Calculator Library / Safety
Enter hours worked and case counts to get the standard safety rates with exact intervals, compare them with a prior period or a benchmark, and see a monthly control chart. Built for the small counts safety data really has.
Tool
Rate = cases × 200,000 ÷ hours worked
Enter the hours worked and the case counts for a period. The calculator gives the standard rates with 95% intervals that show how much a small count can move by chance. Add a prior period or a benchmark to compare, and monthly figures for a control chart. Load the example to see a worked year.
Everything runs in your browser. Nothing is sent to a server, and entries are not saved between visits. Definitions of a recordable case, a DART case, and a lost-time case depend on your jurisdiction’s recordkeeping rules. Count cases the way your rules require. The example is made-up data.
Rates
Picture
Enter monthly hours and cases to draw the chart.
Instructions
Enter the hours worked and the counts of recordable, DART, lost-time, and first aid cases and near misses. The calculator gives the total recordable incident rate (TRIR), the DART rate, the lost-time case rate, the lost-time injury frequency rate (LTIFR), the severity rate, and the near-miss ratio. Each rate comes with an exact 95% interval, because safety counts are small and chance alone moves them a lot.
It can also compare your rate with a prior period or a benchmark using an exact test, and draw a control chart of monthly rates. It matches the worked example in the Safety KPIs guide.
| Output | Calculation | What it tells you |
|---|---|---|
| TRIR | Recordable cases × 200,000 ÷ hours worked | Recordable cases per 100 full-time employees in a year |
| DART rate | DART cases × 200,000 ÷ hours | How often cases lead to days away, restriction, or transfer |
| LTIFR | Lost-time cases × 1,000,000 ÷ hours | Lost-time cases per million hours, common outside the US |
| Severity rate | Days lost × 200,000 ÷ hours | How severe cases are, not just how often they occur |
| 95% interval | Exact Poisson interval for the count, scaled to the rate | How far the rate could be from its long-run value by chance |
| Comparison with a prior period | Exact conditional test: given the total cases, the share in the current period is compared with the share of hours | Whether a change is more than chance |
| Comparison with a benchmark | Exact Poisson test of the observed count against the expected count at the benchmark rate | Whether you differ from the benchmark |
| Control chart (u chart) | Rate for each month with limits at the average plus or minus three standard errors, from the Poisson model | Whether a month is unusual |
The pre-loaded data are a year at an illustrative 140-employee plant: 262,000 hours worked and 6 recordable cases. The TRIR is 4.58, with a 95% interval of 1.68 to 9.97. The year before had 9 cases in 255,000 hours, a TRIR of 7.06, so the rate seems to have fallen by about 35%. The exact test gives p = 0.45: with counts this small, a change of that size is well within chance.
Against an illustrative benchmark of 3.5 the plant would expect about 4.6 cases and had 6, with p = 0.48: the plant cannot be told apart from the benchmark. The monthly chart shows no month above the upper limit, so the year looks stable. The lesson is to pair these lagging rates with leading indicators, which move before cases occur and have far more data.
TRIR counts all recordable cases per 200,000 hours worked. DART counts only the cases that led to days away from work, restricted work, or job transfer. LTIFR counts lost-time cases per 1,000,000 hours and is common outside the United States. They answer different questions, so report them together.
Because injuries are rare events and the counts are small. With 6 cases, the 95% interval for the TRIR runs from about 1.7 to 10.0. A change from 7 to 4.6 is well inside that uncertainty. The interval keeps you from over-reading a single period.
Not necessarily. Use the exact comparison: a fall that could easily happen by chance is not evidence of a change. Also check that reporting has not fallen. Leading indicators, which have larger counts, show whether the safety system is actually improving.
It shows whether a month is unusual compared with the average, using limits that allow for the number of hours in each month. A point above the upper limit is a signal to look for a special cause. With low counts the lower limit is zero, so the chart can warn of a bad month but cannot confirm a good one.
Use the one your benchmark and your regulator use. The 200,000-hour base is the US convention for TRIR and DART. The 1,000,000-hour base is common for LTIFR elsewhere. The calculator shows TRIR and DART per 200,000 hours and LTIFR per 1,000,000.
No. The calculator runs entirely in your browser, nothing is sent to a server, and your entries are not saved when you leave the page.