Accident severity rates calculating

Frequency rates tell how many accidents took place per 200,000 hours worked. They do not tell how serious the accidents were. For example, one lost-time accident may involve a day off and another may require 10 days away from work. For this reason, another accident rate was created, called an accident severity rate. To calculate an accident severity rate you need to know the number of work days lost and the number of employee hours worked. The formula for calculating a severity rate is ... 

The second view is macroscopic. In case more than one event is evaluated, an aggregation of the single events is possible in order to assess the overall effects. If the sample under investigation happens to contain accident and non-accident events, an accident rate or prevention rate can be calculated as ratio of frequency of accidents (or one minus accidents) with a measure by frequency of accidents without the measure. Summary statistics can also be computed in non-accident events by statistically evaluating the indicators defined on the physical level. In comparison to a baseline without measure the change due to a specific safety measure can be evaluated at the desired level of detail. Within the accident group, rates for specific injury severities as well as a fatality rate can be estimated. 

Frequency rate = number of accidents x 200,000 divided by the total employee hours worked. North Americans use 200,000 hours as a base, the Europeans tend to use 100,000 hours, and still others use 1,000,000 hours in their rate calculations. The reason 200,000 hours is used in North America is because it roughly equals the number of hours worked by 100 employees during a normal work year. Using 200,000 as a base makes it easy to estimate the site s frequency rate by simply knowing the number of employees at work. For example, if your site has 200 employees and you had six recordable injuries, you have a frequency rate of about 3 if you had 400 employees with six recordable injuries, you have an accident rate of about 1.5. Commonly used frequency and severity rates are ... 

Property damage severity rates can also be calculated by determining the total cost of property damage accidents x 200,000, divided by the total employee hours worked. 

The short-term permeation rate of hydrogen from the ZrHi 7 layer at an accident is also calculated. The hypothetical scenario in which the temperature of the ZrHi 7 layer is suddenly increased to 1,260°C (the criterion of the cladding surface temperature at the LOCA is created, and the void reactivity is analyzed). The results are shown in Fig. 7.11. It can be seen that several hours are needed to lose one third of the hydrogen from the layer. That is much longer than the time required for the control rod insertion and borated water injection. Therefore, the hydrogen loss does not impose any severe problem for reactor safety. 

Before the accident at Three Mile Island Unit 2 (TMI-2) in 1979, protective action decisions would often be based on real-time environmental measurement of dose rates following a release. Once this dose rate was measured, a projected effective dose could be calculated. This effective dose could then be compared to the intervention levels and the appropriate protective action could be selected. There is a serious problem with this approach by definition, environmental measurements are obtained ctfter a release. Thus, they cannot be used to initiate protective actions before the release. Moreover, even if field measurements are taken shortly after release initiation, much time can be consumed in the process of selecting and implementing appropriate protective responses. After gamma dose-rates are assessed, it is necessary to select an action, obtain the concurrence of off-site authorities, and transmit warnings to the population at risk—who must prepare to evacuate and then drive out of the risk area. The result is that for severe releases the protective action may be taken too late to be effective. 

Three typical examples of the results of these calculations which are of interest in reactor accident considerations are shown in Fig. 7.24. The tendency of these results can be summarized as follows At pH 5, 100 °C and an initial I2 concentration of 10 g-atom/1, iodate formation proceeds very slowly, reaching the same concentration as I2 after about 1 day the equilibrium state of the reaction would only be established after about 100 days. This means that over a comparatively long period of time one has to deal with rather high fractions of the molecular species I2 and HOI. Raising pH to about 7 at the same temperatme results in a much faster IO3" formation, with the equilibrium state already being established after about 10 minutes. At lower temperatures, the reaction rates are correspondingly lower. In solutions with very low total iodine concentrations, the I2 fraction decreases very quickly at pH 7 however, HOI disproportionation proceeds rather slowly with the consequence that the equilibrium state of reaction (3) is attained only after several days. 

A particular problem with all of these comparisons is that there is no consistency about what constitutes an accident and it should be remembered that this was one of the problems with any comparison of incidence and frequency rates. One way of improving comparisons is to calculate a rate which takes into accoxmt the severity of the accidents, i.e. the number of days lost per accident, to give the mean duration rate ... 

This is a very basic method for calculating upper and lower control limits. Other methods can also be used to calculate these hmits. To ensure that the statistical control charts are as reliable as practical, keep several guidelines in mind. First, use accident rates that have as many sets of data as possible. For example, an aU injury/iUness frequency rate works better than a lost time frequency rate. Second, try to use at least twenty sets of data in calculating the base rate or average. In our example we only used five data points for practical purposes. However, they represented 60 individual monthly frequency rates. 

Next, we ask whether the LTI-rate is a valid indicator of the risk of losses dne to accidents. This criterion is more problematic, since the LTI-rate is insensitive to the severity of the injuries. An eye injury resulting in a few days of absence and a severe fall injury with many months of sick leave connt equally when calculating the LTI-rate. It is questionable whether the LTI-rate is a valid indicator of the risk of losses due to accidents. Other SHE performance indicators that we will look into in the next Section are better suited as to validity, since they account for the degree of harm (e.g. number of days of absence). 

Truck accidents occur relatively infrequently. A trucking firm with an accident rate three times the industry average has a reportable accident rate of one-and-a-half accidents per million miles. An accident is "reportable if it involves a death, a serious injury, or property damage severe enough that a tow truck has to be called. Seventy percent of trucking firms operate less than 100,000 miles a year, so it is obvious that even very dangerous carriers could be in business for many years without an accident. The FHWA s problem is compounded in that they only know the number of miles operated by forty percent of the carriers. Another forty percent of carriers self-report the number of trucks they own. The FHWA doesn t have any information on the remaining twenty percent of carriers. Consequently, the FHWA cannot calculate accident rates for most carriers. 

---