{
  "abstract": "Microscopic risk assessment of highway weaving sections relies largely on single-vehicle kinematic indicators, yet what such indicators actually detect is rarely tested against a measure of vehicle-to-vehicle interaction. This study analyses 12,256 complete vehicle trajectories collected by UAV at seven highway weaving sections, and evaluates every trajectory along two independent dimensions: a kinematics-based composite indicator combining longitudinal acceleration, lateral acceleration and longitudinal jerk with entropy-derived weights, and an interaction-based measure derived from time-to-collision with the leading vehicle. Because only seven sections are available, all section-level results are assessed by exact permutation as well as by parametric tests. Three findings follow. First, cumulative risk, the time integral of instantaneous risk, is mechanically dominated by traversal time (r = 0.989 with duration, r = 0.948 with weaving length at section level) and is unsuitable for comparing sections of differing length; time-averaged risk intensity is used instead. Second, individual behaviour dominates risk variation and both dimensions agree about it: merging and diverging movements carry higher risk and more conflicts than through movements, and heavy vehicles fewer conflicts than passenger cars (1.1% against 2.7% of trajectories below TTC = 3 s). Higher entry speed shortens exposure by 30.9% and lowers cumulative risk by 25.4% while raising risk intensity by 12.7%, so it reduces exposure rather than severity. Third, and centrally, the two dimensions respond differently to section configuration: task complexity is more strongly associated with interaction-based conflict rate than with kinematic risk intensity (permutation p = 0.028), although with seven sections neither association is significant on its own (p = 0.052 and p = 0.915), and the negative association between weaving length and risk intensity is directionally consistent across three methods but does not reach significance under permutation (p = 0.083). Kinematic indicators therefore characterise the demand placed on the individual driver, and no association between them and the interaction burden a section imposes was detected in these data; the two families of measure must be reported together. Risk tiers built on either kinematic feature are nonetheless externally valid, with conflict rate rising monotonically from 1.2% to 8.3% across intensity tiers.",
  "abstract_shown": true,
  "affiliations": [
    "School of Foreign Studies, Yiwu Industrial & Commercial College, Jinhua, China",
    "School of Industrial Affiliate, Xiangyang Auto Vocational Technical College, Xiangyang, China"
  ],
  "article_type": "Research Article",
  "arxiv": null,
  "author_total": 4,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "Zhenghua Huang"
    },
    {
      "affs": [
        1
      ],
      "name": "Zhibing Yuan"
    },
    {
      "name": "Mengyuan Wang"
    },
    {
      "name": "Peng Fu"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/e1ra9q7cpzbgr.bib",
  "cited_by": 0,
  "doi": "10.1371/journal.pone.0361559",
  "fulltext_url": null,
  "id": "e1ra9q7cpzbgr",
  "issns": [
    "1932-6203"
  ],
  "landing_url": "https://doi.org/10.1371/journal.pone.0361559",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "journal",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "plos",
      "/Engineering and technology/Measurement/Time measurement"
    ],
    [
      "plos",
      "/Medicine and health sciences/Epidemiology/Medical risk factors"
    ],
    [
      "plos",
      "/Physical sciences/Mathematics/Arithmetic"
    ],
    [
      "plos",
      "/Physical sciences/Mathematics/Discrete mathematics/Combinatorics/Permutation"
    ],
    [
      "plos",
      "/Physical sciences/Mathematics/Probability theory/Probability distribution/Skewness"
    ],
    [
      "plos",
      "/Physical sciences/Physics/Classical mechanics/Acceleration"
    ],
    [
      "plos",
      "/Physical sciences/Physics/Classical mechanics/Kinematics"
    ],
    [
      "plos",
      "/Physical sciences/Physics/Classical mechanics/Motion"
    ]
  ],
  "openalex": null,
  "pdf_url": null,
  "pmcid": null,
  "pmid": null,
  "published": "2026-11-05",
  "publisher": "Public Library of Science",
  "ris": "https://research.draconitic.com/paper/e1ra9q7cpzbgr.ris",
  "sources": [
    {
      "first_seen": "2026-11-06T12:20:42Z",
      "native_id": "10.1371/journal.pone.0361559",
      "source": "plos",
      "url": "https://doi.org/10.1371/journal.pone.0361559"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "plos:combinatorics",
      "subfield": "combinatorics"
    },
    {
      "field": "mathematics",
      "origin": "plos:probability theory",
      "subfield": "probability"
    },
    {
      "field": "medicine",
      "origin": "plos:epidemiology",
      "subfield": "epidemiology-public-health"
    }
  ],
  "title": "Microscopic risk assessment of highway weaving sections based on large-scale UAV trajectory data: What kinematic and interaction measures each detect",
  "title_text": "Microscopic risk assessment of highway weaving sections based on large-scale UAV trajectory data: What kinematic and interaction measures each detect",
  "type": "peer-reviewed",
  "updated_at": "2026-11-07T12:20:24Z",
  "url": "https://research.draconitic.com/paper/e1ra9q7cpzbgr",
  "venue": "PLOS One",
  "versions": []
}