Page 400 - Pipeline Risk Management Manual Ideas, Techniques, and Resources

Appendix F1375

Appendix F
Receptor Risk

Eva1 uat ion

The following is extracted from Attachment C ofAppendix 9B associated with 2,395 reportable spills during this 25-year
of Ref. [86]. This reference is an environmental assessment period. The fatality rate is calculated as:
(EA) of a proposed 700 mile gasoline pipeline across the state
of Texas. This extract illustrates the use of specific conse- Fatality rate = 11 fatalitied2395 reportable spills
quence determinations in a QRA. The references to Cases 1 = 0.00459 fatalities per reportable spill
through 4 in the text below refer to four different pipeline fail-
ure probabilities that were used in this EA [86]. These failure The injury rate was calculated in a similar manner. The fatality
probability cases are described in Chapter 14’s Case Study D and injury rates were 0.00459 and0.0238 per spill, respectively.
(we apologize for the confusing use of ‘cases’ and ‘Case This approach assumes that there is no more than one fatal-
Studies’). That Case Study also originates from Ref [86] and ityiinjury per reportable spill even though this is not the case.
shows how the assumptions and calculations shown below were This assumption introduces conservatism into the fatality/
used to produce probabilities of specific damage states. Tables injury rate estimates since the “fatalitieshnjuries per reportable
referenced below can be found in Case Study D and references spill” rates overstate the rate which is really sought: the fre-
noted can be found in Ref [86]. quency of “one-or-more fatalitieshnjuries per reportable spill.”
This extract is included in this book because it details the These two rates are referred to interchangeably in EA discus-
considerations that went into assessing various receptor vulner- sions, but are always based on the conservative calculations
abilities. These can generate ideas and concepts useful to the described here.
designer of any risk assessment methodology. The overall risks of fatalities and injuries from pipeline spills
were determined from the overall leak rate expressed as leaks
per mile per year. For example, the estimated average number
Details of assumptions and calculations of LPP pipeline leaks predicted over the next 50 years, using
used in QRA for environmental industry average reportable leak rates as a basis, is 35. The
assessment of proposed LPP pipeline equivalent number of fatalities that can be expected for this
Fatalities and injuries same length of pipeline over 50 years is 0.00459 (fatalities per
spill) x 35 (spills per 50 years) = 0.16. The annual frequency is
The fatality and injury rates for the 4 cases described inTables 3 calculated as the project-life frequency divided by the project
and 4 of this appendix were calculated from data in the DOT life of 50 years.
Database (DOT, 1999). The fatality and injury rates for the The fatality and injury rates for Case 2 were calculated in a
period 1975-1999 were derived from the total number of fatali- similar manner, using the estimated leak count of 26.8 deter-
ties and injuries associated with pipelines carrying refined mined from the pre-mitigation reportable leak rate of 0.00077
products and crude oil during this period. These rates, leaks per mile per year (10 leaks in 450 miles over 29 years).
expressed as fatalitieshnjuries per reportable spill, are calcu- The average leak rate of 0.00199 leaks per mile used in
lated as the total number of fatalities or injuries divided by the Case 3 includes all leaks: those less than 50 barrels in volume
total reportable spills (spill volumes 2 50 barrels, mostly) in the in addition to reportable leaks. In estimating the fatality and
period 1975-1999. There were 11 fatalities and 57 injuries injury rates, it was assumed that there were no injuries or