What makes vehicle-as-a-weapon attacks difficult to prevent?

Ordinary vehicle access can create an approach to a crowded space. A site review should examine approach routes, pedestrian activity, access controls and response arrangements together.

What does hostile vehicle mitigation (HVM) design involve?

HVM design establishes the relevant vehicle scenarios, examines site access and vehicle dynamics, and sets requirements for physical protection and operations. The Crime Science taxonomy informs scenario selection; the site assessment establishes the design basis.

When does a project require hostile vehicle mitigation analysis?

Consider HVM analysis where vehicle access and pedestrian activity create a credible exposure. The scope depends on the site, its use and existing controls. A general incident taxonomy alone cannot establish whether a site needs barriers.

Vehicle as a Weapon

Study at a glance: The Crime Science paper analysed 135 incidents from 2010 to April 2025 and identified six attack clusters. This page summarises that study.

A vehicle security strategy needs a clear account of the scenarios it addresses. The published taxonomy gives owners, designers and security advisers a common starting point for that discussion.

Why we did this research

Our aim was to give project teams a more specific basis for discussing vehicle threats, so the design conversation can address the setting, the people using it and the decisions still open.

What we did

The study used open-source incident records, Factor Analysis for Mixed Data and k-means clustering. The three retained dimensions describe temporal attack complexity, tactical sophistication and operational lethality.

Classifying vehicle-as-a-weapon attacks: an inductive taxonomy and design implications, by Codee Roy Ludbey, Daniel Versaci, Samyak Bharthur and Shreemen Prabhakaran, was published in Crime Science on 9 February 2026.

What we found

The six clusters distinguish heavy-vehicle, light-vehicle, ideological, protest-related and mixed-motivation incidents. They provide a descriptive taxonomy, not a model of attack probability.

The historical sample and uneven English-language reporting limit coverage. Incomplete barrier and street-furniture records limit conclusions about mitigation performance. Cluster frequencies should not be read as the likelihood of an attack at a particular site.

For the full cluster definitions and results, refer to the published paper.

What it means for design and governance

In our practice, the taxonomy informs the scenarios we examine. A project still needs its own assessment of access, vehicle movements, pedestrian use, existing protection and response arrangements before requirements can be set.

The commissioning decision is to agree which scenarios need testing, what evidence is available and who will own the resulting design and operational decisions.

Use the scorecard to prepare for a review

The 6 Profiles HVM Scorecard is a Core42 self-assessment informed by the taxonomy. Its questions, scoring and site recommendations are a separate advisory application; they were not validated by the Crime Science paper. Use the result to prepare for a site-specific review.

Benchmark your site Read the Crime Science paper

Related publication: Vehicle-As-A-Weapon Threats in Cities: A Global Analysis of Attack Patterns, Terrorism and Political Violence (2026). The figures and method summarised above refer to the Crime Science study.

Related research: Security 2030 — how senior security leaders describe their organisations' readiness for the threats ahead. For a live project, see our HVM consulting practice.