A plant manager in Gainesville called us two springs ago, midway through a conversation about adding twelve workers to a packaging line. The budget made sense, the timeline worked, and the shift details were already sorted. Then he stopped: "What's your TRIR?" We had the number ready. He said he needed the last three years, in writing, before his procurement team would sign off on the contract.
It's a question we get more often now than we did five years ago. Most clients want it before the agreement goes to legal. Some want it before they'll schedule a site visit.
TRIR (Total Recordable Incident Rate) measures how many OSHA-recordable injuries occur per 100 full-time equivalent workers. The formula: (recordable incidents × 200,000) ÷ total hours worked. DART (Days Away, Restricted, or Transferred) uses the same denominator but counts only cases serious enough to affect what an injured worker could do. The 2024 BLS private industry average was 2.3 for TRIR; warehousing and transportation ran notably above that. Most clients in light industrial settings set prequalification thresholds between 1.5 and 3.0.
The TRIR Calculation: Formula and What Counts
The formula OSHA uses:
TRIR = (Number of recordable incidents × 200,000) ÷ Total hours worked
The 200,000 figure normalizes results to 100 workers putting in 2,000 hours a year. A 15-person shop and a 500-person distribution center can be compared on even footing without adjusting for headcount separately.
A recordable incident is any work-related injury or illness that results in one of the following: medical treatment beyond first aid, days away from work, restricted work activity, job transfer, loss of consciousness, or a significant diagnosis from a healthcare professional. A first-aid-only case does not count and should not be on your OSHA 300 log, even if the worker went to an urgent care clinic for assessment.
Here's what the calculation looks like on an actual Georgia facility. A light industrial warehouse has 50 full-time workers and 10 part-time workers each averaging 24 hours a week. Over 52 weeks, total hours worked are (50 × 2,000) + (10 × 24 × 52) = 112,480 hours. Three recordable incidents that year gives a TRIR of (3 × 200,000) ÷ 112,480 = 5.3. That's well above average.
Only count hours actually worked. Paid time off, sick leave, FMLA, and bereavement do not go in the denominator. This is where a lot of calculations come in understated, which becomes a problem when a client or their vendor management platform audits the methodology behind your submitted number.
What DART Measures and How It Differs
DART runs the same math: (DART cases × 200,000) ÷ total hours worked. The difference is in the numerator.
A DART case must involve at least one of three outcomes: days away from work, restricted work activity, or a job transfer because of the injury or illness. A sprained wrist resulting in a two-week light-duty restriction is a DART case. A laceration requiring stitches and a prescription, but no work restriction or lost time, is a recordable for TRIR but not a DART case.
Clients compare TRIR to DART to get a read on severity. A TRIR of 3.0 with a DART of 0.9 says: there are a moderate number of minor recordables, but they're not keeping people off full duty. A TRIR of 3.0 with a DART of 2.6 tells a different story. The second pattern concerns a prequalification reviewer more than the first, even though the TRIR is the same.
We track the DART-to-TRIR ratio for accounts we manage over time, and we've found that a slow drift upward in that ratio, even when the TRIR holds flat, often predicts a more serious incident before it shows up in an annual count. It's not a metric we advertised to clients for a long time because it felt like noise. Turns out it's one of the cleaner early signals we have.
2024 BLS Benchmarks for Warehousing and Light Industrial
The Bureau of Labor Statistics January 2026 release (USDL-26-0101) put the total recordable case rate for private industry at 2.3 per 100 FTE workers in 2024. That's the lowest since the data series began. Manufacturing came in at approximately 2.8. Transportation and warehousing, which covers most light industrial and 3PL operations, ran considerably above the private sector average, consistent with the physical demands of the work.
The national DART rate for private industry came in around 1.4 in 2024. For warehousing and transportation, it tracks higher.
| Sector | Approximate TRIR (2024) | Source | |---|---|---| | All private industry | 2.3 | BLS USDL-26-0101, Jan 2026 | | Manufacturing | ~2.8 | BLS Table 1, 2024 | | Transportation and warehousing | ~4.5 | BLS Table 1, 2024 | | Professional services | ~0.7 | BLS Table 1, 2024 | | Private industry DART | ~1.4 | BLS USDL-26-0101, Jan 2026 |
Look up your specific NAICS code in the BLS Table 1 data for the precise figure. Industry averages vary at the four-digit NAICS level in ways that the broad sector numbers don't show.
For a warehouse or light industrial account, a TRIR below 2.0 puts you in strong standing with most clients. A rate between 2.0 and 3.5 is common territory and typically passes on most prequalification reviews. Above 3.5 and you'll start hitting threshold screens, particularly with clients who use formal vendor management platforms.
We track our own TRIR across Georgia placements each year. On years when we've grown fast and pushed newer placements into physically demanding accounts, the number moves. That's a real pattern, not an excuse, and we explain it directly when clients ask for historical data.
Why Clients Check TRIR Before Signing Anything
That Gainesville plant manager wasn't being difficult. His company used ISNetworld for vendor management, and his procurement team ran every supplier through a scoring process that required three years of TRIR and DART history, OSHA 300A summaries, and an EMR letter from their workers' comp carrier. Vendors above the client's threshold got flagged. Some were declined outright. Existing relationships didn't override the platform grade.
ISNetworld, Avetta, and Veriforce operate as the practical gatekeepers for much of the light industrial supply chain. They compile your safety metrics, score your safety programs, and generate a grade your potential clients can pull. If your TRIR is above a client's threshold, you're not on their approved vendor list. That's the whole process.
The commercial reason clients care: when a worker on their floor gets hurt and they're directing that worker's day-to-day tasks, the injury goes into their OSHA 300 log. Their TRIR moves. Their workers' comp experience modifier moves. A staffing vendor with a poor safety history becomes a risk that lands, at least in part, on the client's incident rate and insurance cost, not just yours.
We've seen industrial food processing and distribution clients in North Georgia with hard caps at 1.5. Manufacturing clients with active OSHA compliance programs set thresholds near 1.0 for high-hazard roles. The range 1.5 to 3.0 covers most of the light industrial vendor reviews we participate in, with the lower end applying to clients who run formal safety programs with worker safety incident tracking as part of their operations KPI stack.
Where Temporary Workers Fit Into Your Rate
OSHA's multi-employer policy is clear: if you control the work being performed, you record the injury. The staffing agency records it too. Both logs carry the same case.
That means a single serious injury involving a temp worker can move a small operation's TRIR considerably. A 30-person facility with 10 temp workers, all supervised by the client's floor leads, running 60,000 total hours a year needs only one DART-qualifying injury to post a DART rate of 3.3, which sits above the 2024 private industry average of 1.4 and above the manufacturing average. One incident. On a temp worker the agency employed but the client supervised.
The practical implication: operations managers using significant temp labor need to include all supervised hours in their TRIR denominator, not just the hours of direct employees. If you're not doing this, your rate is understated, and a vendor management platform reviewing your submitted OSHA hours alongside your incident count may flag the discrepancy before you do.
The staffing agency carries its own workers' comp policy, which protects the client from a direct premium impact on those placements. But the OSHA recordable stays on both logs. How that co-employment structure divides the financial liability between agency and client is covered in our post on workers' comp and staffing liability in Georgia, which walks through the insurance structure and experience rating implications specifically.
Calculating Your Own Numbers This Week
What you need: your OSHA 300 log for the last full calendar year and total hours worked for the same period from payroll.
Three steps: count your recordable incidents from the 300 log; sum all hours actually worked (excluding PTO, sick time, and leave); divide: (incidents × 200,000) ÷ hours worked. For DART, repeat using only the cases on your log that involved days away from work, restricted duty, or job transfer.
If you're operating with fewer than 50 workers, treat your annual rate as a directional signal rather than a benchmark comparison. A single serious injury at a 20-person operation can push a year's TRIR above 10.0. Client safety reviewers who do this work regularly understand the small-sample problem. If your rate is elevated on a small sample, that context is worth addressing directly in any prequalification submission rather than letting the number stand alone.
For operations working to reduce the recordable count itself, our post on reducing new-hire safety incidents in light industrial settings covers the onboarding and supervision factors driving most first-60-day recordables. That's where most operations see the spike.
If you're heading into a prequalification review and want our TRIR numbers on Georgia accounts before the conversation starts, we'll put them in writing. We staff warehouse, packaging, recycling, and light industrial operations across Gainesville, Lawrenceville, Smyrna, Conyers, and the Atlanta MSA. Get Started and we'll put the numbers on the table.
