Improving the Safety of Marine Pipelines (1994) / Chapter Skim
Currently Skimming:
Appendix E A Risk Analysis Approach
Appendix E A Risk Analysis Approach
Pages 116-139
The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.
From page 116... ...
It may be sufficient in some cases and can provide a basis for initial sensitivity analyses. As outlined in Chapter 3, this model is based on a zonation of the considered area by superposition of three maps (water depth, vessel traffic density, and pipeline density)
|
From page 117... ...
The mean of the amount spilled per unit of surface and per year in the min-zone and the mean annual property damage to pipelines and vessels. Zonation of the Region The area of interest is divided into min-zones defined by the superposition of three different partitions: .
|
From page 118... ...
What is shown here is a hypothetical region. A harbor site was assumed because the vessel traffic is more likely to be high in its vicinity.
|
From page 119... ...
and, possibly, specific distributions of pipe sizes, ratios of old to new and inactive to live pipelines, as well as the proportion of those carrying gas versus oil or condensate (unless these characteristics are considered geographically uniform) · The density of vessel traffic, indexed by k (1, 2, or 3)
|
From page 120... ...
The risks associated with each failure mode can then be computed and summed over all accident types and all min-zones to assess the overall risk in the Gulf. In particular, one can compute independently the mean annual number of human casualties, and the mean annual volume of oil spilled due to leaks (which depends on the size of a hole in a live oil pipeline of specified diameter for a given spill duration)
|
From page 121... ...
The mean annual volume of oil spilled depends on the mean annual number of impacts of various objects (mostly anchors) on live, exposed pipelines transporting oil and the mean amount of oil spilled per event.2 The annual number of impacts depends on the density of vessel traffic, the density of live pipelines, and the proportion of pipelines that are exposed and transport oil in the different specified ranges of diameters.
|
From page 122... ...
The mean annual volume spilled depends on the mean number of corrosion holes per year in min-zone i (allocated among buried and exposed pipelines) , which itself depends (cumulatively)
|
From page 123... ...
The mean annual amount of oil spilled also depends on the nature of the incident/accident initiator, on the detection time, the diameter of the pipe, the pressures, and the time to response. The property damage (as well as the number of casualties)
|
From page 124... ...
Characteristics of Min-Zone i 1: Index of min-zone = {j, k, 1} (pipeline density, vessel traffic density, water depth indexes for min-zone i) se: Proportion of exposed pipelines in min-zone i hi: Proportion of live pipelines in min-zone i; hi may depend on al gi: Proportion of live pipelines transporting gas or liquefied gas in min-zone i (ltransport oil)
|
From page 125... ...
O: Impact of an anchor or other dropped object on a live exposed pipeline hi: Mean annual number of object impacts on live exposed pipelines per square mile of min-zone i as a function of j, k, l, al and hi Oi: Mean annual number of object impacts on live exposed pipelines transporting oil in min-zone i q: Index of breach diameter ranges in object impact incidents Ohq: Occurrence of a hole of diameter q caused by object impact on a live pipeline pm~ohq10~: Probability of a hole of diameter q (discretized distribution) conditional on impact of object on a live exposed pipeline in the diameter range m tr: Spill duration ranges ~ Example: to: <1 day; t2: from 1 to 5 days; t3: > 5 days)
|
From page 126... ...
in min-zone i CHBi: Mean annual number of corrosion holes in buried live pipelines transporting oil in min-zone i c: Index of corrosion hole diameter ranges when detected and fixed dc: Corrosion hole diameter ranges when detected and fixed NCEiCm: Mean annual number of corrosion holes in exposed oil pipelines of diameter m, that reach diameter c (discretized) in min-zone i NCBiCm: Mean annual number of corrosion holes in buried oil pipelines of diameter m, that reach diameter c in min-zone i p~dClCH)
|
From page 127... ...
: Probability that a storm or hurricane of severity s that affects min-zone i causes nb breaches in marine pipelines in min-zone i (depending on the pipeline density and the water depth) XHi: Mean amount of oil spilled per year in min-zone i due to damage caused by storms and hurricanes to live pipelines transporting oil XH: Mean volume of oil spilled in a breach caused by a storm or a hurricane in a live pipeline transporting oil in min-zone i (assumed here to be averaged over pipe diameters, severity of the damage to the pipe and detection time)
|
From page 128... ...
The mean annual number Ci of collisions in min-zone i between vessels and exposed live pipelines transporting gas depends on the surface of the min-zone, the collision rate per square mile and the proportions of pipelines that are live and transport gas (it is assumed here that these two ratios are independent)
|
From page 129... ...
Eq. 3 The mean annual number of deaths in explosions or sudden fires due to vessel/ pipeline collisions in min-zone i is the product of the average number of explosions by the average number of deaths per explosion (the latter can be either infered from existing statistics, or computed as the result of a model involving different types of vessels)
|
From page 130... ...
and the property damage to live pipelines. The mean annual number of impacts of dropped and/or dragged anchors or other objects on live, exposed pipelines transporting oil in min-zone i is: 0i = Si X °i X (1 -gi)
|
From page 131... ...
10 Similarly, the mean value of the annual property damage due to impacts of dropped and/or dragged anchors or other objects on all live exposed pipelines in min-zone i is the product: PDOi = 0i x PDO Eq. 11 IEc: Corrosion of Pipelines Transporting Oil The simplified model used here relies on the events and random variables described in Figure E-7.
|
From page 132... ...
x dm Eq. 13 The mean annual amount of oil spilled through corrosion holes in exposed live pipelines transporting oil in min-zone i is sum of the amounts of oil spilled for all pipe diameters, sizes of corrosion holes, and spill durations: XCEi = ~ Hi, Hi, tNCEiCm x XCE(6c, tr)
|
From page 133... ...
nb nh s Eq. 20 The mean annual amount of oil spilled due to breaches in pipelines caused by hurricanes and storms in min-zone i is obtained by restricting this number to live pipelines transporting oil, and multiplying the result by the mean amount spilled per event: XHi =NHBi xbi X(l-gi)
|
From page 134... ...
lEe: Accidents and Incidents Involving Pipelines at the Site of a Platform It is assumed here that either a preliminary PRA has been performed, accounting for the specific configuration of each platform and pipeline system, or that global rates of different categories of incidents have been assessed for different types of platforms. For initiating events involving, for instance, severe sustained fire loads across the whole platform (of the type of the Piper Alpha accident)
|
From page 135... ...
v Eq. 28 OVERALL RISK The annual probability of at least one casualty in the Gulf due to pipelines away from platforms (explosions or sudden fires following collisions between a vessel and an exposed live pipeline transporting gas)
|
From page 136... ...
Similarly, the mean annual amount of property damage (including costs of repair of corroded pipelines) in the Gulf due to pipeline failures caused by any of the five considered accident initiators (vessel collision, dropped objects, corrosion, hurrianes and storms, and pipe failure at the site of a platform)
|
From page 137... ...
· Probability of at least one death due to explosion or sudden fire in a collision between a vessel and an exposed live pipeline transporting gas in each diameter range (can be derived from the probability distribution of the number of casualties in collisionrelated explosions) · Mean spilled volume following a collision of a vessel with a live pipeline of given diameter transporting oil · Mean value of the property damage per collision between a vessel and an exposed empty pipeline · Mean value of the property damage per collision between a vessel and an exposed live pipeline transporting oil or gas given that no explosion or sudden fire occurs Mean value of the property damage per collision between a vessel and an exposed live pipeline transporting gas given that an explosion or a sudden fire occurs Object impact: Define ranges of sizes for diameter of breaches Define ranges of spill duration · Mean annual number of events (impacts on live exposed pipelines of dropped objects including anchors)
|
From page 138... ...
- Mean annual number of occurrences of initiating events of each type at the site of each class of platform · Probability of at least one death conditional on the occurrence of each type of initiating event on each class of platform (PRA result) Average number of workers on each class of platform
|
From page 139... ...
APPENDIX E result) 139 Expected number of casualties per type of accident and class of platform Mean amount of oil spilled per type of accident and class of platform (PRA Mean property damage per type of accident and class of platform (PRA result)
|
Key Terms
This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More
information on Chapter Skim is available.