67.1.1 Risk-Based Inspection for Pipelines

The use of risk-based inspection (RBI) for pipelines is increasing in demand as government regulators worldwide are requiring operators to use risk assessment in high-consequence areas. Historical statistics attribute piping failure as the leading cause of large property losses globally [1]. Data from numerous corporations indicate that the most frequent failures (approximately 35–40%) are from piping systems.

There are a number of international design codes and standards covering pressurized equipment that should be referenced when considering a piping program. The Norsk Sokkels Konkuranseposisjon (NORSOK) standards, developed by the Norwegian petroleum industry, were intended to supersede individual company specifications as a reference for government regulators. ISO 31000, Risk Management—Principles and Guidelines, is a family of international risk management standards that provide generic guidance for any risk management implementation. Other ISO and OGP standards focus more specifically on risk and integrity management specifications and techniques.

This chapter focuses on in-service American Petroleum Institute (API) codes and standards that were developed by representatives from multinational oil and gas companies through an ANSI-accredited balloting process. API 510, 570, and 653 and associated documents define inspection requirements for pressurized equipment, while recommended practices API 580 and 581 provide the necessary guidance on RBI programs and methodology. RBI methods and applications for inspection planning, execution standards, and recommended practices are also described by American Society of Mechanical Engineers (ASME). Finally, the assessment of fitness-for-service and remaining life (RL) using API 579-1/ASME FFS-1 can be used to justify continued service when damage is found during inspection.

Risk-based principles allow low-risk and low-consequence systems to run to failure. This risk concept implies that other systems are high in risk and that the consequence of failure (COF) is unacceptable; as such, these systems should receive attention even when the probabilities of failure are not significantly high. Risk-based principles challenge accepted design codes that are based on deterministic design and fitness-for-service codes, particularly where worst-case scenarios are used without consideration for COF. Differences in inspection requirements and remedial action will occur when deterministic methods are compared with recommendations from risk-based methodologies.

The purpose of a RBI program is to develop risk assessment in order to implement an integrity inspection program over a defined period with the available data and detailed technical data. With inspection planning focused on high-risk items, inspection generates the necessary technical data required for risk assessment updating and fine-tuning future priorities. RBI methodologies can be used on older facilities that have little or poor data available; on newer assets where technical data are available, but little inspection history exists; and on facilities with good quality data and inspection history.

A RBI program should be considered for the following pipeline assets:

• Trunk lines from manifold stations to process facilities;
  
• Oil and gas NGL pipelines;
  
• Manifold stations;
  
• Process facilities for oil and gas separation.

Since facilities range widely in age, there is often limited construction, process, and inspection history available. Inspection and maintenance programs are sometimes reactionary, lacking a cohesive strategy or long-term plan based on expected in-service damage to equipment. Construction and process data and inspection histories are marginal or missing completely. The quality of available information on the assets determines whether a qualitative or quantitative RBI approach should be used. Less data generally require a more qualitative approach, at least initially. The approach adopted will depend on the following:

• Design and construction data readily available;
  
• Known equipment condition based on the inspection history;
  
• Knowledge of past, current, and future process operational conditions;
  
• Complexity and age of systems considered.

Regardless of the approach used, a continuous improvement process similar to one of the following should be implemented:

1. Conduct the initial RBI assessment using the available data.
   
2. Use inspection results to develop inspection requirements and priorities.
   
3. Implement initial inspection and plan over defined multiyear time period.
   
4. Update RBI assessment after each inspection.
   
5. Update the inspection plan from the updated assessment.

As the program develops and matures, a more quantitative analysis can be incorporated, as desired.

67.1.2 Scope

This chapter is primarily intended to be used for the planning of in-service inspection for pipeline static mechanical pressure systems when considering failures by loss of containment of the pressure envelope. Failure modes, such as failure to operate on demand, leakage through gaskets, flanged connections, valve stem packing along with valve passing, and tube clogging, are not addressed. The focus is on pipeline pressure systems involving the following types of components:

• Piping systems comprising straight pipe, bends, elbows, tees, fittings, and reducers
  
• Pressure vessels and atmospheric tanks
  
• Pig launchers and receivers
  
• Heat exchangers
  
• Unfired reboilers
  
• Valves
  
• Pump casings
  
• Compressor casings

Excluded from the scope are

• structural items, including supports, skirts and saddles;
  
• seals, gaskets, flanged connections;
  
• failure of internal components and fittings;
  
• instrumentation.

In this chapter, “equipment” is defined as any of the aforementioned, while a “system” refers to a combination or group in the aforementioned list.

67.1.3 Risk Analysis

Traditional risk analysis and risk management is based on applying a systematic approach to use available knowledge to assess, mitigate (to an acceptable level), and monitor risk. The goal is to identify all reasonably foreseeable detrimental impacts on asset integrity and reliability. Risk is assessed by identifying possible failures, estimating the probability of failures (POFs), and assessing the COFs. The results of the risk assessment are then used to reduce the impact of failure to as low as reasonably practicable (ALARP). The risk management process is a continuous activity that reflects changes in the operating environment and the need to monitor and maintain the performance of the resources.

The major concepts of a traditional risk analysis process comprise five tasks:

1. System definition
   
2. Hazard identification
   
3. Probability assessment
   
4. Consequence analysis
   
5. Risk results

A typical risk analysis scenario includes the following set of events: loss of containment; detection, isolation, and mitigation. Depending on the nature of the process and the detail of the study, a risk analysis can include thousands of different scenarios with evaluation of both the probability and the consequence of the set of events in each scenario.

67.1.3.3 Probability Assessment

A probability assessment is conducted to estimate the probability of occurrence for the scenarios identified in the previous phase of the risk analysis. If a scenario occurs fairly frequently, it is best to use historical data to estimate the event’s probability. However, if the events are rare and sufficient data do not exist to estimate their probability based on the historical data alone, a building-block approach is used. Probability estimates for all elements of the scenario are obtained and combined to predict the overall scenario probability.

The most common measure of probability for a scenario is frequency. A single fixed value is assigned for the frequency/probability term and is considered constant over time. Frequency can be used for a single event or a series of events. A year is most often used as the standard time interval for a frequency analysis. Frequencies may be very small numbers, such as one in a million years for infrequent events, or they may be relatively high values, such as once a month or four times a day. If, for example, a pipe is known to leak about every 5 years, it would have a leak frequency of one in 5 years or 0.2 per year. The term “recurrence period” is sometimes used to refer to the reciprocal of the frequency. In our example, the recurrence period for the leak would be 5 years.

To obtain the frequency of the scenario, multiply the frequency of the leak by the probability of all events that follow. The resulting likelihood is the scenario’ s frequency. The mathematical representation of the likelihood of the sequence, in terms of frequency, is as follows:

For example, consider a leak that does not ignite. If the estimated leak frequency is 10⁻⁵ and the probability of that leak not igniting was 0.5, the overall frequency for the no ignition scenario is 10⁻⁵ × 0.5 = 5 × 10⁻⁶.

67.1.4 The RBI Approach

RBI is a methodology that uses risk as a basis for prioritizing and managing the efforts of an inspection program, including recommendations for monitoring and testing inspection plans. It provides focus for inspection activity to specifically address threats to the integrity of the asset and the equipment’s capacity to operate as intended. The POF and COF are assessed separately and then combined to calculate risk of failure. Risk is compared and equipment risk prioritized for inspection planning and risk mitigation.

A RBI program includes the following:

• High-risk systems or processes within an operation prioritized by risk.
  
• Calculated risk value or category associated with an equipment item within a system or process based on a consistent methodology.
  
• Prioritized equipment rank based on risk.
  
• Development of an appropriate inspection program to address key drivers of equipment risk.
  
• A method to systematically manage risk of equipment failures.

A RBI program is not a fully quantitative risk analysis, but a technique that combines the disciplines of risk analysis and mechanical integrity. However, the techniques used in RBI are similar to those seen in traditional risk analysis. RBI methodology follows the five tasks of a traditional risk analysis described in the previous section, but it uses a modified and in some cases simplified approach:

• System definition: the system definition phase of an RBI program establishes ground rules and collects all pertinent information, as outlined in the system definition section.
  
• Hazard identification: while hazard identification is a critical step in a traditional risk analysis, the RBI approach focuses on the pressure boundary of equipment and assumes that failures are due to identifiable mechanisms of degradation in that boundary. Secondary causes of a leak, such as instrument failure or human error, are included implicitly in the RBI program’s treatment of management systems.
  
• Probability assessment: probability and consequence are limited to a carefully defined and limited number of scenarios. The RBI study expends a considerable effort in the assessment of the failure frequency and is generally considered to vary with time. The POF increases as degradation progresses and can be reduced when inspection is performed to better define the actual condition of the equipment.
  
• Consequence analysis: RBI consequence analysis is a relatively coarse determination compared with a traditional risk analysis. The effects considered may impact plant personnel or equipment, population in the nearby residences, and the environment. Depending on the ground rules established and material released, all of the consequence effects are usually calculated and the largest used for final risk.
  
• Risk results: risk is presented in a risk matrix (Figure 67.1), risk categories, iso-risk (Figure 67.2), or risk values for the purpose of prioritization.

67.1.5 Risk Reduction and Inspection Planning

RBI focuses on the development of inspection programs to address damage types that inspection should detect; the appropriate inspection technique to detect the damage; and the application of RBI tools to reduce risk and optimize the inspection program.

Inspection influences risk primarily by reducing the POF. Many conditions (design errors, fabrication flaws, and malfunction of control devices) can lead to equipment failure, but in-service inspection is primarily concerned with the detection of progressive damage. RBI considers that the POF is due to damage as a function of four factors:

• Damage mechanisms and resulting type of damage (cracking, thinning, etc.);
  
• Rate of damage progression;
  
• Probability of detecting damage and predicting future damage states with inspection techniques;
  
• Tolerance of the equipment to the type of damage.

RBI differs from conventional inspection management because it provides the concepts and methods to support decision making even when data are missing or uncertain. Rational decisions can be made using both quantitative and qualitative RBI methods.

The benefits of using RBI for inspection planning are as follows:

• Systematic overview of a system or process with an explicit and documented breakdown of the risks while clearly showing the risk drivers and recommending appropriate actions.
  
• Focused inspection efforts on items where the safety, economic, or environmental risks are identified as being high while the efforts applied to low-risk systems are reduced.
  
• Use of probabilistic methods to account for uncertainties in process parameters, corrosively, and degradation rates to calculate the extent of degradation.
  
• Consideration of consequences of failure to focus activities where they will have the most significant impact.
  
• Proactive and focused effort to ensure that overall system or process risk does not exceed the risk acceptance limit set by the authorities and/or operator.
  
• Optimized inspection or monitoring method opportunities according to the identified degradation mechanisms and the agreed inspection strategy.

The inspection planning process directs resources to accurately determine the condition of the equipment through the most effective and efficient inspection methods, balancing the necessary down time and cost of inspection against the effectiveness and benefits of inspection.

67.2.1 API Industry Standards for RBI

Operators considering a piping inspection program should be aware of several important industry design codes and standards that cover pressurized equipment. The API in-service codes and standards, including API 510, API 570, and API 653, and their associated documents should be used to define inspection requirements for pressurized equipment and tanks. Recommended practices API 580 and API 581 provide the necessary guidance on RBI programs and methodologies. RBI methods and applications for inspection planning, execution standards, and recommended practices are also described by ASME PCC-3-2007 Inspection Planning Using Risk-Based Methods. Finally, assessment of fitness-for-service and RL using API 579-1/ASME FFS-1 can be used to justify continued service when damage is found during inspection.

ASME PCC-3-2007 provides guidance on the preparation and implementation of a RBI plan based on API RP 580 RBI, the earliest industry standard on the practice. ASME PCC-3 and API RP 580 are closely aligned as changes, additions, and improvements are made to each document to enhance applicability for a broader spectrum of industries. ASME-PCC-3-2007 provides recognized and generally accepted good practices that may be used in conjunction with postconstruction codes, such as API 510 and API 570.

The purpose of API RP 580 is to provide users with the basic elements required for developing, implementing, and maintaining an RBI program. The risk analysis principles, guidance, and implementation strategies presented in this document are broadly applicable but were specifically developed for applications involving fixed pressure-containing equipment and components. API RP 580 provides guidance to owners, operators, and designers of pressure-containing equipment for developing, implementing, and assessing inspection programs. API RP 580 emphasizes safe and reliable operation through cost-effective inspection using a spectrum of complimentary risk analysis approaches (qualitative through fully quantitative) as part of the inspection planning process.

API RP 580 includes an introduction to the concepts and principles of RBI for risk management as well as the steps that should be followed in applying risk management principles within the framework of the RBI process. These steps include the following:

1. Understanding the design premise;
   
2. Planning the RBI assessment;
   
3. Data and information collection;
   
4. Identifying damage mechanisms and failure modes;
   
5. Assessing POF;
   
6. Assessing COF;
   
7. Risk determination, assessment, and management;
   
8. Risk management with inspection activities and process control;
   
9. Other risk mitigation activities as needed;
   
10. Reassessment and updating;
    
11. Roles, responsibilities, training, and qualifications;
    
12. Documentation and recordkeeping.

The expected outcome from an RBI process should be the association of risks with appropriate inspection, process control, or other risk mitigation activities to manage the risks. The RBI process is capable of generating

• A ranking of all equipment evaluated by relative risk.
  
• A detailed description of the inspection plan to be employed for each equipment item, including the following:
  
  
  
  • Inspection method(s) that should be used (e.g., visual inspection, ultrasonic, radiography, and smart pigging operations).
    
  • Extent of application of the inspection method (e.g., percent of total area examined or specific locations).
    
  • Timing of inspections/examinations (inspection intervals/due dates).
    
  • Risk management achieved through implementation of the inspection plan.
  
  
• Description of any other risk mitigation activities, such as repairs, replacements or safety equipment upgrades, and equipment redesign or maintenance.
  
• Expected risk levels of all equipment after the inspection plan and other risk mitigation activities have been implemented.
  
• Identification of risk drivers.

API RP 581 [2] was developed to provide users with a specific approach for developing, implementing, and maintaining a RBI program. The methodology outlined in API RP 581 [2] was developed through a joint industry project initiated in May 1993 and conducted by API to develop practical methods for RBI. The original sponsor group was organized and administered by API and included the following original members: Amoco, ARCO, Ashland, BP, Chevron, CITGO, Conoco, Dow Chemical, DNO Heather, DSM Chemicals, Exxon, Fina, Koch, Marathon, Mobil, Petro-Canada, Phillips, Saudi Aramco, Shell, Sun, Texaco, and UNOCAL. The results of this project were published in a RBI base resource document (BRD) to the sponsor group based on the development of quantitative and qualitative RBI methodologies.

The methodologies developed in the project continue to be improved and maintained in API RP 581 [2]. The target audience in the proposal for the API sponsor group project stated that the methods in it were “to be aimed at an inspection and engineering function audience.” The methodology was specifically not intended to “become a comprehensive reference on the technology of quantitative risk assessment (QRA).”

For failure rate estimations, the proposal promised “methodologies to modify generic equipment item failure rates” via “modification factors.” For consequence calculations, safety, monetary loss, and environmental impact were to be included. Existing algorithms in AIChE chemical process quantitative risk analysis (CPQRA) guidelines are “complex and are best suited for use in a computerized form” for safety evaluations. It was proposed that “for ease of use, that safety consequences be limited to the evaluation of burning pools of liquids, ignited high-velocity gas and liquid releases, explosions of vapor clouds, and toxic impacts.”

The result of the sponsor group project and subsequent activities has been the development of simplified methods for estimating failure rates and consequences of pressure boundary failures. The methods are aimed at persons who are not expert in QRA. Subsequent computer programs were developed to further ease the application of the BRD and now API RP 581 [2] methodologies.

The approach outlined in API RP 581 [2] has been successfully implemented in major refineries, petrochemical/chemical plants, and upstream gas plants. In recent years, an increasing emphasis on pipeline process safety management and mechanical integrity using RBI has been considered for all upstream and pipeline applications following established and emerging OSHA, DOT, and API regulations and recommended practices.

67.2.2 Basic Risk Categories

Equipment or system failure and the subsequent release of hazardous materials can lead to many undesirable effects, including the unavailability of equipment. A RBI program considers these effects based on four basic risk categories:

• Flammable events can cause damage in two ways: thermal radiation and blast overpressure. Most of the damage from thermal effects tends to occur at close range, but blast effects can cause damage over a larger distance from the blast center.
  
• Toxic releases are addressed when they affect personnel. Acute, as opposed to chronic, exposure is considered. These releases can cause effects at greater distances than flammable events. Unlike flammable releases, toxic releases do not require an additional event (e.g., ignition, as in the case of flammables) to cause an undesirable event.
  
• Environmental risks are an important component to any consideration of overall risk in a processing plant. An RBI program focuses on acute environmental risks rather than chronic risks from low-level emissions. Environmental damage can occur with the release of many materials; however, the predominant environmental risk comes from the release of large amounts of liquid hydrocarbons.
  
• Business interruption can often exceed the costs of equipment and environmental damage and, therefore, should be accounted for in an RBI analysis. Equipment replacement costs (accounted for in flammable damage estimates) are often low when compared to the business loss associated with critical equipment for an extended period of time.

The risk calculation is performed for each scenario, for all four risk categories, if desired. The risk for each equipment item is then found by summing the individual risk components from each scenario (hole size) calculation.

67.2.3 Alternative RBI Approaches

RBI begins with the identification of high-risk assets followed by the assessment of equipment condition, evaluation of maintenance and inspection programs, study of operating protocols, and estimation of life consumption of these priority assets. This process takes into account the probability and consequences of mechanical failure. The information is used to modify and optimize inspection and maintenance programs, audit procedures, operating limits, and safety information.

The POF due to loss of containment should consider each of potential degradation mechanism that could lead to loss of containment. POF (loss of containment) can be estimated quantitatively, if any degradation model exists, or qualitatively by means of engineering judgment.

Given that loss of containment has occurred, the potential consequences depend on the size of the hole, which can vary from a pinhole to a full-bore rupture. The methodology’s handling of hole size probability distribution by degradation mechanism should be understood when considering the potential consequences of any loss of containment, since the approach used can significantly affect the outcome. Finally, the probability that loss of containment can lead to ignition or pollution is an important issue.

RBI can be carried out using methods that are qualitative or more quantitative. In practice, most RBI efforts are carried out using a blend of both the methods, referred to as semiquantitative, and is often represented as a continuum as shown in Figure 67.3. The qualitative assessments can be conducted at a system or equipment level, while quantitative assessments are conducted on the equipment level, as follows:

1. RBI of systems assessments is qualitative. The system risk ranking allows for prioritization of groups of equipment for inspection.
   
2. RBI at the equipment level can be conducted qualitatively:
   
   
   
   1. Team work sessions are used to collect experience-based data.

      

      
      
   2. Team work sessions are used to collect experience-based data degradation mechanisms assessment, COF assessment, POF assessments, and risk and inspection planning.
      
      
      
      • Each team work session follows a standard format using guidance forms and question sets to assure consistency between studies.
        
      • Degradation mechanisms considered in each assessment, at a minimum, consider internal, external degradation mechanisms, and mechanical damage.
        
      • Team work sessions include qualified subject matter experts (>10 years’ experience), including an RBI expert.
        
      • POF and COF assessments should use only qualified specialists to assign numerical values/ ranges based on engineering judgment.
      
      
   3. Risk matrices with decision procedures are used for inspection planning.
   
   
3. RBI at the equipment level can be conducted quantitatively:
   
   
   
   1. Analysis is supported and driven by equipment-specific data collected for the assessment.
      
   2. POF determinations are model based, where values vary widely from <10−5 to 1 with results often presented on a logarithmic scale. POF values <10−5 are difficult to model, observe, and represent an insignificant risk.
      
      
      
      • Degradation mechanisms considered in each assessment, at a minimum, consider internal, external degradation mechanisms, and mechanical damage.
        
      • RBI team personnel should include qualified personnel knowledgeable in process operations, materials/corrosion, and inspection as well as an RBI expert.
      
      
   3. COF are model based and are expressed in terms of damage to equipment and potential personnel injury.
      
   4. Environmental COF are model based and are expressed in terms of mass or volume of a pollutant released to the environment or in financial terms as the cost of cleaning up the spill, including consideration of fines and other compensation.
      
   5. Financial COF is expressed in financial terms and is often a combination of other consequence effects for equipment, personnel, environment, and business interruption.

Regardless of the level of assessment conducted, the team should be comprising specialists with considerable experience with the specific installation(s) being assessed and include relevant expertise in the RBI methodology being used. Since qualitative assessments are based on rankings and numerical assignment during experienced-based team sessions, a higher level of experience and specialist involvement is required for a credible assessment. In addition, the documentation of data and any assumptions that were the basis for the assessment should be documented sufficiently for the study to be reproduced at a later date, even if the original team members are not available to participate. Conversely, RBI assessments based on actual design, operating, and inspection data require less documentation and assumptions, making future reassessments more reproducible.

67.2.4 Qualitative Approaches to RBI

Qualitative approaches are similar to quantitative analysis but require less equipment-specific data and can take less time to complete. While the results yielded are not as precise as those of the quantitative analysis, this approach provides a basis for prioritizing a RBI inspection program. However, as previously mentioned, a higher level of experience and specialist involvement is required for a credible qualitative assessment. The basis for all numerical assignments or rankings should be thoroughly documented as well as any assumptions that were the basis for the assessment in the event that the original team members are not available to participate.

Qualitative models can be interpreted as an expert judgment-based approach in which a numerical value is not calculated or assigned. A descriptive ranking is given, such as low, medium, or high, or a numerical ranking such as 1, 2, or 3. Qualitative RBI approaches cover a wide spectrum of methodologies. The simplest approach would be to compile the data/information required to do a quality damage review for the POF and categorize the hazards for COF to generate a risk matrix location to recommend a relevant inspection frequency.

The advantage of using a qualitative approach is that the assessment can be completed quickly and at a lower initial cost; there is little need for detailed information and the results are easily presented and understood. A disadvantage is that the results are more subjective than quantitative analysis because they are based on the opinions and experience of the RBI team and are thus harder to update the following inspection. Qualitative models provide a risk ranking of items since there is no change in risk over time and setting an acceptable risk target is not possible. As a result, an inspection interval is assigned based on position in a risk matrix that only changes if an input value, and therefore, matrix location changes.

Methodologies that would be considered qualitative generally have the following characteristics:

• Parts of the RBI assessment use category assignments and ranking methodologies. For example
  
  
  
  1. POF assessment is quantitative and COF assessment is qualitative (and vice versa).
     
  2. COF and POF assessments are quantitative, but risk ranking and time to inspection assessment are qualitative (i.e., do not change over time).
  
  
• Assignment of POF and/or COF categories is done by simple algorithms based on a selected set of most relevant parameters.
  
• Assignments of POF and COF are made by engineering judgment.

A qualitative analysis of a system can be conducted to quickly prioritize groups of equipment for more detailed risk analysis. The result of any qualitative analysis positions equipment within a risk matrix and provides a method for risk rank prioritization. This level of qualitative analysis is performed at any of the following levels:

• Operating plant: for example, oil/gas separation plant processing unit;
  
• Major area or functional section within a plant: for example, amine section of a separation unit;
  
• System: a major piece of equipment and associated equipment—for example, an amine absorber column, including the feed preheat, overhead, and reflux and reboiler equipment;
  
• Equipment or component: a major piece of equipment and associated equipment—for example, an amine absorber column or column top.

Note that the aforementioned four levels are most representative of an oil/gas separation plant. Pipeline applications would generally only apply to the system and equipment levels.

Qualitative approaches conducted at the system level are generally influenced by the number of equipment items in the system or equipment being studied because the quantity and type of inventory is considered to assign COF. As a result, assessments conducted at the system level should be based on similar equipment counts when used to compare systems for prioritization purposes. Equipment level qualitative assessments do not require any special considerations for risk prioritization purposes.

Qualitative RBI procedures have the following three functions:

1. Screening the system/equipment to select the level of analysis needed and to ascertain the benefit of further analyses (quantitative RBI or another technique).
   
2. Rating the degree of risk for system/equipment and assigning them a position on a risk matrix.
   
3. Identifying areas of potential concern for the system/equipment that requires enhanced inspection programs.

The qualitative analysis determines a risk ranking for systems or equipment by categorizing the two elements of risk: POF and COF. The process involved and the physical boundaries of the study area must be defined before the qualitative analysis is conducted. The following sections provide an overview of the factors that should be considered in any qualitative analysis.

67.2.5 Quantitative RBI Analysis

Quantitative models are model-based approaches where numerical values are calculated. The advantage of a quantitative approach is that the results can be used to calculate with some precision when the risk acceptance limit is reached. Quantitative methods are systematic, consistent, documented, and are relatively easy to update with inspection results. A quantitative approach generally uses a program or software to calculate risk and develop inspection program recommendations. The models are initially data-intensive but use of models removes repetitive, detailed work from the traditional inspection planning process. The types of data required for quantitative RBI analyses are represented in the flow diagram shown in Figure 67.4. API RP 581 [2] outlines a thoroughly documented quantitative approach widely used and generally accepted in the oil and gas and petrochemical industries. API RP 581 [2] is equally applicable to offshore, midstream, and pipeline industries if a quantitative methodology is desired and supported by the data requirements. The following description of quantitative RBI analysis generally follows the methodology outlined in API RP 581 [2].

67.2.5.1 Probability of Failure Analysis

The POF analysis is based on using industry generic failure frequencies (G_ff) by equipment type that is modified by two factors, reflecting identifiable differences from a generic value to an equipment-specific probability of the equipment/component being studied. A calculated equipment/component equipment factor (F_Equip) reflects the specific operation, potential in-service damage mechanisms, and condition based on inspection. The F_Equip is a combination of a DF modified by an inspection effectiveness factor (IF). The DF is a factor calculated based on the potential in-service damage mechanisms for the equipment. The inspection IF modifies the DF based on the history of inspection, the effectiveness of those inspections, and the result/findings. The management systems factor (F_MS) is based on an evaluation of the facility’s management practices that affect the mechanical integrity of the equipment. The management system evaluation tool was based on API 750 guidelines and is included as a workbook of audit questions in the API RP 581 [2].

The adjusted failure frequency or POF is calculated by multiplying the G_ff by the two modification factors, as shown in the following equation:

The database of G_ff was based on a compilation of available equipment failure histories from multiple industries. From these data, G_ff were developed for each type of equipment and pipe diameter.

The DF evaluates the specific operating environment of the equipment/component and determines a factor unique to the equipment operating history and inspection condition. DF calculation includes a series of damage modules that assess the effect of specific failure mechanisms on the POF. The DF for thinning is determined by calculating an ar/t factor, where

• a = age
  
• r = corrosion rate
  
• t = furnished thickness

The ar/t factor is defined as the ratio of wall loss (years in service age, a; times damage rate, r) and furnished thickness, t. The ar/t factor is used with inspection effectiveness to determine a thinning DF using the table in API RP 581 [2] Part 2, Table 5.11. A similar table is provided for cracking mechanisms that can be found in API RP 581 [2] Part 2, Table 7.4. The damage modules serve the following functions:

• Screen the operation to identify the active damage mechanisms.
  
• Establish a damage rate in the environment and DF associated with damage, DF.
  
• Quantify the effectiveness of the inspection program, IF.
  
• Calculate the F_equip modification factor to apply to the generic failure frequency.

The F_MS adjusts for the influence of the facility’s process safety management system on the mechanical integrity of the plant. The outline of the Management Systems Workbook from API RP 581 [2] is given in Table 67.1. The complete audit of the workbook can be found in API RP 581 [2] Part 2, Table 4.4. The F_MS adjustment is applied equally to all equipment/component items in the study. The F_MS provides discrimination between risk of equipment/components under different management practices or at different locations and management systems. However, the evaluation process can be used to improve the effectiveness of any PSM program, thereby reducing overall risk.

67.3.1 Introduction

High-risk assets have the potential to cause serious disruption in operations or cause unplanned and extended production outages. The priority treatment of high-risk assets must be balanced by providing sufficient maintenance and inspection for other lower-risk equipment to avoid the unintended increased risk over time. The objective is to optimize the use of pipeline maintenance and inspection resources for all equipment/components over time to manage risk at the desired levels.

In any system or process, a relatively large percentage of risk in an operating plant is associated with a small percentage of the equipment items. RBI shifts inspection and maintenance resources to provide a higher level of coverage on the high-risk items and an appropriate effort on lower risk equipment, by systematically combining both POF and COF to establish a prioritized list of equipment/components based on total risk. From the prioritized list, an inspection program may be designed that manages (reduces or maintains) the risk of equipment failure. The potential benefit of an RBI program is to improve the operating efficiencies of systems or processes, while improving, or at least maintaining, the same level of risk.

Figures 67.5–67.7 show the impact of an RBI assessment on the inspection program. Figure 67.5 presents a comparison of the risk reduction relative to increased inspection activity associated with a typical, non-RBI program, and risk using RBI. Both plans will decrease risk as the inspection activity increases to a point where the rate of risk reduction with additional activity is low. However, the RBI-based program achieves an overall lower level of risk with the same inspection as a result of optimization of resources based on risk. It should be noted that a certain level of residual risk is inherent in any process involving hazardous materials and additional inspection will not reduce the risk below that residual risk.

Figure 67.6 shows how refocusing inspection activities on the highest risk equipment achieve a lower risk. The lower risk is achieved in this case without additional costs to reduce the risk. Figure 67.7 demonstrates how unnecessary or low-value inspection activities can be reduced while maintaining the same risk level when inspection priorities are based on risk.

Risk reduction or risk optimization is achieved several ways, including the following:

• Change the process, operations, and/or inventories.
  
• Replace equipment with other equipment that is desirable for the intended service.
  
• Improve management systems, training, workloads, and so on.
  
• Improve inspection programs.

RBI focuses on risk reduction through inspection planning by the development of inspection programs that address the types of damage expected and the inspection methods appropriate to detect the expected damage.

Inspection influences risk by reducing the POF through improved knowledge of the equipment condition. Many conditions (design errors, fabrication flaws, and malfunction of control devices) can lead to equipment failure, but in-service inspection is primarily concerned with the detection of progressive damage. The POF due to such damage is a function of the following factors:

• Damage mechanism and resulting type of damage (cracking, thinning, mechanical damage, etc.);
  
• Damage rate;
  
• Probability of detecting damage and predicting future damage states with inspection technique(s);
  
• Tolerance of the equipment to damage type.

RBI differs from conventional inspection management by providing the methods to support decision-making when the data are missing or uncertain. Rational decisions can be made using RBI methodologies and the application of decision-making applies risk reduction through inspection.

67.3.2 Inspection Techniques and Effectiveness

The purpose of an inspection program is to define and perform the activities necessary to detect in-service deterioration of equipment before failures occur. An effective inspection program is developed by systematically identifying

• What types of damage are possible?
  
• What equipment is susceptible and what locations to look for damage?
  
• What inspection methods are effective to detect and measure damage?
  
• What frequency of inspection is required to properly detect and measure the damage?

Equipment data that must be available to properly assess the risk include equipment design and construction, the operating process conditions, and the maintenance and inspection history. The following basic data are required to identify damage mechanisms of interest:

Design and construction data

• Equipment type (heat, mass, or momentum transfer) and function (shell and tube exchanger, trayed distillation column, centrifugal pump, etc.);
  
• Material of construction (ASTM specification, grade, and year);
  
• Heat treatment;
  
• Furnished thickness.

Operating process data

• Temperature
  
• Pressure
  
• Liquid level, %
  
• Corrosive components of process (such as H₂S, CO₂, amine, water, and organic chlorides)
  
• Flow rate
  
• Solids contents
  
• Equipment history:
  
• Previous inspection data
  
• Failure analysis
  
• Maintenance activity
  
• Replacement/repair information
  
• Modifications

67.3.3 Damage Types

Damage types are the physical characteristics of damage that can be detected by an inspection technique. Damage mechanisms are the corrosion or mechanical actions that produce the damage. Table 67.2 describes some familiar damage types and their characteristics. Tables 67.3–67.6 illustrate the damage mechanisms by broad categories and the types of damage associated with the mechanisms.

Damage may occur uniformly throughout a piece of equipment, or it may occur locally, depending on the mechanism at work. Uniformly occurring damage can be inspected and evaluated at any convenient location, since the results can be expected to be representative of the overall condition. Damage that occurs locally requires a more focused inspection effort. This may involve inspection of a larger area to ensure that localized damage is detected. If the damage mechanism is sufficiently well understood to allow prediction of the locations where damage will occur, the inspection effort can focus on those areas.

67.3.3.2 Inspection Frequency/Interval

Inspection frequency/interval is determined by combining the factors of RBI presented in previous sections:

1. Damage mechanism and resulting type of damage (cracking, thinning, etc.);
   
2. Rate of damage progression;
   
3. Tolerance of the equipment to the type of damage;
   
4. Probability of detecting damage and predicting future damage states with inspection technique(s).

Traditional inspection programs perform inspection at recommended frequencies, often selected based on some fraction of the equipment’s RL defined as:

Inspection planning using risk of loss of containment provides a tool to conduct inspection based on risk targets rather than fixed frequencies. The result is inspection intervals that vary through the life of the equipment depending on the in-service damage potential and observed rate of damage. A risk criteria defining acceptable risk based on prudent levels of safety, environmental, and financial risks must be developed for RBI decision-making. Risk acceptance may vary for different risk exposures. For example, risk tolerance for a financial risk may be higher than for a safety/health risk.

67.3.4 Probability of Detection

Inspection techniques vary in their accuracy, depending on operator skill and test conditions. Accuracy can be measured by repeating the examinations of known flaws of various sizes and recording the results. The application of these data to in-service plant inspections is somewhat limited since they are based on examination of prepared test blocks in a comfortable laboratory environment. However, they provide two very important pieces of information:

1. They establish that there is a probability of detection (POD). Even under controlled conditions, nondestructive testing has limitations and reveals an increased POD for flaws as the size of the flaws increase.
   
2. They establish the basis for the maximum effectiveness that can be expected from an inspection. “Real-world” probabilities of detection may approach this effectiveness but could not be expected to exceed it.

Several organizations have tried to quantify the POD by performing round-robin type tests:

• Nordtest in Europe (ultrasonics [UT], MPI [MT], and radiography [RT]);
  
• PISC (Italy);
  
• EPRI (USA);
  
• CIPS in the United States (UT-nuclear applications);
  
• Nippon Steel Japan (MPI);
  
• US Navy (UT and radiography submarines).

As discussed in the previous section on inspection effectiveness, POD data can be used in a quantitative assessment of inspection effectiveness through extreme value analysis, if sufficiently detailed information on the distribution of damage states is also available.

The POD data, where available, are also useful in helping assign the inspection methods to the appropriate inspection effectiveness category.

67.3.5 Reducing Risk through Inspection

Presenting the RBI results in a risk matrix is a very effective way of communicating the distribution of risks throughout a system or process without showing numerical values. An example of risk matrix is shown in Figure 67.1, different sizes of matrices may be used (e.g., 5 × 5 and 4 × 4). Risk categories may be assigned to the boxes on the risk matrix. The example in Figure 67.1 shows risk categorization of high, medium-high, medium, and low. A risk matrix depicts risk results at one particular point in time.

Another method of showing risk results for quantitative methodologies and when showing numeric risk values is meaningful, an iso-risk plot is used, as shown in Figure 67.2. An iso-risk plot shows POF and COF for equipment plotted on a graph with an iso-risk line (line of constant risk). If this line is the acceptable level of risk, the equipment above the line fail the established risk acceptance criteria and should be mitigated to a value below the line.

The most common presentation of the RBI results is to rank the equipment/components by risk (from high to low) in a tabular format.

Risk matrix and iso-risk plots graphically show that the equipment/components toward the upper right-hand corner of the matrix or plot have a higher priority for inspection planning because these items have the highest risk. Similarly, items residing toward the lower left-hand corner of the matrix or plot will have a lower priority. The risk matrix or plot can be used as a screening tool during the inspection prioritization process.

Established risk thresholds that divide the risk matrix, plot, or prioritized risk table into acceptable and unacceptable regions of risk can be developed using corporate safety and financial policies. Inspection is conducted on or before the date that equipment reaches the risk target. As damage progresses, inspection is performed to increase confidence in the current and expected future condition of equipment. Increased confidence reduces risk but never as low as the original risk level due to the damaged state. When inspection no longer effectively lowers the risk below the risk target, other risk mitigation activities are required to manage the risk such as FFS evaluations or plans for the equipment repair/replacement.

Regulations and laws may also specify or assist in identifying the acceptable risk thresholds. Reduction of some risks to a lower level may not be practical due to technology and cost constraints. An ALARP approach to risk management or other risk management approach may be necessary for these items.

Obviously, inspection does not mitigate damage mechanisms or in and of itself does not reduce risk. However, the knowledge of the equipment/component condition gained through effective inspection better quantifies the actual risk. Inspection does not prevent an impending failure unless the inspection initiates a risk mitigation activity that changes rate of damage and POF. Inspection serves to identify, monitor, and measure the damage rate to more accurately calculate the remaining useful life of equipment. Correct application of inspection methods improves the ability to predict the future condition of the equipment. The better the predictability, the less uncertainty there will be as to when a failure may occur. Risk mitigation activities (repair, replacement, changes, etc.) are planned and implemented prior to the predicted failure date. The reduction in uncertainty and increase in predictability through inspection translate directly into a better estimate of the POF and, therefore, reduce risk. Figure 67.8 shows the mitigation of risk over time that is possible by using a well-planned inspection program (Figure 67.9).

Inspection is not the only risk mitigation activity reducing risks associated. Some damage mechanisms are very difficult or impossible to monitor with inspection activities (e.g., metallurgical deterioration that may result in brittle fracture, some types of stress corrosion cracking [SCC] and even fatigue). Other damage caused by mechanical damage is event-driven and are not impacted by inspection.

Risk mitigation (by the reduction in uncertainty) achieved through inspection presumes that the organization will act on the results of the inspection in a timely manner. Risk mitigation is not achieved if gathered inspection data are not properly analyzed and acted upon. The quality of the inspection data and the analysis or interpretation significantly affects the level of risk mitigation. Proper inspection methods and data analysis tools are therefore critical.

In any system or process facility, a relatively large percentage of risk in an operating plant is associated with a small percentage of the equipment items. RBI permits the shift of inspection and maintenance resources to provide a higher level of coverage on the high-risk items and an appropriate effort on lower risk equipment, by systematically combining both likelihood and COF to establish a prioritized list of pressure equipment based on total risk. From that list, users can design an inspection program that manages (reduces or maintains) the risk of equipment failure. The potential benefit of a RBI program is to increase operating times and run lengths of process facilities, while improving, or at least maintaining, the same level of risk.

67.4.1 A Continuum of Approach

Traditional inspection codes give prescriptive or time-based requirements for inspection recommendations, focusing on the POF over time, that is, inspection would be in a defined number of years based on estimated RL or some other criteria. Alternatively, risk-based methodology considers both the probability and COF to plan inspections, providing a basis for making informed decisions on inspection frequency and the extent of inspection.

A qualitative RBI approach uses data analysis as well as expert local knowledge and judgment to prioritize inspection, while quantitative analysis is based on a probabilistic or statistical model that calculates risk over time as the potential of in-service damage increases. Quantitative RBI requires large amounts of up-to-date and accurate data that are often not available; on the other hand, qualitative RBI analysis requires much less data to implement. Because it relies on less data and a simpler assessment methodology, qualitative RBI provides a more conservative analysis with greater inspection frequency. However, a qualitative program is fully compliant with API 580, the recommended practice for implementing RBI for the fixed equipment and pipeline industry and can prepare the operator for a more in-depth quantitative analysis.

In most facilities, a large percent of the total risk is concentrated in a relatively small percent of the equipment, and these potential high-risk areas may require greater attention through a revised inspection plan. A qualitative program can help identify areas of higher risk and potentially destabilizing conditions for which a more focused quantitative approach is necessary. Thus, both qualitative and quantitative approaches can be used in a complimentary way to develop an efficient and reliable inspection program that continues over the life of the equipment. A qualitative program can be the first step toward a semiquantitative or quantitative program as more detailed technical data become available.

67.4.2 Qualitative Versus Quantitative Examples

These are the key elements to every RBI program, whether the approach is qualitative or quantitative:

1. Management system for maintaining documentation, history, and updated analysis;
   
2. Documented methodology for POF;
   
3. Documented methodology for COF;
   
4. Risk ranking for all systems/equipment;
   
5. Detailed inspection plan and other mitigating activities.

In the following examples, a low-pressure separation drum in sour gas service is assessed using both qualitative and qualitative methodologies to illustrate the difference in results between the two approaches.

Background: A low-pressure separation drum is in sour gas service

67.4.3 Qualitative Example

A simple qualitative example will be used to demonstrate the use of RBI. No RBI was performed until 2005 and the first in-service inspection was scheduled for 2007.

67.4.3.3 Risk Results

The risk results for the qualitative example are summarized in the following table:

Analysis	Basis	Results COF Category
COF	Flammable	4
	Health	3
	Final	4 POF category
POF	Thinning, 10 pmy	4
	Thinning, 5 mpy	2
	Cracking	3 Risk category
Risk	Thinning, 10 mpy	High
	Thinning, 5 mpy	Medium-high
	Cracking	High frequency
Inspection planning	Thinning, 10 mpy	2
	Thinning, 5 mpy	6
	Thinning, 5 mpy RL	0.3
	Cracking	4

API Fluid Name	Applicable Materials	COF, Flammable	COF, Toxic
C1–C2	Methane, ethane, ethylene, LNG, fuel gas	4	0
C3–C4	Propane, butane, isobutane, LPG	4	0
C5	Pentane	3	0
C6–C8	Gasoline, naphtha, LSR, heptane	3	0
C9–C12	Jet fuel. Light kerosene	3	0
C13–C16	Diesel, heavy kerosene, Atm gas oil	3	0
C17–C25	Gas oil, crude	2	0
C25+	Residuum, heavy crude	2	0
Water	Water	l	0
Steam	Steam	l	0
Acid-LP	Low pressure acid/caustic	3	0
Acid-MP	Medium pressure acid/caustic	3	0
Acid-HP	High pressure acid/caustic	3	0
H2	Hydrogen only	3	0
H2S	Hydrogen sulfide	3	4
HF	Hydrofluoric acid	0	4
Aromatics	Benzene, toluene, xylene, cumene	3	0
Ammonia	Ammonia	0	4
HCl	Hydrochloric acid	0	5

Notes:

• H₂S toxic category based on >3% in process stream. Reduce COF, toxic one category if ≤3%.
  
• HF toxic category based on >5% in process stream. Reduce COF, toxic one category if ≤5%.
  
• C₁–C₂, C₃–C₄ pressurized liquids that rapidly vaporize with atmospheric boiling temperatures below 50 °F or where the atmospheric boiling point is below the operating temperature can lead to brittle failures.
  
• C₅ and C₆–C₈ fluids should be classified as COF 4 when the operating temperature is above the atmospheric boiling point.
  
• Flammable services operating above their autoignition temperature should be assigned COF, flammable category 4.

Since the COF does not change over time in this example, the largest of the flammable and health COF category is used to set the inspection frequency for thinning and cracking mechanisms. If process conditions change, modifications can easily be made to the model to reflect those changes. Updating the expected corrosion rate of 10 mpy with an in-service measured rate reduced the POF category from 4 to 2 and results in a reduction from high to medium-high.

67.4.3.4 Inspection Planning

Using Figure 67.10, the change in risk category increases the inspection interval from 2 years to 6 for thinning. The recommended interval for cracking is 4 years using Figure 67.11.

Because qualitative RBI is not quantitative enough for POF to change over time, the inspection frequency is assumed to be constant through the equipment useful life. This assumption is conservative at the start of service but becomes less conservative as the equipment reaches the end of life. As a result, the qualitative results must be based on more conservative frequencies.

Using the RL interval, adjustment factor approach (Figure 67.12) adjusts the inspection interval as the equipment approaches end of life. This method requires a RL calculation based on the excess wall thickness (furnished thickness—minimum required thickness) after each inspection. The next inspection is based on the RL × inspection factor from Figure 67.14. In this example, the measured corrosion rate is 5 mpy with a 0.090 in. corrosion allowance at the start of service or 18 year expected life. The first inspection would be expected to occur by using the RL × 0.3 or 18 × 0.3 = 5.4 years. However, after 10 years of service with 5 mpy all loss, the RL is 8 years. Therefore, the inspection interval at year 10 is RL × 0.3, since there is no change in risk over time. But by using the RL adjustment factor approach, the new interval is 8 × 0.3 = 2.4 years. Applying the RL method for setting RBI frequency/intervals results in setting more realistic intervals over the equipment life, that is, longer intervals at the start of service and shorter intervals as the equipment approaches its end of life.

The results for thinning and cracking are shown on a risk matrix in Figures 67.13 and 67.14. Implemented correctly, a qualitative RBI study logically leads to a more conservative analysis and more inspection than a quantitative RBI analysis would suggest.

67.4.4 Quantitative Example

The aforementioned example will be studied using a quantitative approach to demonstrate the use of quantitative RBI. No RBI was performed until 2005 and the first in-service inspection was scheduled for 2006.

67.4.4.1 POF Calculation

The drum operated for 6 years with no inspection. The expected corrosion rate was 10 mpy and applied over the first 6 years of service with a starting thickness 0.375 in. The ar/t factor was calculated and DF determined using the inspection history. For demonstration purposes, the ar/t and DF’s were calculated in the table before and after each B level thinning inspection assuming an inspection is performed at a 3-year frequency after the first 6 years of service.

Age	Corrosion Rate	(ar/t)	DF, No Inspection	No. of B Inspections	DF	Thinning POF Category
5	0.01	0.13	90	0	90	3
5	0.005	0.07	1	1	1	1
9	0.005	0.12	6	1	2	1
9	0.005	0.12	6	2	1	1
13	0.005	0.17	170	2	7	2
13	0.005	0.17	170	3	2	1
17	0.005	0.23	400	3	6	2
17	0.005	0.23	400	4	2	1

Similarly, the cracking DF and POF must be calculated. While H₂S was expected at fairly low concentrations and the vessel was not PWHT’d during fabrication, no inspection for H₂S blistering or cracking was planned. Wet H₂S damage was identified as a potential damage mechanism during the RBI damage review, so no inspection had been identified prior to the study. A susceptibility of medium was assigned for both blistering and cracking. A medium susceptibility has a severity index, SI, of 10. The severity index is used to determine a cracking DF with the inspection history. The DF is used to calculate the DF_final using the following equation:

Age	Susceptibility	No. of C Inspection	DF	Cracking POF Category
5	Medium	0	58.75	3
9	Medium	0	112.1	1
13	Medium	0	168.0	1
17	Medium	1	1	2
17	Medium	1	33.50	1

The final POF is the sum of the individual DFs combined with the G_ff and F_MS as shown in the following table:

Year	Age	Thinning DF	Cracking DF	Combined DF	FMS	Gff	POF	POF Category
2001	5	90	58.73	167.73	1	1.56E−05	2.32E−03	4
2006	5	1	58.73	59.73	1	1.56E−05	9.32E−04	3
2010	9	2	112.12	114.12	1	1.56E−05	1.78E−03	4
2010	9	1	112.12	113.12	1	1.56E−05	1.76E−03	4
2013	13	7	168.01	175.01	1	1.56E−05	2.73E−03	4
2013	13	2	168.01	170.01	1	1.56E−05	2.65E−03	4
2017	17	6	225.68	231.68	1	1.56E−05	3.61E−03	4
2017	17	2	1.00	3.00	1	1.56E−05	4.68E−05	1

Table 67.13 is used to determine the POF category from the combined DFs.

67.4.4.3 Risk Results

The risk results for the quantitative example are summarized in the following table:

Analysis	Basis	Results
		COF Area	COF Category
COF	Flammable, equipment	943.66	B
	Flammable, injury	2,443.84	C
	Toxic	226.84	B
	Final	2443.84	C
		DF/Risk	POF Category
POF at 2017	Thinning	6	2
	Cracking	225.68	4
	Total	231.68	4
	POF	3.61E−03	4
Risk	Total	Risk Value, ft2/year	Risk Category 4C—high
		8.83
Inspection Planning risk after inspection		Thinning—1B	2	1
		Cracking—1C	1	1
		Total DF	3	2
		POF	4.68E−05	2
Risk		Risk Value, ft2/year	Risk Category
		1.14E−01	2C—low

Since the COF does not change over time unless the process conditions are changed significantly, the largest of the flammable (equipment and injury) and toxic COF consequence areas are used to set the inspection frequency active damage mechanisms. If process conditions change, modifications can easily be made to the model to reflect those changes.

67.4.5 Optimizing the Inspection Program

The aforementioned examples demonstrate the use of RBI to evaluate options for an inspection program and show how inspection planning can be optimized using the following:

• Increasing activity level or frequency if insufficient reduction in risk occurs, or decreasing activity level or frequency if no gain in risk reduction results from the higher level of inspections.
  
• High DFs may result when an RBI program is initially implemented. Equipment showing high DF should be prioritized for inspection based on their risk and then DF/POF.
  
• Equipment with a history of multiple inspections and confirmed low-corrosion rates are candidates for reduced inspection activity or increased intervals.
  
• Equipment with large uncertainty in the corrosion rate require shorter intervals and possibly more thorough inspections to maintain the target risk, at least until sufficient history has been established.
  
• Equipment approaching end of life due to corrosion or other deterioration requires increased inspection activity to be sure that the actual deterioration rates do not exceed expected rates. Additional inspection will not significantly reduce the DF or risk once the RL has been consumed.

67.4.6 Example Problem Conclusions

The purpose of this example is to show both qualitative and quantitative RBI approaches. A wide spectrum of RBI can be used, as discussed in Section 67.2.3 and shown in Figure 67.3. In addition, the example can be used to demonstrate the fact that a more qualitative approach is more conservative than a quantitative one. While qualitative approaches require less time and cost to implement, they generally recommend more equipment inspection as a result of that conservativeness. Conversely, quantitative approaches require more time and data to implement. However, updating risk after implementation of an inspection plan and re-establishing the inspection requirements can require less time and cost to maintain.

67.5.1 Summary

Government regulators are increasingly demanding that risk assessment techniques be applied to high-consequence areas. International standards organizations such as ISO, NORSOK, OGP, and API have each developed international design codes and standards that should be considered when implementing a risk management program for pipelines.

RBI is a structured methodology that establishes an inspection strategy over a defined period of time using detailed technical data. Based on an analysis of the probability and COF, RBI focuses inspection planning on high-risk items and generates the necessary data to determine inspection frequency and future priorities.

A RBI program should be considered for the following pipeline assets:

• Trunk lines from manifold stations to process facilities;
  
• Oil and gas NGL pipelines;
  
• Manifold stations;
  
• Process facilities for oil and gas separation.

The quality of available information on a facility’s assets determines whether a qualitative or quantitative RBI approach should be used. The approach adopted will depend on the following:

• Design and construction data readily available.
  
• Known equipment condition based on inspection history.
  
• Knowledge of past, current, and future process operational conditions.
  
• Complexity and age of systems considered.

Less data generally require more qualitative approach, at least initially.

67.5.2 Ten Criteria for Selecting the Most Appropriate Level of RBI

RBI methodology can be used on older facilities that have little or poor data available; on newer assets that have technical data available, but little inspection history; and on facilities with good quality data and inspection history. This recommends that the following criteria be used to determine whether a qualitative–quantitative should initially be used:

1. Maturity of the mechanical integrity and process safety management programs: RBI requires data for equipment design (U-1 or piping specification data), process operating conditions (temperature, pressure, material balance, including hydrocarbon, nonhydrocarbon, toxic compositions, and liquid/gas phase information), and inspection history (inspection locations, methods, and coverage) and inspection results such as measured thickness at a specific date.
   
2. Implementation of mechanical integrity program and process safety management program initiatives: as discussed earlier, a mature mechanical integrity program may support a more quantitative approach to RBI. Conversely, in the early stages of mechanical integrity implementation, an emphasis on collected design, process, and inspection data may make a quantitative RBI approach unrealistic. In these cases, a quantitative RBI program limits the amount of design, process, and inspection data needed for a high-quality risk analysis. But possibly more importantly, using RBI at this stage could result in a cost-effective method to risk rank and prioritize systems for implementation of the mechanical integrity program. As more data becomes available, a more quantitative assessment can be added to the RBI program, depending on the other factors considered in this section.
   
3. Maturity of inspection program: a very mature program with all potentially active damage mechanisms properly identified can use RBI for optimizing inspection and manage/reduce risk. A program focused on optimizing risk generally results in increasing inspection for higher risk equipment and reducing inspection for lower risk equipment. The savings related to reduced inspection most often can be applied to the increased inspection areas with little negative impact on the overall cost. In the situation where little inspection has been performed, RBI may identify a much higher level of inspection required than that for a more mature program. This example highlights a situation where more inspection may be recommended when a damage mechanism (wet H₂S damage potential in sour service) had not been previously identified. A quantitative approach is not required to properly identify equipment in this case.
   
4. Corrosiveness of service: consideration of process conditions driving POF due to in-service damage needs to be considered. The key aspect of the POF is a credible corrosion/damage review performed by a knowledgeable specialist. Identify potential damage mechanisms for operation that include internal corrosion (thinning due to general or localized corrosion); external corrosion or cracking (atmospheric conditions and CUI potential as well possible cracking if austenitic materials are used); internal SCC (for sour applications with H₂S or H₂S/CO₂, amine treating); other damage mechanisms (brittle fracture considerations for processing noncondensing hydrocarbons or cold service equipment). In noncorrosive services, RBI can justify doing little inspection to maintain acceptable risk levels.
   
5. Process operation: considers the process conditions driving COF due to flammable and toxic consequences as well as business interruption. The key aspect of the COF is relative consequence ranking of equipment COF does not change over time unless the process conditions change considerably. In upstream and pipeline applications, the change in process stream composition over time, and therefore COF and risk over time, is possibly as important as the COF ranking between assets at any point in time.
   
6. Complexity of the process: reminding us that it is all about risk prioritization for inspection planning, and that absolute quantitative values are not required or even desired. Operators are looking for the best approach to properly and effectively risk rank equipment and provide the proper relative rankings. The less time and money spent on the risk analysis in order to get the desired result, the more that can be spent developing and improving the inspection program.
   
7. Personnel: resources available to implement and maintain the risk program
   
8. Systems and training: required over time to maintain the RBI program.
   
9. Range and number of systems and assets: what and how much equipment is being included in the RBI program. Usually piping, pressure vessels, heat exchangers, filters, externally fired heaters are included.
   
10. Cost of implementing and maintaining the RBI program versus value of program: if the inspection program is very mature, there may be less dollar value in investing a lot of time and money to optimize it. Conversely, if the inspection program is very immature (very little high-quality inspection has been done with consideration for in-service damage), implementation of an integrity program may require more extensive inspection initially regardless of risk. Short-term benefit may not justify big investment.

Regardless of the approach used, a continuous improvement process should be implemented:

• Conduct the initial RBI assessment using available data.
  
• Use inspection results to develop inspection requirements and priorities.
  
• Implement initial inspection and plan over defined multiyear time period.
  
• Update RBI assessment after each inspection.
  
• Update the inspection plan from the updated assessment.

As the RBI program develops and matures, a more quantitative analysis can be incorporated, as desired.

67.5.3 Justifying Costs

Operators can justify the cost of implementing RBI because an RBI program compliments and supports other company initiatives:

• Mechanical integrity program;
  
• Process safely management initiatives;
  
• Business initiatives;
  
• Specialized inspection programs;
  
• Positive material identification;
  
• Corrosion under insulation (CUI).

Mechanical integrity and process safely management initiatives can use RBI to help prioritize the work over time, precluding the onerous task of implementing these programs in short periods to avoid exposures.

RBI can be used to prioritize risk and perform the activities resulting in the largest risk reduction for the risk reduction expense. Similarly, RBI can be used to support any business initiatives or specialized inspection programs (such as wet H₂S inspection programs and CUI detection/improvements) in an efficient, cost-effective manner.