Docstoc

Risk Management Failures Case Study

Document Sample
Risk Management Failures Case Study Powered By Docstoc
					       Risk Management: A Case Study in Operating PCCP
                Brian J Mergelas1, David L. Atherton2, Xiangjie Kong3

Abstract

The likelihood and consequence of failure of a single pipe plays a large role in the
effective operation of a pipeline system. Condition assessment on each pipe is possible
using recent and historical data and identifying potential failure points along a pipeline.
Cost-effective rehabilitation strategies and plans are based on the prioritization of such
analyses. Predictive methods for pipeline failure and consequences are possible as well,
allowing proactive preventive maintenance. The stages of collection, assessment and
analysis, prioritization and action form a risk management strategy, demonstrated here as
a composite case study.

Introduction
This paper represents a composite of several of the owners and operators of Prestressed
Concrete Cylinder Pipe (PCCP) who have experienced failures or who may have
concerns regarding the potential for such failures. Prestressed Concrete Cylinder Pipe is
used throughout North America and the world to convey large volumes of water. The
estimated replacement value of the PCCP in North America alone approaches
$50,000,000,000. As our infrastructure ages it will become more and more necessary to
be proactive about how we manage the operations. With the introduction of new
legislation, notably GASB 34, Utilities will be required to account for the present value
of their infrastructure. It is clear now, more than ever, that assessing the condition of our
critical infrastructure and developing effective ways to manage the risk inherent in
operating any large system will be crucial both in maintaining reasonable cost and in
enabling the extension of the safe economic life of the pipes.

Prestressed concrete cylinder pipe is comprised of a thin watertight steel cylinder that is
either embedded or lined with concrete before it is spirally wrapped with high strength
steel wire under very high tension. These prestressing wires are designed to ensure that
no matter what worst-case internal pressure and soil loading is applied to the pipe, the
concrete core will always remain in compression and the pipe will continue to operate
safely.




1,3
  The Pressure Pipe Inspection Company Ltd, 4700 Dixie Rd. Mississauga, Ontario, Canada L4W 2R1
Phone: (905) 624-1040, email: mergelas@ppic.on.ca
2
 Queen’s University, Kingston, Ontario, Canada K7L 3N6
Phone: (613) 533-2701, email: dla@physics.queensu.ca




                                                1
Failures of PCCP have occurred when many of the wires loose their prestress and
operating pressures then exceed the reduced critical limits. Breaks in the prestressing
windings may be caused by third party damage, construction damage during installation,
corrosion and or embrittlement of the wires due to environmental factors.

PCCP is frequently used to convey water over long distances. Rupture of these critical
arteries of our infrastructure can have serious consequences on many fronts including
health & safety, economic and political. This kind of failure can be classified as Class III
failures based on Makar’s classification system [1]:

   Class I failures: Routine pipe breaks that are repaired as they occur. Example: day-
   to-day small diameter gray cast-iron failures.

   Class II failures: Pipe failures that would ordinarily be considered class I, but that
   will be investigated more thoroughly as part of a larger program of research into the
   behaviour of the system or a local area. A full or partial analysis of the failure is
   carried out for internal use only. Examples: small–diameter gray cast-iron failures
   during a sampling campaign, pipes in a region with higher than expected failure rates.

   Class III failures: Pipe failures with serious consequences in terms of number of
   customers losing water supply, damage to surrounding environment and/or other
   direct, indirect or social costs. This failure analysis should be conducted on the
   assumption that it will be reviewed by other interested parties, such as utility
   managers, local politicians or litigants. Examples: large-diameter pipe failures,
   small-diameter failures on lines that lead to hospitals or water-intensive industries,
   failures that may result in litigation.

In order to prevent such catastrophic failures and ensure the long-term viability and
continued supply of fresh water to our communities, we need to consider risk
management strategies that are both efficient and effective.


Risk Management
Risk is involved in operating any pipeline. A basic risk assessment model (Figure 1) has
been described by Muhlbauer [2]. The oil and gas industry leads the water industry in its
ability to quantify and manage the risk in operating a transmission pipeline. Risk
Management practices used in the oil and gas industry typically include such steps as
sectionalization of the pipeline, assessment of the condition of the pipeline, the potential
for third party damage, the design and operating parameters of the line, and the
consequence of sudden failure as well as monitoring techniques and preventive actions to
reduce the possibility of future problems.

There are several levels of concerns in the risk management practices of the oil and gas
industry which are not structurally relevant to the water industry. The number one reason
for failure of an oil and gas transmission main is third party damage. Thus one of the


                                             2
main risk management strategic involves communication to the general public regarding
the location of buried infrastructure. In addition, the consequence of failure of a water
pipeline does not include many of the hazardous environmental concerns associated with
an oil or gas main break. There is little risk of contamination or a flammable explosion,
although the failures themselves can be catastrophic since the release of large quantities
of high-pressure water can cause considerable physical damage.

The basic Risk management strategy for water supply pipelines could be described in the
following list:

   •   Assess condition
   •   Determine risk of failure
   •   Assess potential consequences of failure, including specific effects of location and
       time
   •   Determine cause of failure
   •   Develop a mitigating strategy (including repairs/ replacement/ modification of
       operating procedures)
   •   Monitor and Reassess condition

In general distressed pipes in a line will continue to degrade over time. One solution is to
replace ALL pipes that contain wire breaks as soon as they are identified. While this
should lead to a lower overall risk of operating a pipeline, it certainly is a costly course of
action to take. There are now various levels of decision-making processes that are being
employed by utilities to both minimize their operational risk while also optimizing the
investment of expenditures on repairs to preserve or even enhance the asset value of the
waterline infrastructure.

Case Study
In the following examples we describe the actions taken by several different utilities in
order to capture the range of possible actions currently being utilized and employed.

   Utility A – After a catastrophic failure in a section of pipe an investigation revealed
   several factors leading to concern regarding the condition of the pipes in this section.
   These factors included the soil conditions, the age of the system and irregularities in
   the rate of installation of this section of the pipeline. Based on this information the
   utility replaced several miles of pipe and incurred significant non-budgeted costs.

   Utility B – Assessed the condition of the pipeline using RFEC/TC. The inspection
   identified those sections that showed wire breaks and also identified the number (and
   location) of wire breaks in all of the distressed pipes.

   At the time, no information was available regarding the probability of failure but,
   based on how the distressed pipes were grouped, a decision was made to line long
   lengths of the pipeline with a steel liner. Several isolated pipes in deep cover were



                                              3
repaired using internal Carbon fiber. (Other utilities have used external tendons to
repair their pipes.)

The decision regarding which pipes to repair or replace was made using an ad hoc
analysis based on the number of breaks and the location and frequency of distressed
pipes (Figure 2).

This was again a significant investment but the selection of sections with numbers of
distressed pipes enabled considerable cost savings compared with replacement of the
entire pipeline.

Utility C – Assessed the condition of the pipeline using RFEC/TC. The inspection
identified those pipe sections that showed wire breaks and identified the number (and
location) of wire breaks in all of the distressed pipes.

Applied advanced structural analysis of PCCP using known modes of pipe failure to
predict behaviour of distressed pipes. A failure prediction curve built for each pipe
class and the expected maximum pressure and soil loading seen by each pipe was
used to categorize the distressed pipe into repair priorities.

In some cases the priority of a pipe (risk of failure) was lowered by changes to the
pipeline operation and / or control system. The failure predictions included cracking
of the concrete core which could lead to weakening of steel cylinders which might
otherwise be strong enough to contain the internal pressure of some low pressure
lines.

Decisions on which pipes to repair were made based on the priority or probability of
failure as well as the distribution and location of distressed pipes. The utility also
included other parametric data sets such as environmental or geographic, access and
land use considerations to make the best possible decision regarding rehabilitation
(Figure 3). This mode of data usage represents the status of many owners and
operators today.

Utility D – Performed multiple inspections using RFEC/TC to monitor the rate of
increase of wire breaks (Figure 4). Ideally this rate of increase can be used to assess
the frequency of inspection required to manage the system. Given a fixed set of
assumptions regarding the operating condition of the line it is possible to predict the
maximum number of breaks allowable before a pipe enters a higher priority class.
One may use the trends seen in the rate in number of breaks to determine a “safe”
interval between inspections.

In addition to multiple inspections with the RFEC/TC system the utility chooses to
perform strategic acoustic monitoring of the pipeline. Regions where there are
distressed pipes that will remain in the ground are monitored to determine ongoing
acoustic activity. In addition some sections may be monitored to provide more
information to supplement the RFEC/TC inspection.



                                         4
It is important to realize the following points regarding the rates of increase:

The number of breaks as determined by RFEC/TC is dependent on many factors
including the inspection conditions and the understanding of the analysis procedure. It is
important to maintain continuity when trying to compare inspection results over several
years. Since the knowledge regarding this technology continues to grow rapidly there is
significant merit in considering re-analysis of data from previous inspections while
performing a re-inspection. It is also important that the configuration of the test
equipment is maintained from one inspection to the next so that any subtle changes in the
reading from the pipe may be properly interpreted.

While acoustic sensors can be placed on PCCP and are able to detect active wire breaks
and other damage induced in the pipeline, these sensors are also sensitive to wire
movements. As such they measure acoustic activity not necessarily the rate of increasing
number of wire breaks. It is expected that, as more and more coordinated projects
utilizing the two technologies are performed, it may be possible to definitely distinguish
between wire breaks and slippage leading to further loss of pretension, however,
RFEC/TC can always be used to establish actual number of breaks in any pipe at the time
of an inspection and acoustic technology can be used to provide supplement monitoring
information. The two technologies complement each other. It is beneficial to have both
available to provide the state of art condition assessment and monitoring information to
manage water pipelines most effectively.


Conclusions
To effectively manage the operating risk for a PCCP pipeline one must balance the
relative cost of risk reduction strategies with the benefit of such activities.

By assessing the actual condition of a pipeline using RFEC/TC one can gather the
information required to start to predict the probability that individual pipe sections might
fail given present operating conditions. Repeated inspections using RFEC/TC is an
excellent way of establishing the rate at which the pipe distress is increasing. Thus we
can begin to predict what the future risk of operating the pipe will be. This information
can be used to determine a better course of action regarding repairs, rehabilitation and
future inspections.

The ultimate goal of Risk Management program is to provide a cost effective way of
maintaining the integrity of the Water Supply system in a safe and responsible way.
Well-maintained real estate lasts longer than poorly maintained property. It often
appreciates in value and revenue. The use of state of the art inspection and monitoring
information to manage waterlines effectively can not only reduce costs, improve safety,
reliability and extend lifetime but can also enhance the asset value and revenue through
enhanced property values and increased ratable tax base. It is typically a very high-
return, safe investment.


                                              5
References:

1. J. Makar, “Investigating Large Gray Cast-Iron Pipe Failures: A Step-by-Step
   Approach”, AWWA Infrastructure Conference, March 2001.

2. W.K. Muhlbauer, “Pipeline Risk Management Manual, Second Edition”, Gulf
   Publishing Company, 1996.




                                          6
                        Relative
                         Risk
                         Score
                                                               Dispersion
                                                                 Factor



                                                 Leak Impact
                                                   Factor



                         Index                                  Product
                          Sum                                   Hazard




                                                                Incorrect
      Third Party           Corrosion              Design      Operations
     Damage Index            Index                 Index          Index




                                   Data Gathered
                                   From Records
                                   And Operator
                                     Interviews




Figure 1. Basic risk assessment model [2].




                                             7
Figure 2. An ad hoc analysis can be done based on the distribution plot with the location
and frequency of distressed pipes.




Figure 3. Decisions can be made considering not just the distribution of distressed pipes
but also other parametric data sets such as geographic information or local soil
conditions.



                                            8
                        500

                        450

                        400
     Number of Breaks




                        350
                              Maximum number of breaks allowable for projected operating pressure
                        300
                                                                                                 ?
                        250

                        200

                        150

                        100

                        50

                         0
                                   April,99           Oct. 00              March, 02        Sept., 03

                                                                    Time


Figure 4. Multiple RFEC/TC inspections can be performed to monitor the rate of increase
of wire breaks (the dotted line and 4th point is predicted based on three inspections).




                                                                9

				
DOCUMENT INFO
Description: Risk Management Failures Case Study document sample