Page 296 - Pipeline Risk Management Manual Ideas, Techniques, and Resources

Risk assessment model 13/273
detected in I day, even though the same amount of product is Station stuffing As an opportunity to detect and react to a
spilled in either case. Unknown and complex interactions leak, the staffing level of a facility can be evaluated by the
between small spills, subsurface transport, and groundwater following relationship:
contamination, as well as the increased ground transport oppor-
tunity, account for the increased chronic hazard. Opportunity to detect = [(inspection hours) + (happenstance detection)]
One application of such an amplification factor established
an equivalency by saying that a 200,000-barrel (bbl) contain- where
ment area with very good leak detection capabilities is roughly Inspection hour =an inspection that occurs within each
equivalent to a 500-bbl containment area with very poor leak hour
detection capabilities-from a risk perspective. The larger con- Happenstance detection = 50% ofmanned time per week.
tainment area has a greater potential leak volume due to its
larger stored volume, but either can produce a smaller, but con- In this relationship, it is assumed that station personnel
sequential leak. Making these two scenarios equivalent empha- would have a 50% chance of detecting any size leak while they
sizes the importance of leak detection capabilities and limits were on site. This is of course a simplification since some leaks
the ‘penalty’ associated with higher storage volumes. This would not be detectable and others (larger in size) would be
equivalency seems to be reasonable, although any ratio will suit 100% detectable by sound, sight, or odor. Additional factors
the purposes of a relative assessment. With a desired amplifica- that are ignored in the interest of simplicity include training,
tion factor fixed, various combinations of containment volume thoroughness of inspection, and product characteristics that
and leak detection capabilities can be assessed, used to produce assist in detectability.
spill scores, and then compared on a relative basis. An alternate approach to evaluating the staffing level as it
Improvements to the spill score are made by reducing the relates to detection is to consider the maximum interval in
product containment volume in the case of volume-dependent which the station is unmanned:
spills, and by reducing the source (e.g., pressure, density, head,
hole, time-to-detect) in the case of rate-dependent spills. Note
that improvements in leak detection also effectively reduce the Worst case = maximum interval unobserved
source, in the leak-rate dependent case.
In assessing station leak detection capabilities, all opportuni- Examples of evaluating various staffing protocols using the
ties to detect can be considered. Therefore, leak detection sys- two techniques are shown in Table 13.6. The last column shows
tems that can be evaluated are shown in Table 13.5. The time to the results of a “maximum interval unobserved” calculation
detect various leak volumes (T, through T,,,,, in Table 13.5, while the next to the last column shows the “opportunity to
representing volumes from 1 bbl to 1000 bbl of spilled product detect” calculation.
and defined in Table 7.1 3) can be estimated to produce a leak The maximum unobserved interval method is simple, but it
detection curve similar to Figure 7.7 for each type of leak detec- appears worthwhile to also consider the slightly more compli-
tion as well as for the combined capabilities at the station. The cated “opportunity” method, since the “max interval” method
second column, reaction time, is for an estimate of how long it ignores the benefit of actions taken while a station is manned,
would take to isolate and contain the leak, after detection. This that is, while performing formal inspections of station equip-
recognizes that some leak detection opportunities, such as 24-7 ment-rounds. The “opportunity” method, while providing
staffing of a station, provide for more immediate reactions similar relative scores, also shows benefits that more closely
compared to patrol or off-site SCADA monitoring. This can be agree with the belief that more directed attention during
factored into assessments that place values on various leak episodes of occupancy (performing inspection rounds) are
detection methodologies. valuable.

Table 13.5 Leak detection opportunities

Leukdetection system Reaction time Ti T,, TI 00 ~Iooo
7 x 24 manning with formal. scheduled “rounds”
5 x 8 staffing with formal, scheduled rounds
7 x 24 staffing, no formal rounds
5 x> 8 staffing, no formal rounds
Other staffing combinations
Occasional site visits (weekly)
Mass balance for facility
Mass balance for station
Pressure point analysis
Acoustic monitoring
SCADA real-time monitoring
Groundwater monitoring
Surface drain system (monitored)
Soil vapor monitoring
Passerby reporting