01 Mar Construction industry driving risk 

Lytx analysed and labelled more than 50 million risky driving events. By combining machine vision and artificial intelligence with professional review, we’re able to present a customised and accurate view of risk to our clients; with greater than 95% accuracy across more than 60 risky driving behaviours. 

Construction industry driving risk

Lytx’s construction fleet telematics findings include the five risky behaviours seen most often among construction drivers, the most improved driving behaviours, and insights on how construction-industry driving habits compare to those in other industries.

This data was captured from construction fleet tracking of all sizes and types of fleets within the industry, including commercial and residential construction, contractors, builders, excavation, lumber, roofing fleets, and more.

How construction fleets stack up against other industries

Lytx compared the prevalence of behaviours seen in construction fleets against behaviour averages of fleets across all its other protected industries. Comparatively, construction fleets stood out in the following areas:

• Posted speed violation, which occurred 16% less often
• Smoking, which occurred 69% more often
• Other distraction, which occurred 32% more often

Construction collision insight

Lytx also found that 91% of collisions in the construction segment were low impact. Late response, one of the most prevalent behaviours in this space, contributed to 32% of low-impact rear-end collisions.

About the data

These insights were derived from Lytx’s proprietary database of construction fleet management data, including 192,000 risky construction driving events captured last year. For comparisons across industries, Lytx calculated behaviour averages from its global database, which contains driving data from trucking, distribution, concrete, utilities, services, transit, government, and waste industries. 

Lytx maintains the fastest-growing proprietary database of professional driving data in the world, currently surpassing 200 billion kilometres of driving data. The data is anonymised, normalised, and in instances of behaviour prevalence, generalisable to construction fleets at large.