) on freeways. While existing literature suggests that MLCs and DLCs are driven by different motivations—such as exiting versus gaining speed—few studies have statistically compared the risk-taking behaviors associated with each. The research aims to determine whether distinct statistical models are necessary for each lane change type, which is critical for improving microscopic traffic simulations and autonomous vehicle algorithms. The authors analyzed vehicle trajectory data from the Next Generation Simulation (NGSIM) database, specifically from two sites in California: Interstate 80 in Emeryville (Dataset A) and U.S. Highway 101 in Los Angeles (Dataset B). They identified passenger car lane changes, classifying moves involving auxiliary lanes as MLCs and moves between main lanes as DLCs. The study examined four decision variables representing gaps between the subject vehicle and surrounding traffic: the front gap in the original lane before the change (GPB), the front gap in the target lane after the change (GPA), the rear gap in the target lane after the change (GFA), and the total gap in the target lane (D). The researchers conducted descriptive statistical comparisons, hypothesis tests on mean differences, and Kolmogorov–Smirnov tests to compare both observed and fitted cumulative probability distributions for each variable. The results indicate that for the three variables associated with the target lane (GPA, GFA, and D), there were no significant differences in means or distributions between MLCs and DLCs. These variables were best described by log-normal distributions with similar parameters for both lane change types. However, the front gap in the original lane (GPB) showed significant statistical differences in both means and distributions between MLCs and DLCs across both datasets. In Dataset B, the mean GPB for MLCs was substantially larger than for DLCs. The authors suggest this discrepancy arises because GPB is a critical input for DLC decisions but is likely not a primary factor in MLC decisions, where drivers are focused on reaching the target lane regardless of the gap behind them in the original lane. The study concludes that MLCs and DLCs should be modeled separately in traffic simulation tools and autonomous driving systems. While both models can share the same decision variables for the target lane (GPA, GFA, and D), they require different inputs and logic regarding the original lane gap (GPB). This finding provides statistical justification for developing distinct behavioral models, which can enhance the accuracy of microscopic simulations and improve the safety and naturalness of connected and autonomous vehicles by better replicating human driver risk-taking behaviors.

Key finding

The gap between the subject vehicle and the preceding vehicle in the original lane is the only decision variable showing statistically significant differences in means and distributions between mandatory and discretionary lane changes.

Methodology

dataset

Sample size: 500

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• lane changing
• gap acceptance

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: behavioral performance data
• Theoretical Contribution: computational model