NTSB Safety Recommendation Letter P-78-058 through P-78-063

Page 1Official PDFNATIONAL TRANSPORTATION SAFETY BOARD WASHINGTON, D.C. ISSUED: October 25, 1978 Forwarded to: \ M r . L. D. Santman Director ( SAFETY RECOMMENDATION (5) Materials Transportation Bureau Department of Transportation P-78-58 through 63 Washington, D.C. 20590 The National Transportation Safety Board has completed a s p e c i a l study, t i t l e d "Safe Service L i f e f o r Liquid Petroleum P i p e l i n e s " (NTSB- PSS-78-1), which t h e Board undertook t o determine t h e f e a s i b i l i t y of developing a s a f e service l i f e model. The study included an a n a l y s i s of t h e DOT Form 7000-1 "Pipeline Carrier Accident Report" data t h a t is permanently f i l e d a t t h e Office of Pipeline Safety Operations (OPSO). These accident r e p o r t s have been collected on a regular b a s i s since 1968. During the examination of t h i s accident d a t a , t h e Safety Board was unable t o determine t h e existence of any OPSO accident d a t a a n a l y s i s plan beyond t h a t of providing cumulative accident t o t a l s by various categories f o r t h e annual OPSO report. The l a c k of a plan t h a t s t r i v e s t o provide more insight than yearly cumulative accident t o t a l s , combined with the f a c t t h a t mileages f o r various categories of e x i s t i n g p i p e l i n e s a r e not requested, makes more sophisticated data analyses impossible t o complete. Therefore, t h e Safety Board w a s only a b l e t o extend t h e OPSO cumulative accident t o t a l s from a series of 1-year periods t o a 9-year span. This extension served t o more c l e a r l y i l l u s t r a t e t h e accident trends i n t h e Liquid pipeline industry and showed t h e apparent effec- tiveness of t h e corrosion-prevention programs t h a t r e q u i r e cathodic protection, coated pipe, and e l e c t r o n i c inspections. These preventive measures have apparently caused corrosion accidents t o drop from a reported high of 229 i n 1968 t o 50 accidents i n 1976. This s i g n i f i c a n t decline has c u r r e n t l y placed corrosion-caused accidents a s t h e second leading cause of accidents, behind p i p e l i n e ruptures due t o e x t e r n a l l y operated equipment. 2484#

Page 2Also, the OPSO data showed that the reported ground cover over pipelines at rupture sites in the six leading States was less than 24 inches in 51 percent of the accidents. The industry standard now calls for at least 30 inches of ground cover. Unfortunately, pipeline mileage data by depth of ground cover are not available so that It was impossible to determine whether greater depth would be effective in reducing the frequency of these ruptures. A n investigation into the accuracy and completeness of the accide reporting on the DOT Forms 7000-1 indicated some accident investigators do not understand how to correctly complete the 7000-1 forms nor do they always provide the requested data. The accuracy problem was illustrated through selections of either incorrect or redundant "origin of liquid or vapor release" entries that pertained to the accident causes "defective pipe" and "defective welds." The Safety Board concludes that clearer and more concise instructions and definitions would be effective in obtaining accurate data. The completeness problem was shown by the lack of responses in data fields such as "Grade of Pipe," "Pressure Test Duration,'' "Maximum Test Pressure," "Year of Pressure Test ," and "Time Between Corrosion Tests." The Federal government currently does not have an automated capa- bility for systematically maintaining a surveillance of the 125,000 miles of interstate liquid petroleum pipelines. The purpose of such a surveillance should be to provide guidance relating to situations where the pipelines may be deteriorated and for safety purposes should operate at a reduced pressure, transport a less hazardous product, or cease operation until upgraded or replaced. A computerized system could be implemented to compute leak rates-per-mile of pipe for each pipeline carrier as well as t o perform basic statistical tests to determine which carriers have leak rates that vary significantly from a nationwide rate. Additional computations could be performed to determine risk levels to persons and properties near the pipelines in question. The great losses arising from liquid propane gas (LPG) accidents clearly shows the need for more definitive regulation. The Safety Boar has been calling for more stringent controls of LPG pipelines since 1971. No action had been taken by the DOT until August 1978 when two Notices of Proposed Rulemaking (NPRM) to strengthen Federal regulatio dealing with LPG pipelines were issued. DOT indicates that additional NPRM's covering other areas also will be issued in 1978. The SafetyBoard believes that solution of the LPG problem should dominate regu- lation of liquid pipelines and strongly urges DOT to expedite of its LPG rulemaking.#

Page 3- 3 - Therefore, the National Transportation Safety Board recommends that the Office of Pipeline Safety Operations: Publish a plan that describes how the OPSO will use accident report data to formulate safety regulations and to develop a safe service life model for pipelines. (Class 11, Priority Action) (P-78-58) Redesign the Liquid Pipeline Accident Report System to include data similar to that collected in the Natural Gas Accident Reporting System. (Class 111, Longer Term Action) (P-78-59) Provide clear instructions and definitions to insure the accuracy and consistency of the data recorded on the liquid pipeline acci- dent report forms. (Class 111, Longer Term Action) (P-78-60) Computerize the redesigned Liquid Pipeline Accident Report System. Include the capability to: a. compute the historical accident/leak rate-per-mile of pipe for each carrier as well as the nationwide rate; b. make periodic comparisons of each carrier's accident/lealc rate against the nationwide accident/leak rate; c. compute and plot selective accident/leak rates based on pipeline parameters such as age, specified yield strength, depth of cover, product transported, etc; d. selectively retrieve and summarize accident/leak data pertain- ing to any given accident or classification of accidents; e. produce summarized reports reflecting the above-listed information. (Class LIZ, Longer Term Action) (P-78-61) Conduct audits of the completed liquid pipeline accident reports to insure that mandatory data is provided. (Class 111, Longer Term Action) (P-78-62) Expedite completion of the rulemaking to strengthen the Federal regulations concerning LPG pipelines. (Class 11, Priority Action) (P-78-63) K I N G , Chairman, DRIVER, Vice Chairman, McADAMS and HOGUE, Members, concurred in the above recommendations.#