An Introduction to Chemical Process Safety

This chapter is from the book

Chemical Process Safety: Fundamentals with Applications, 4th Edition

Learn More Buy

This chapter is from the book

This chapter is from the book 

Chemical Process Safety: Fundamentals with Applications, 4th Edition

Learn More Buy

1-7 Accident and Loss Statistics

Accident statistics are one metric to determine the effectiveness of any safety program. However, accident statistics are lagging indicators and are usually more indicative of personal safety rather than process safety.

Several methods may be used to calculate accident and loss statistics. All of these methods must be used carefully, because each method has strengths and weaknesses and no single method is capable of measuring all of the required aspects. The methods most commonly used to measure accident statistics are as follows:

• Total number of fatalities or injuries/illnesses.

• Fatality rate, or deaths per person per year

• Fatal injury rate based on total hours or total workers

• Incidence rate

All of these measures are lagging indicators, since they are tabulated after an accident has occurred.

The U.S. Occupational Safety and Health Administration (OSHA; www.osha.gov) has legal authority over U.S. workplace safety. OSHA is responsible for ensuring that U.S. workers are provided with a safe working environment. Many countries have government organizations similar to OSHA.

All U.S. workplaces are required by law to report to OSHA all occupational deaths, illnesses, and injuries. An injury includes medical treatment (other than first aid), loss of consciousness, restriction of work or motion, or injuries causing a transfer to another job. These accident statistics are tabulated by the U.S. Bureau of Labor Statistics (BLS; www.bsl.gov) and are made available to the public—albeit usually more than a year after the calendar year the data were collected. Table 1-6 provides sources for accident statistics; please refer to these sources for more updated statistics than presented here.

Table 1-6 Sources of Accident Statistics

The total number of fatalities is most commonly used as a lagging indicator, but does not take into account the number of people working in a particular occupation. For instance, many more auto-related fatalities occur in a big state like Texas than in a small state like Vermont.

The total number of injuries/illnesses is also dependent on the number of workers. However, it has an additional problem since it requires a definition of an injury or illness.

The fatality rate, or deaths per person per year, is independent of the number of hours exposed to the hazard and reports only the fatalities expected per person per year. The exposed population may be carefully defined to ensure that it includes only those exposed to the hazard. This approach is useful for performing calculations on the general population. Fatality rate is calculated as follows:

Click to view larger image

The fatal injury rate is defined in two different ways. The first approach is in terms of the number of fatalities per 100,000 full-time equivalent workers employed. Thus, the worker-based fatal injury rate is calculated using the following equation:

Click to view larger image

A similar approach can be applied to a general population. This fatal injury rate is defined in terms of 100,000 people and applied to a general, exposed population. It is calculated using the following equation:

Click to view larger image

A work-related fatal injury rate can be defined in terms of the total hours worked by 100,000 full-time equivalent workers. For 100,000 workers working 40 hours per week and 50 weeks per year, this results in (100,000 workers × 40 hours/week × 50 weeks/year) = 200,000,000 hours. Thus, the hours-based fatal injury rate is defined by the following equation:

Click to view larger image

Hours-based fatal injury rates (Equation 1-4) are generally considered more applicable than worker-based fatal injury rates (Equation 1-2). Hours-based rates use the total number of employees at work and the total hours each employee works. Worker-based rates will be similar for groups of workers who tend to work full time, but differences will be observed for worker groups who tend to include a high percentage of part-time workers.

The incidence rate is based on the cases per 100 workers. A worker year is assumed to contain 2000 hours (50 work weeks/year × 40 hours/week). The incidence rate, therefore, is based on 200,000 hours of worker exposure to a hazard (100 worker years × 2000 hours/year). The incidence rate is calculated from the number of incidents and the total number of hours worked during the applicable period. The following equation is used to calculate the incidence rate:

Click to view larger image

The incidence rate is typically used for accidents involving injuries or illnesses, although it was used for fatalities in the past. The hours-based fatal injury rate is commonly used for fatalities, whereas the incidence rate is used for injuries since fatalities occur much less frequently than injuries. Using a different number of hours for these two rates brings both rates within comparable numerical values.

OSHA also uses the incidence rate for illnesses; days away from work (DAW); and days away from work, job restriction, or job transfer (DART). Table 1-7 defines these terms in relation to occupational injuries. There are many other ways to present accident statistics depend-ing on what you wish to achieve. For instance, for airline transportation, the usual method is to report fatalities per million miles traveled.

Table 1-7 U. S. OSHA Definitions for Occupational Injuries

Source: www.osha.gov.

Table 1-8 provides OSHA statistics on the total number of fatalities, the hours-based fatal injury rates, and the total recordable incidence rates for the United States in 2015, ordered from the highest number of fatalities to lowest. In 2015, a total of 4836 occupational fatalities occurred. The peak number of fatalities was 5840 deaths recorded in 2006; the low was 4551 fatalities in 2009, primarily due to the recession of 2008—fewer workers means fewer fatalities. The total number of fatalities has been increasing slowly over the past few years (4821 in 2014) due to an increase in the number of workers, but likely at a diminished pace owing to improvements in occupational safety programs.

Table 1-8 2015 U.S. Occupational Statistics for Selected Industries, Ranked from Highest to Lowest Number of Fatalities

Source: U.S. Bureau of Labor Statistics, www.bls.gov/iif/.

Several conclusions can be reached from Table 1-8. Construction (overall) has the highest number of fatalities (937), but fishing, hunting, and trapping has the highest hours-based fatal injury rate (54.8). The difference depends on the number of workers employed in each area. Construction has a larger number of workers than fishing, hunting, and trapping, resulting in the total fatalities for construction being higher and the hours-based fatal injury rate being lower. Interestingly, hospitals have the second highest total recordable incidence rate (8.1), followed by agriculture, forestry, fishing, and hunting (5.7). Table 1-8 also shows that the traditional chemical engineering industries are near the bottom in terms of occu-pational injuries and fatalities. This group includes chemical manufacturing (28 fatalities), plastics and rubber products manufacturing (17 fatalities), oil and gas extraction (6 fatalities), and chemical and allied products merchant wholesalers (3 fatalities). The hours-based fatal injury rates and total recordable incidence rates for these industries are lower than those of many other occupational activities that are commonly considered as safer. For example, colleges, universities, and professional schools had a total of 17 fatalities in 2015. Many specific chemical companies achieve total recordable incidence rates as low as 0.2, compared to the industry average for chemical manufacturing of 2.1.

Table 1-9 provides details on the nature of the fatalities. Clearly, transportation accidents account for the largest number of fatalities in the workplace (2054 fatalities). This is followed by falls, slips, and trips (800 fatalities). With respect to the nature of the fatal injury, most of the injuries are due to multiple traumatic injuries and disorders—occupational fatalities usually involve widespread injury to many areas in the human body. With respect to the worker activity involved with the fatality, transportation accounts for the largest number of fatalities, followed by constructing, repairing, and cleaning and using or operating tools or machinery.

Table 1-9 Details on the Nature of Occupational Fatalities in 2015

Source: U.S. Bureau of Labor Statistics, www.bls.gov/iif/.

Note under the “Primary Source” heading in Table 1-9 that 233 fatalities occurred due to exposure to chemicals and chemical products. However, if you look further into the U.S. Bureau of Labor Statistics data, you find that only 5 of these deaths occurred in the chemical manufacturing industry and only 1 in operations of chemical and allied products merchant wholesalers. One can easily conclude that few fatalities in the chemical industry are due to chemical exposures; instead, most of the chemical fatalities occur in industries that are not considered chemical in nature.

Table 1-10 provides more details on fatalities in the chemical industry. Surprisingly, retail gasoline stations account for the largest number of fatalities (39 fatalities—due mostly to robberies). Within chemical manufacturing, fertilizer manufacturing (6 fatalities) and basic chemical manufacturing (5 fatalities) account for the largest number of fatalities. Petroleum refineries had 4 fatalities in 2015, while crude petroleum and natural gas extraction had 6 fatalities.

Table 1-10 2015 Fatal Occupational Injuries Related to the U.S. Chemical Industry

Source: U.S. Bureau of Labor Statistics, www.bls.gov/iif/.

The Marsh and McLennan companies annually publish a report entitled 100 Largest Losses in the Hydrocarbon Industry.⁵ The most recent report tabulates losses from 1974 to 2015 and is based only on the property value losses from the ground up. It does not include the financial losses due to fatalities/injuries, environmental factors, lawsuits, fines, or business interruption—these additional losses could easily multiply the losses by many times. Table 1-11 shows the percentage of losses by industry sector and Table 1-12 shows the total property damage losses by event type. Reviewing these data, the first conclusion is that these losses are huge—totaling more than $33 billion, an amount that does not include losses to human life and environmental losses. Second, 87% of the losses occurred in upstream oil and gas production, refining, and petrochemicals. Finally, $25 billion in losses—75% of the total dollar losses—are from explosions and fires.

Table 1-11 Percentage of Property Damage by Industry Sector

Source: The 100 Largest Losses 1974–2015, 24th ed. (New York, NY: Marsh and McLennan Companies, March 2016), p. 10.

Table 1-12 Property Damage Values Based on Event Type

Source: The 100 Largest Losses 1974–2015, 24th ed. (New York, NY: Marsh and McLennan Companies, March 2016), p. 10.

Table 1-13 is a list of non-occupational fatalities in the United States for the year 2014 ranked from the highest number of fatalities to lowest. Also shown is the deaths per 100,000 people, as defined by Equation 1-3. In 2014, there were 136,053 non-occupational fatalities due to unintended injuries—compared to 4836 occupational fatalities. Poisoning accounted for the highest number of fatalities, although this includes 38,718 poisoning deaths by drug overdose. This alarmingly large number of fatalities is dramatically increasing each year. Motor vehicle deaths numbered 35,398—a total that has been increasing slowly for the past few years. In 1972, the number of motor vehicle fatalities reached a peak of 56,278. In 2014, 58 people died from electrocution by exposure to electric transmission lines, while 25 died from lightning.

Table 1-13 Non-Occupational Fatalities in the United States Due to Unintentional Injuries, 2014

Source: Injury Facts (Itasca, IL: National Safety Council, Itasca, IL, 2015), www.nsc.org.

Comparing Table 1-13 with Table 1-8 shows that the total number of fatalities in the workplace is much lower than the non-occupational fatalities in the general population: The number of workplace fatalities is comparable to the number of deaths by choking. Choking, falls, motor vehicle deaths, and poisonings all exceed the total number of workplace fatalities by a large margin. Also note that the general population is much larger than the total number of workers in the general population. Nevertheless, this comparison does provide an indication of the magnitude of workplace deaths compared to non-occupational deaths.

In summary, accident statistics show that:

• The chemical industry has much lower fatalities and hours-based fatal injury rates than many other occupational activities that are commonly considered to be safer.

• The numbers of transportation and motor vehicle fatalities are high in both occupational and non-occupational environments.

• Chemical industry incidents, although infrequent, can result in huge property losses.

The chemical industry includes chemical plants, refineries, and other industrial sites using chemicals. Despite the relatively small number of fatalities that occur in the chemical industry, the potential always exists for a major incident—though such an event remains unlikely. Clearly, no unintended injury or fatality is acceptable in the workplace or elsewhere. All safety programs must drive toward zero injuries.