INTEGRATING RISK BASED INSPECTION WITH FITNESS FOR PURPOSE presented by John Aller Bill Zhu Doug Paneitz Capstone Engineering Services, Inc. Houston, TexasABSTRACT Risk Based Inspection (RBI) is 2 relatively new technology being applied in the Chemical Process Industry. It provides a systematic process for factoring risk into decisions concerning how, where and ‘when to inspect a process plant, The intent of RBI is to focus inspection resources on critical equipment in ways which will prevent failures, particularly catastrophic ones, Such a tool shows the potential to ‘make chemical plants safer in a cost effective way. One of the advantages of RBI is that it can provide a dynamic (time dependent) means of predicting when as well 2s where to expect in-service deterioration. When damage is confirmed, a Fitness for Purpose evaluation can be used to decide on an appropriate maintenance strategy. This paper will briefly discuss one method for assessing risk in Pressure Vessels and Piping, present actual results from a study on 2 30 year old ethylene plant, and demonstrate the link between Risk Based Inspection and Fitness for Purpose. Copyright 1996 by Capstone Engineering Services, Inc. BINTRODUCTION Risk Based Inspection is receiving increasing attention and application in the Refinery and Petro- chemical Process Industry since the first reported application in 1992. [ 1 } Its application corresponds to the increased industry attention on developing and implementing more effective Process Safety Manage- ment (PSM) systems. [2] PSM requires that all process equipment items containing hazardous chemi- cals be managed in a systematic way to ensure their mechanical integrity. ‘The intent of Process Safety ‘Management is to prevent catastrophic releases or minimize their consequences. Some equipment items have the potential for catastrophic releases, while others do not. One of the functions of Risk Based Inspection is to identify the items which have the potential for catastrophic failure and to focus inspection and maintenance resources to prevent that equipment from failing, Figure 1 Risk Based Inspection R Risk with Typical Inspection Program t s 7 # Risk Using RBI INSPECTION COSTS: OVERVIEW OF RBI Risk Based Inspection is a method for using risk as a basis for prioritizing and managing the efforts of an inspection and maintenance program. It has been demonstrated that the application of an effective risk based program can simultaneously reduce the risk in an operating unit as well as reduce the cost of the inspection effort. Such a concept is illustrated in Figure 1. In an operating plant, a relatively large percentage of the risk (80-95%) is associated with a small percentage (5-20%) of the equipment items. Risk Based Inspection manages a shift of inspection and maintenance resources to provide a more signifi- cant effort concentrated on the high risk items and an appropriate level of effort directed toward the Tower risk equipment, 4IN Probability Ranking ° Eo) G8 A Consequence Ranking Figure 2 - Risk Rating Risk Based Inspection defines the risk of operating equipment using a prioritization process which combines two separate factors: the likelihood and the consequence of failure. ‘The prioritization process can vary from qualitative to quantitative, but the intent is to characterize each equipment item with a likelihood and consequence of failure. A semi-quantitative method for risk rating, using a 5 x 5 matrix as shown in Figure 2, is a method which has been demonstrated to have an appropriate balance between being simple to implement while still providing adequate discrimination between the items being evalu- ated. ‘The likelihood of failure of a specific piece of equipment from inspectable causes is a function of the identified failure mechanisms for that item, the rate of deterioration, and the effectiveness of the prior inspection history in detecting and properly characterizing the damage to the equipment. ‘The safety consequence of failure of the item is a function of the type of fluid the equipment con- tains, how much might be released in the event of a failure, and the effect of such a release. Well defined methods have been developed to characterize the safety consequence of a release. 15Risk Based Inspection Flow Chart ‘opening Condes “Restle Cnoeuece Tone Coneemce Risk Rating Inspection Planning implementation [Ua ara] Figure 3 Once the Risk Ranking has been identified for a piece of equipment, an appropriate inspection and maintenance strategy can be developed, An understanding of the identified failure mechanisms and their severity will allow the selection of appropriate inspection methods, ‘The inspection frequency and cover- age can also be adjusted, based on the results of the risk analysis. ‘The Risk Rating estimates the probability of failure along with the failure consequence. For fixed equipment, the most significant failure mode is usually loss of containment, which was the basis for this study. Using the method described above, both the consequence and probability rankings are calibrated in order-of-magnitude steps. This approach is considered a semi-quantitative method. Setting up an effective risk based inspection program in a process plant requires a thorough know!- edge of the potential failure mechanisms. The identified failure mechanisms in an ethylene plant can include a variety of mechanisms, depending on the composition of the feed stock and the specific process. ‘Typical ethyiene plant degradation mechanisms might include corrosion under insulation (CUD), internal corrosion, amine and caustic stress corrosion cracking, fatigue, liquid mercury embrittlement and brittle fracture ‘The RBI methodology should be able to effectively deal with all of the damage machanisms that are identified in a plant corrosion study. The reader is referred to several papers in the reference section for more details on the application of Risk Based Inspection. [ 3-10 } 16BACKGROUND OF STUDY A recent Risk Based Inspection study was conducted on a 30 year old ethylene plant. The intent of the study was to help the plant set a fixed equipment inspection and maintenance strategy prior to and during an upcoming shutdown. ‘The objectives of the study included the identification of corrosion “hot spots” (locations likely to have active damage mechanisms), an assessment of prior inspection effectiveness, assignment of internal and external corrosion rates for all equipment items in the study, identification of items to be included on ‘a pre-turnaround inspection, and items which should be internally inspected during the turnaround. ‘The work process started at a Process Flow Diagram (PFD) level by assigning “areas” and “sys- tems” to the Inside Battery Limits (ISBL) portion of the plant, An “area” is a logical section of the operating unit, from feed to product storage. A total of 13 areas were assigned to the plant. A “system” is an interconnected group of equipment and piping which contain essentially the same composition, The study identified 62 systems in the unit, ‘The “system” is used to establish process inventories, based on the equipment in or directly attached to the system. Equipment items were assigned to systems and areas, Data was downloaded from plant databases as well as gathered from paper files, and used to estab- lish the scope of the work. Data sheets were printed for each equipment item, They were used for additional data collection, data entry into an RBI program, and quality assurance. Risk Ratings were calculated for each item, using the approach described in the previous section. During the first phase of the study, pressure vessels, (drums, towers and driers), piping circuits representing each of the systems, and heat exchangers (shell side) were analyzed. Based on the age of the plant, prior inspection information, and the analyst experience, reasonable corrosion rates were assigned to the equipment and piping circuits. The materials of construction, corro- sion rates and design data were used to perform the likelihood of failure analysis for the equipment items. Both internal and external corrosion were evaluated. Process conditions and system inventories were used to calculate the consequence of failure for each item. The likelihood and consequence were used to establish the Risk Rating for each item. ‘The preliminary corrosion study determined there were relatively few active damage mechanisms in the ethylene plant. By far the biggest problem identified was the corrosion under insulation (CUD. RESULTS OF RISK BA‘ RITIZATION A total of 75 pressure vessels, 18 columns, 97 exchangers and 105 piping circuits were evaluated in the study. The columns were evaluated as two sections (top and bottom) to more accurately model the consequence of failure. For this part of the study, the exchangers were modeled on the shell side only. ‘The following table summarizes the distribution of criticality ratings for the various equipment types. 17Table 1 DISTRIBUTION OF CRITICALITY RATINGS Criticality Category ‘Equipment Type High ‘Med. High] Medium] Low ‘Vessels i n 24 * Exchangers 6 47 28 16 Piping ie 21 29 55 0 Totals 38 147 10 a caro) Distribution by Risk Category 160 140 120 100 80- 60. 40. 20: 0. High M-High Medium Low Figure 4 Figure 4 shows the distribution of the risk for all equipment items by the four criticality categories. The distribution is skewed toward the higher risk, since about 90% of the items included in the study contain flammable products, About 10% of the items contain cooling water or steam, mostly the shell sides of the heat exchangers, The equipment count in the high and medium-high risk categories is rela- tively high, even for an ethylene plant. This in turn decreases the “medium” count. The low category has been forced to a low count by selecting predominately PSM Covered (higher consequence) equipment ‘An appropriate inspection and maintenance strategy will manage all equipment so none remain in the High Criticality Category and the count in the Medium High Category is reduced significantly 18‘Combined Risk Distribution Internal and External Corrosion 313 Figure 5 All items were evaluated for internal corrosion and corrosion under insulation, Figure 5 shows the distribution of the total equipment count by the combined (internal and external) likelihood of failure categories. A Likelihood Category of “4” indicates equipment is essentially in “like new” condition. Approxi- mately 70% of the equipment items fall into this category. The other 30% of the items have an increased probability of failure. Thirty two items (80% piping) are in Likelihood Category 1, which predicts a very high likelihood of failure The increased likelihood of faiture is dominated by the external corrosion (CUI) predicted by the RBI corrosion study. The age of the plant and a relatively low confidence in the prior inspections have contributed to the significant number of items with an increased likelihood of failure from extemal corro- sion. Very few of the higher risk items are driven by intemal corrosion. The low likelihood of failure from internal corrosion results from few of the equipment items containing corrosive process fluids. In addition, there is a relatively high confidence in the internal corrosion rates because of good prior inspec- tion data for intemal corrosion. 19Risk Category Distribution Internal vs External Corrosion 180. 180: 140; 120! 400! 80; 60; 40) 20. 0 Figure 6 Brittle fracture and fatigue were identified as potential failure mechanisms in a few items, but the screening step demonstrated no significant increase in likelihood of failure as result of either mechanism, The results of the consequence of failure analysis are also described in Figure 5. In this study, the consequence analysis was primarily based on the flammable consequence of hydrocarbon streams. Acid and caustic streams were analyzed as reactive (to the human body) chemicals, although the consequence of the release of a such chemicals is generally much lower than for flammable or toxic materials, They ‘were all “D” or “E” consequences. The chlorine system was the only toxic analyzed, and it was an “A” consequence ON U) INSULATION PROGRAM Asa result of the study, the decision was made to go into an aggressive on-stream CUI inspection program using Visual Inspections, Real Time and Profile Radiography. The results of the RBI study were used to help prioritize the RT activities. A high percentage of the items predicted to have significant damage from CUI (Probability Ranking of 1 or 2) were actually found to have damage. The most signifi- cant findings of the study were four towers with severe corrosion in the area of the skirt to shell attach- ment. ‘The evaluation of the corroded areas is discussed in the following section. FITNESS FOR SERVICE EVALUATION Visual inspections and UT examination verified that four towers had significant wall loss in the cylindrical shell to skirt attachment area. The following discussion focuses on one of the towers, the Caustic Wash Tower. The nominal wall thickness of the skirt was 2.032", but remaining wall thicknesses were found to be as low as 0.185”, ‘The strength of the skirt support was questioned, especially under possible high wind load conditions. A fitness for purpose analysis of the bottom shell course, bottom head, and skirt was conducted to develop repair recommendations 20In this project, the strength of the tower under operation conditions was evaluated according ASME Code, Section VIII, Division 1, Rules for Construction of Pressure Vessels, 1995 Edition. [ 11 ] ‘The wind load analysis was performed via ASCE 7-93/88, Minimum Design Loads for Buildings and Other Structures. (121 ‘The minimum wall thickness of the cylindrical shell and ellipsoidal heads under pressure were calculated with the follow data and results: Table 2 Minimum Wall Thickness Required by Pressure [Parameter Unit | Symbol | Cylindrical Shell | Ellipsoidal Head Design Pressure psi P 614 614 (Design Temperature - T 515 SiS Static Head psi 00 0.0 Inside Diameter in D 66.0 66.0 Head Factor - K NIA 10 Material SA-516 Grade 70 | SA-516 Grade 70 Joint Efficiency E 1.00 1.00 Allowable Stress at Design Temperature | psi s 17,500 17,500 |Corrosion Allowance in CA 0.0 0.0 [Minimum Wall Thickness in t 1.183 1.162 The allowable longitudinal tensile stress, under combined moment, pressure and weight loading, is the ‘ASME maximum allowable stress. [ 13] When combining pressure and weight loading with the load due to wind or seismic forces, due to the temporary nature of these loadings, the allowable stress is 1.2 times the ASME maximum allowable stress, as stated in the ASME Code UG-23 (4). ‘The maximum allowable longitudinal compressive stress is determined in accordance with the ASME Code, Section VII, Division 2, Paragraph UG-23 (b). According to UG-23 (b), this allowable stress should be the smaller of the following values: (the maximum allowable tensile stress value; Gi) the critical buckling stress with a safety factor of 2. For the combination of wind load or earthquake loading with other loading, the allowable longitudinal compressive stress is 1.2 times the smaller values of (i) or (ii) as stated in ASME UG-23 (4). ‘These allowable stress values are calculated and compared with the calculated combined moment, pres- sure, and weight induced longitudinal stress at the bottom of each segment. ‘Two stages of calculations were performed. Stage I was to determine the minimum wall thickness of the skirt. Stage I was to decide the minimum wall thickness of the corroded shell bottom course. In both stages, the minimum wall thickness of the skirt and the bottom course were determined by an iterative procedure. 21Table 3 Minimum Wall Thickness Requirement Component Calculated Measured Min. Wall Thickness,” [in] ‘Wall Thickness [in] Skirt 0.355 0.185 Bottom Head 1,183 1.375 Shell Bottom Course 1.183 1.375 * Corrosion Allowance was not included in the minimum thickness values. ‘The measured wall thickness of the skirt was less than the minimum wall thickness. Weld repair of that corroded area to the minimum wall thickness plus corrosion allowance will be performed during the tumaround. The actual wall thickness of the bottom course and the bottom head are still above the ‘minimum wall thickness, therefore no immediate welding repair of the shell was recommended. How- ever, the tower will be blasted and painted during to shutdown to protect the metal surface and prevent farther corrosion. The measured wall thicknesses of each of the corroded towers has been input into the Risk Based Inspec- tion Program and the updated risk of the corroded vessels will be used to establish their future inspection plans.LITERATURE CITED 10. ML. 12, 13, JE. Aller, R. Dunlavy, K.R. Riggs and D, Perry, “The Risk Based Management System: A New Tool for ‘Assessing Mechanical Integrity”, PVP-Vol. 251, Reliability and Risk in Pressure Vessels and Piping, ASME, New York, (1993) OSHA 29 CFR 1910.119, “Process Safety Management of Highly Hazardous Chemicals”, U.S. Depart- ‘ment of Labor, Occupational Safety and Health Administration, Washington, DC. (February 24, 1992) JE. Aller, RL. Dunlavy, S.A. Leppard, “A Tool for Risk Based Management of Mechanical Integrity”, First International Symposium on Process Industry Piping, The Materials Technology Institute of the Chemical Process Industries, Inc., St. Louis, (December 14-17, 1993) IT. Reynolds, “Risk Based Inspection Technology Improves Safety of Pressure Equipment”, Oil and Gas Journal, (January 16, 1995) LE, Aller, N.C. Horowitz, J.T. Reynolds, B.J. Weber, “Risk Based Inspection for the Petrochemical Industry”, PVP-Vol. 296, Risk and Safety Assessments, Where is the Balance?, ASME, New York, (1995) MJ. Conley, .E. Aller, A. Tallin and B.J, Weber, “Risk Based Management of Piping Systems”, Second Intemational Symposium on the Mechanical Integrity of Process Piping, The Materials Technology Insti- tute of the Chemical Process Industries, Inc., St. Louis, (January 30 - February 1, 1996) “Workshop on Experiences in the Application of Risk Analysis to Operating Equipment”, Pressurc Vessel Research Council of the Welding Research Council, Inc. New York, (April 17-18, 1996). John Reynolds, “Petrochemical Industry Standards activity Aimed at Improving the Mechanical Integrity of Process Piping”, Second Intemational Symposium on the Mechanical Integrity of Provess Piping, The Materials Technology Institute of the Chemical Process Industries, Inc. St. Louis, (January 30 - February 1, 1996) ASME Research Task Force on Risk-Based Inspection Guidelines, “Risk Based Inspection - Development ‘of Guidelines, Volume 1, General Document, CRTD-Vol.20-1”, American Society of Mechanical Engineers, Washington, D.C. (1991) JE. Aller and M.N. Tayloe, Materials Engineering and Risk Management in Chemical Plant Operations, MTI Publication No. 44, The Materials Technology Institute of the Chemical Process Industries, In., St Louis, (1995) ‘The American Society of Mechanical Engineers, “ASME Boiler and Pressure Vessel Code, Section VItI, Rules for Construction of Pressure Vessels, Division 1”, 1995 Edition “American Society of Civil Engineers, “ANSU/ASCE 7-93, Minimum Design Loads for Buildings and Other Structures”, 1993 Edition “The American Society of Mechanical Engineers, “ASME Boiler and Pressure Vessel Code, Section I Materials, Part D — Properties”, 1995 Edition 23