Page 260 - Pipeline Risk Management Manual Ideas, Techniques, and Resources

Incorrect operations index 11/237
customer disruption potential, much less the wide variety of potentially damaging events of soil shrink-swell, subsidence,
materials and joint types commonly seen in distribution sys- erosion, landslide, scour, and others as described in Chapter 5.
tems. It has even been reported that certain physical inspec- Differences in pipe material properties will complicate the
tions may actually increase leak rates in older, low-pressure modeling of distribution system pipeline vulnerability to land
pipelines-the act of temporarily removing earthen cover and movements. Larger diameter pipelines made from more flexi-
side support can actually increase leak rates in certain situa- ble materials and joining processes that create a more continu-
tions [40]. As already noted, distribution system leakage is ous structure, such as a welded steel pipeline, have historically
normally more tolerable with some amount of leakage accept- performed better in seismic events.
able even for some newly installed systems. Leaks often In colder regions, failure considerations exist that are not
replace inspection as the early warning system for distribution present in more temperate climates. These are related to soil
pipelines. movements from frost action and subsurface temperature
It is normally conservatively assumed that some deteriora- changes. Seasonal changes in moisture content and tempera-
tion mechanisms are active in any pipeline (even though this is ture effects have been correlated with both water and gas
certainly not the case in many systems). As time passes, these distribution system break rates in many studies. These are
mechanisms have an opportunity to reduce the pipe integrity. A often shown to be at least partially related to soil movements
good risk assessment model will show this possibility as and resulting changes in stresses on the buried pipe. Where
increased failure probability over time. An assumed deteriora- such correlations are established, they can be used in risk
tion rate is confirmed by inspection in hydrocarbon transmis- assessment and break forecasting efforts as well as in com-
sion pipelines and often by the presence of leaks in other parative risk assessments between regions with differing
systems. An effective inspection has the effect of “resetting the climates.
clock” in terms of pipeline integrity since it can show that loss Score this variable as described on pages 105-1 10.
of integrity has indeed not occurred (or deficiencies can be
cured when detected) and that it is appropriate to assume a cer-
tain level of system integrity or strength. IX. Incorrect operations index
Careful monitoring of leaks also confirms assumed deterio-
ration in the case of some distribution systems. Integrity is As noted in Chapter 6, human error potential is perhaps the
often not thought to be compromised unless or until leaks are most difficult failure mode to assess. An important point in
seen to be increasing over time in such systems. Only an unac- assessing this is the supposition that small errors at any point in
ceptably high andor increasing leak rate, above permissible a process can leave the system vulnerable to failure at a later
original installation leak rates, would be an indication of loss of stage. With this in mind, the evaluator must assess the potential
integrity. So, leak detection surveys can be credited as a type of for human error in each of four phases in pipelining: design,
integrity verification when results are intelligently and appro- construction, operation, and maintenance. A slight design error
priately used to assess integrity. may not show up for years when it is suddenly the contributor to
Although visual inspections with cameras are sometimes a failure. By viewing the entire process as a chain of interlinked
used to inspect pipe interiors, and some tools exist to assess the steps, possible intervention points can be identified. These are
integrity of steel reinforcements of some concrete pipes, the opportunities where checks or inspections or special equipment
use of sophisticated internal inspection devices such as intelli- can be inserted into the process in order to avoid a human-error-
gent pigs is relatively rare in distribution systems. This variable type failure.
will therefore not play a significant risk-mitigating role in most It is a valid observation that human error is also a factor in
cases. If a distribution system does use these devices or other each ofthe other failure mechanisms. Partly as a modeling con-
means for inspecting the pipe wall, the scoring can be kept con- venience, this index is designed to capture all types of human
sistent with the transmission pipeline model Post-installation error potential in a single part ofthe model. This recognizes that
pressure testing can be assessed as an integrity verification the same variables would apply in most other failure modes and
technique as discussed in Chapter 5. The tracking and evalua- it makes sense to evaluate such variables in a single place in the
tion of leak rates can also be assessed as part of this variable model. This approach has the added benefit of facilitating more
scoring. Opportunities for direct assessment of excavated pipe efficient mitigation since human error issues can be more read-
can provide indications of current integrity and can be used ily assessed and addressed in a whole scale fashion. So, in this
with zones of influence (see Chapter 8) or statistical sampling index, variables that are thought to increase or decrease the
thinking to credit these efforts as integrity verifications. Formal potential for human-error precipitated failures are examined
assessments of coating or pipe condition should be minimum (Table 11.6).
requirements for awarding of point credits when scoring these
activities. The evaluator may also choose to include the inspec- Table 11.6 Incorrect operations index possible variables and
tion information from other variables such as leak surveys, cor- weights
rosion control surveys, and effectiveness of coating and
cathodicprotection systems. Variable Wright
Design 30
Land movements Construction 20
Operations 35
The risk variable of land movements assesses the potential for Maintenance 15
damaging geotechnical events. This includes seismic events Incorrect operations index total 100
such as fault movements and soil liquefaction in addition to