Page 263 - Pipeline Risk Management Manual Ideas, Techniques, and Resources
P. 263

1 V240 Distribution Systems
           other  pressure  control  devices  is  critical.  The  evaluator   Product hazard
           should seek evidence that regulator activity is monitored and
           periodic overhauls are conducted to ensure proper perform-   Note that a chronic component of a product hazard is often
           ance. Other pressure control devices should similarly be closely   enhanced where a leaking product can accumulate in buildings,
           maintained. The care of an odorization system in a gas distribu-   beneath  pavement, etc.  This  is  generally  considered when
           tion system should also be included with maintenance proce-   assigning RQ points to substances such as methane. The evalu-
           dures.                                     ator is encouraged to review pages 138-142  to ensure that the
            Score the maintenance practices as described in Chapter 6.   reasoning behind the RQ assignments are appropriate for the
                                                      evaluation.
                                                       In the case of water systems, the main product hazard will be
           X.  Sabotage                               related to the more mechanical effects of escaping water. This
                                                      includes flood, erosion, undermining of structures, and so on.
           The risk of sabotage is difficult to fully assess because such   The potential for people to drown as a result of escaping water
           risks are so situation specific and subject to rapid change over   is another consideration.
           time, The assessment would be subject to a great deal ofuncer-   The product hazard variable can be assessed as described on
           tainty,  and recommendations may  therefore be  problematic.   pages 136142.
           Note,  however, that many current risk variables and possible
           risk reduction measures overlap the variables and measures that   Spill size and dispersion
           are normally examined in dealing with sabotage threats. These
           include security measures, accessibility issues, training, safety   One of the chief concerns of gas distribution systems operators
           systems, and patrol.                       is the potential for a hazardous material to enter a building
            The likelihood ofapipeline system becoming a target of sab-   intended for  human  occupancy.  In  a  city  environment, the
           otage is a function of many variables, including the relationship   potential is enhanced because gas can migrate for long dis-
           of  the  pipeline  owner  with  the  community  and  with  its   tances under pavement, route through adjacent conduits (sewer,
           own employees or former employees. Vulnerability to attack is   water lines, etc.), permeable soils, or find other pathways to
           another aspect. In general, the pipeline system is not thought to   enter buildings.
           be more vulnerable than other municipal systems.  The motiva-   For more catastrophic pipe break scenarios, and as a model-
           tion  behind  a  potential  sabotage episode would,  to  a  great   ing simplification, spill size can be modeled as a function of
           extent,  determine  whether  or  not  this  pipeline  is  targeted.   only pipe diameter and pressure as discussed on pages 142-143.
           Reaction to a specific threat would therefore be very situation   The  underlying assumption in  most  consequence assess-
           specific.                                  ments is that higher spill quantities result in higher potential
             Guidance documents concerning vulnerability assessments   damages. The drain volume and flow stoppage time (reaction
           for municipal water systems are available and provide some   time)  are  determining factors for total  volume  released on
           potential input to the current risk model. An  effort could be   water systems. In simplist terms, low spots on large-diameter,
           undertaken to gather this information and  incorporate sabo-   high-flow-rate  pipelines can be  the sites of largest potential
           tage and terrorism threats into the assessment, should that be   spills and larger diameter, higher pressure gas pipeline mains
           desirable.                                 can generally cause greater releases.
             See Chapter 9 for more discussion on sabotage issues and   As discussed in Chapter 7, leak size is also function ofthe fail-
           ideas for risk assessments.                ure mechanism  and the  material  characteristics.  Smaller leak
                                                      rates tend to occur at corrosion (pinholes) or some design failure
                                                      modes (mechanical connections). The most costly small leaks
           XI.  Leak impact factor                    occur below detection levels for long periods of time. Larger leak
                                                      rates tend to occur under catastrophic failures such as external
           In general, the leak impact factor (LIF) for a distribution sys-   force (equipment impact, earthquake, etc.), avalanche crack fail-
           tem  can be scored in  a manner similar to that described in   ures, and with system shocks to graphitized cast iron pipes.
           Chapter 7.  Some key points of consequence assessment are   In assessing potential hole sizes, the failure mechanism and
           reiterated and some considerations specific to distribution sys-   pipe material properties would ideally be considered. As noted,
           tems are discussed below.                  a failure mechanism such as corrosion is characterized by  a
             As in the transmission model, both multiplication and addi-   slow removal of metal and, hence, is generally prone to produc-
           tion operations can be used to better represent real-world rela-   ing  pinhole-type  leaks rather  than  large  openings. Outside
           tionships. For example, a top-level equation,   forces, especially when  cracking is  precipitated, can  cause
                                                      much larger openings. The final size of the opening is a func-
            LIF = @roduct hazard) x (spill) x (dispersion) x (receptors + outage),   tion of many factors including stress levels and material proper-
                                                      ties  such  as  toughness.  Because  so many  permutations of
             captures the idea that the overall consequences are propor-   factors are possible, hole sizes can be highly variable.
           tional to the spill size and product hazard. If either variable is   The risk reduction benefits of a leak detection and response
           zer+no  spill or no product hazard-then  there are no conse-   system can be captured in the spillscore. The ability to reliably
           quences. It also shows that locations where both receptor dam-   minimize the exposure time or area of exposure needs to be
           age and losses due to service interruption (outage) are high are   measured,  at  least  in  some  general way,  in  order  to  score
           the most consequential. As either or both of these are reduced,   these aspects. The leak detectiodreaction  capabilities can be
           so too is overall consequence.             assessed at all points along the pipeline and are a function of
   258   259   260   261   262   263   264   265   266   267   268