Tag
Feature Data
BLS fatal injury rates by trade turn a contractor's hours by occupation into a blended rate that shows which crews carry its fatality exposure.
5.3 Electricians · -19.7% on the period
Roofers had the highest fatal injury rate of any construction trade in 2024
Fatal work injuries per 100,000 full-time equivalent workers, 2024, hours-based
Data
The NWS wind chill formula and the ACGIH work/warm-up schedule turn a cold-day forecast into break minutes and crew hours a contractor can budget.
AI
EarthCam's new connectors let ChatGPT, Claude and other AI assistants answer jobsite questions from detections its cameras have already made.
Data
NCCI's experience mod weights claim count above claim size. This example shows ten $10,000 claims pushing a mod above 1.00 where one $100,000 claim does not.
Data
TRIR and DART divide recordable cases by hours worked. This example computes both for a drywall contractor and compares them to BLS 2024 rates.
Sponsored by ResponseVault
Each signed JHA records a task, a crew and a risk score. Counted weekly, JHA coverage and crew participation show safety gaps before an injury.
AI
A Punchcard and Angus Reid poll of 1,100 Canadian workers finds construction trusts AI with drawings and codes but gets the least employer support.
| Date | Trend | Category | Title | Summary |
|---|---|---|---|---|
| Data | How to weigh crew hours by BLS fatal injury rates, from roofers at 48.7 to electricians at 5.3 Feature | BLS fatal injury rates by trade turn a contractor's hours by occupation into a blended rate that shows which crews carry its fatality exposure. | ||
| Data | How to plan winter crew hours from wind chill and the ACGIH warm-up schedule, with a steel erection example | The NWS wind chill formula and the ACGIH work/warm-up schedule turn a cold-day forecast into break minutes and crew hours a contractor can budget. | ||
| AI | EarthCam connects its jobsite camera AI to ChatGPT and Claude and adds seven detection models | EarthCam's new connectors let ChatGPT, Claude and other AI assistants answer jobsite questions from detections its cameras have already made. | ||
| Data | How to compute your experience mod from the NCCI worksheet, with a $100,000 loss example | NCCI's experience mod weights claim count above claim size. This example shows ten $10,000 claims pushing a mod above 1.00 where one $100,000 claim does not. | ||
| Data | How to compute TRIR and DART from your OSHA 300 log, with a drywall contractor example | TRIR and DART divide recordable cases by hours worked. This example computes both for a drywall contractor and compares them to BLS 2024 rates. | ||
| Data | How to track job hazard analyses as a leading safety indicator | Each signed JHA records a task, a crew and a risk score. Counted weekly, JHA coverage and crew participation show safety gaps before an injury. | ||
| AI | Punchcard survey finds 64% of Canadian construction workers trust AI to read drawings and 87% get no formal AI support | A Punchcard and Angus Reid poll of 1,100 Canadian workers finds construction trusts AI with drawings and codes but gets the least employer support. | ||
| AI | UC Irvine's CORNAV robot planner reads CAD floor plans and look-ahead schedules, raising task success from 13.0% to 72.2% | CORNAV steers site robots from the 2D CAD floor plan and the weekly look-ahead schedule, and kept every planned route out of zones the schedule closed. | ||
| AI | Miter raises $40 million as its contractor payroll platform turns crew photos and voice notes into safety reports | Miter runs contractor payroll and certified payroll and turns crew photos and voice notes into safety reports. It raised $40 million on Sept. 30. | ||
| AI | viAct says AI on Alliad's existing site cameras cut recorded PPE violations 41.7% in three months | viAct reports 41.7% fewer recorded PPE violations in three months on an Alliad hospital job, using cameras already on site. | ||
| AI | Researchers release ConSynth-X, 34,199 synthetic jobsite images for testing site cameras in rain, fog and darkness Feature | A Kent State and UNT dataset turns 3,109 jobsite photos into 11 weather and night versions each, and every version keeps the original labels. |