Security Licensing Basis INL/RPT-25-87822 Framework Development Regulatory Framework Modernization Performance-Based Physical Security SEPTEMBER 2025 Christopher Chwasz, Robby Christian, and Amanda Rynes Idaho National Laboratory Alan Evans Sandia National Laboratory INL/RPT-25-87822 Security Licensing Basis Framework Performance-Based Physical Security Christopher Chwasz, Robby Christian, and Amanda Rynes Idaho National Laboratory Alan Evans Sandia National Laboratory September 2025 Idaho National Laboratory Idaho Falls, Idaho 83415 http://www.inl.gov Prepared for the U.S. Department of Energy Office of Nuclear Energy Under DOE Idaho Operations Office Contract DE-AC07-05ID14517 Page intentionally left blank ABSTRACT This report summarizes a technology-inclusive and performance-based method to determine the physical security licensing basis for a commercial nuclear reactor under the proposed 10 CFR 73.100 for Part 53 licensees. The method focuses on identifying security functions, the contributing security systems and programs to meet those functions, and identifying security events that will provide the foundation for the security licensing basis. This includes a risk-informed performance-based defense-in-depth adequacy method. This method can also be employed to justify performance-based alternative measures to traditional security requirements found in 10 CFR 73.55. v Page intentionally left blank vi CONTENTS ABSTRACT.................................................................................................................................................. v ACRONYMS ................................................................................................................................................ x 1. INTRODUCTION............................................................................................................................... 1 2. BACKGROUND................................................................................................................................. 1 2.1. Physical Security Rulemakings ................................................................................................ 2 2.2. Performance-based Regulation ................................................................................................ 3 3. PERFORMANCE-BASED PHYSICAL SECURITY FRAMEWORK ............................................. 5 3.1. Defining the Regulatory Issue and Context ............................................................................. 5 3.2. Identifying the Security Functions ........................................................................................... 6 3.2.1. Physical Protection System ......................................................................................... 6 3.2.2. Detection ..................................................................................................................... 7 3.2.3. Delay ........................................................................................................................... 7 3.2.4. Response ..................................................................................................................... 8 3.3. Selecting Performance Parameters and Criteria ..................................................................... 10 3.3.1. Irreversible Release Timing ...................................................................................... 13 3.3.2. Site Security Timeline ............................................................................................... 15 3.3.3. Detection Probability and Confidence ...................................................................... 21 3.3.4. Probability of Effectiveness ...................................................................................... 22 3.4. Identifying the Security Margins ........................................................................................... 28 3.4.1. Compensatory Measures ........................................................................................... 28 3.4.2. Defense-in-Depth Analysis ....................................................................................... 29 3.4.3. Timeline Margin ....................................................................................................... 33 3.5. Formulating a Performance-Based Alternative ...................................................................... 34 3.5.1. Target Sets ................................................................................................................ 34 3.5.2. Design Physical Protection System Detection and Communication ......................... 34 3.5.3. Offsite Response Timeline ........................................................................................ 35 3.5.4. Licensing Basis Security Scenarios .......................................................................... 35 3.5.5. Validation .................................................................................................................. 35 3.5.6. Security Margin......................................................................................................... 40 3.5.7. Documentation .......................................................................................................... 40 4. CONCLUSION ................................................................................................................................. 40 5. REFERENCES.................................................................................................................................. 42 vii FIGURES Figure 1. NUREG/BR-0303 objectives hierarchy. ....................................................................................... 4 Figure 2. Detailed decomposition of means objectives. ............................................................................... 5 Figure 3. Functions of a PPS......................................................................................................................... 6 Figure 4. PPS functions................................................................................................................................. 7 Figure 5. Example of delay and adversary timelines. ................................................................................... 8 Figure 6. Use-of-force factors. ...................................................................................................................... 9 Figure 7. 10 CFR 73.100 objective hierarchy. ............................................................................................ 11 Figure 8. Adversary and site security function timeline. ............................................................................ 12 Figure 9. Irreversible Release Timing vs. Site Security Timeline. ............................................................. 14 Figure 10. PPS detection process. ............................................................................................................... 15 Figure 11. Example of running rates. .......................................................................................................... 16 Figure 12. Hypothetical sample of adversary attack tools. ......................................................................... 17 Figure 13. Adversary vs. PPS timeline. ...................................................................................................... 23 Figure 14. Hypothetical PPS response timeline.a ........................................................................................ 24 Figure 15. Sample CDP identification. ....................................................................................................... 25 Figure 16. NEI-18-04 Rev. 1 framework for establishing did adequacy. ................................................... 31 Figure 17. Security timeline highlighting the time margin. ........................................................................ 34 Figure 18. Process for system evaluations. ................................................................................................. 36 Figure 19. Screenshot of Scribe3D TTXs recorder tool. ............................................................................ 38 TABLES Table 1. Hypothetical breach delay times. .................................................................................................. 18 Table 2. Probabilities of detection for 90/95 tests. ..................................................................................... 22 Table 3. Sample adversary timeline with probabilities of detection. .......................................................... 24 Table 4. Hypothetical neutralization analysis: simple numerical table....................................................... 26 Table 5. Hypothetical probability of neutralization determinations.a ......................................................... 26 viii Page intentionally left blank ix ACRONYMS AC&D alarm communication and display ADVANCE Accelerating Deployment of Versatile, Advanced Nuclear for Clean Energy AIPT Adversary Interference Precluded Time ARAPT alarm response assessment performance test CAS Central Alarm Station CCTV closed-circuit television CDP critical detection point CFR Code of Federal Regulation DBT Design Basis Threat DEPO Design Evaluation and Process Outline DID Defense in depth DG Draft [Regulatory] Guide DOE U.S. Department of Energy DOE-ID U.S. Department of Energy-Idaho Operations Office ELSPT enhance limited-scope performance tests FOF force-on-force GUI Graphical User Interface IAEA International Atomic Energy Agency INL Idaho National Laboratory IDS intrusion detection system LMP Licensing Modernization Project LSPT limited-scope performance tests NEIMA Nuclear Energy Innovation and Modernization Act NEI Nuclear Energy Institute NRC Nuclear Regulatory Commission PPS physical protection system RTL response team lead SAS Secondary Alarm Station SMR small modular reactor SSC system, structure, and component TMS time motion studies TTX Tabletop exercise x Page intentionally left blank xi Security Licensing Basis Framework Development 1. INTRODUCTION The advent of advanced nuclear reactors presenting increased safety design and performance, coupled with advances in risk-informed and performance-based licensing has spurred the U.S. Nuclear Regulatory Commission (NRC) to propose a performance-based rulemaking for the licensing of commercial nuclear power plant security against the design basis threat (DBT) of radiological sabotage. Using NRC guidance on development of a performance-based regulation, a performance-based framework is presented here to meet the proposed rulemaking language found in Rulemaking NRC-2019- 0062, “10 CFR Part 53: Risk-Informed, Technology-Inclusive Regulatory Framework for Advanced Reactors” draft text for Title 10 of the Code of Federal Regulations (10 CFR) 73.100. Additionally, this method could be used to justify alternative measures and exemptions from the existing 10 CFR 73.55 and may assist in meeting the proposed rule language and interpretations found in NRC-2017-0227, “Alternative Physical Security Requirements for Advanced Reactor” and NRC-2024-0167, “Law Enforcement Response in Power Reactor Physical Protection Programs.” This report presents a method to develop and maintain a licensing basis for commercial nuclear power physical security that is performance-based, scaled to the security risks of the facility, and allows licensee flexibility in meeting the licensed requirements to demonstrate reasonable assurance of protection against the DBT of radiological sabotage. The licensing basis will be centered around the development of several bounding security scenarios that will drive the determination of performance metrics, while being backstopped by the traditional NRC-observed force-on-force (FOF) performance testing. The number of security scenarios should scale to the size of the facility (both in number of units and number of potential adversary pathways) and the number of radiological sabotage targets. The security scenarios will represent (1) an adversary timeline to cause a high radiological consequence that exceeds the dose limits expected by the regulator and (2) a site protective strategy timeline that may rely on onsite response, offsite response, or a blend of the two to neutralize the adversary or end the security event consistent with the adversary interference precluded time (AIPT) concept. The use of defined scenarios with performance parameters associated with protective strategy detection, delay, and response timeline would allow flexibility by the licensee to meet these parameters individually or holistically within the licensing basis of the facility. The licensed performance times would allow variations and changes that meet the same performance metric, and allow the licensing of an established margin within the physical security system. The result of a licensee adopting this framework would be a licensing basis that provides operational flexibility, and clear indication of when the licensing basis is affected (when the parameters are impacted beyond the licensed values). This distinction would allow straightforward modifications and adjustments under §50.54(p) or the proposed § 53.1565(d)(4), and a clear demarcation when a license amendment is required consistent with § 50.90 or the proposed § 53.1510. 2. BACKGROUND The arena of physical protection of special nuclear material at nuclear power reactors has been one of responsiveness to threat environment impacts. The NRC security orders of the 2000s spurred by the 2001 terrorist attacks led to the codification of the new requirements in 10 CFR 73.55 and other pertinent sections. The application of high-security principles was applied ad hoc to existing nuclear power facilities to meet the requirements of protection against the DBT of radiological sabotage. The “New Reactors” timeframe of the late 2000s through early 2010s brought new reactor designs, some of which considered security within the design of the facility using the existing requirements found under 10 CFR 73.55. The NRC sought greater development of performance evaluation within physical security as early 1 as 2006 as evidenced by the later-canceled SECY-06-0204, “Security Assessment Requirements for New Nuclear Power Reactor Designs” proposed rulemaking. Reflecting the development of four advanced reactor types in the 1990s, the NRC issued the 1998 Commission Policy Statement on the Regulation of Advanced Reactors. The policy statement focused on early engagement between stakeholders, the design characteristics of advanced reactor designs, including safety, reliability, and efficiency, and the need to provide timely feedback. The policy statement also included points related to physical security, including the incorporation of physical security into the design features of the plant, regularly improved guidance on security features, and improved threat assessment approvals. The NRC security orders of the 2000s spurred by the 2001 terrorist attacks led to the codification of the new requirements in 10 CFR 73.55 and other pertinent sections. The application of high-security principles was applied ad hoc to existing nuclear power facilities to meet the requirements of protection against the DBT of radiological sabotage. The “New Reactors” timeframe of the late 2000s through early 2010s brought new reactor designs, some of which considered security within the design of the facility using the existing requirements found under 10 CFR 73.55. The NRC sought greater development of performance evaluation within physical security as early as 2006 as evidenced by the later-canceled SECY-06-0204, “Security Assessment Requirements for New Nuclear Power Reactor Designs” proposed rulemaking. The advent of generation IV reactor designs seeking commercial licensing spurred Congress and the NRC to support additional efforts to make the licensing process reflect the Nuclear Energy Innovation and Modernization Act (NEIMA) of 2019. This provided new transparency requirements for the NRC. The 2019 NEIMA Act had bipartisan support, setting the groundwork for new Part 53 licensing pathway for advanced reactors. The key objectives of NEIMA are modernizing the NRC’s licensing process to reflect a technology that includes a risk-informed and performance-based regulatory framework. The goal is to improve cost transparency and external expertise in new nuclear technologies and promote the use of innovative designs. The provision of NEIMA focuses on the use of risk-informed and performance-based licensing evaluations, which In July 2024, the Accelerating Deployment of Versatile, Advanced Nuclear for Clean Energy (ADVANCE) Act was signed into law. The ADVANCE Act covers a wide range of NRC activities with the intent of providing “reasonable and targeted” measured to reduce the regulatory cost imposed on the advanced reactor industry. The act focuses on including implementation initiatives for efficient, timely and predictable license application reviews, expedited review processes, and developing strategies and guidance for microreactors. The result of NEIMA is the advancement of several NRC performance-based rulemakings, including specific considerations for the performance-based licensing of physical security. 2.1. Physical Security Rulemakings To comply with legislation, requests from industry, and NRC-initiated goals to increase the use of performance-based regulation, the NRC has initiated three rulemakings over the last several years. NRC-2017-0227, “Alternative Physical Security Requirements for Advanced Reactors” was initiated in 2017 to provide a pathway to credit offsite response in response to a security event. The draft proposed rule for comment was published October 23, 2024 and provided an alternative pathway for compliance with 10 CFR 73.55 if the licensee met performance parameters associated with the postulated offsite release resulting from a security event. Accompanying the draft proposed rule language were two draft regulatory guides (DGs), DG-5071, “Target Set Identification and Development for Nuclear Power Reactors,” (draft Regulatory Guide 5.81 Rev. 2), and NRC, DG-5072, “Guidance for Alternative Physical Security Requirements for Small Modular Reactors and Non-Light-Water Reactors,” (Draft Regulatory Guide 5.90), July 2024. DG-5071 would provide updated guidance on the determination of radiological sabotage for advanced nuclear power reactors as a radiological release from any source. DG-5071 sets out the expectations for consequence analysis to support radiological sabotage determinations and the 2 methods to determine and support adversary interference preclusion time, offsite response, and the crediting of operator actions during a security event. NRC-2019-0062, “10 CFR Part 53: Risk-Informed, Technology-Inclusive Regulatory Framework for Advanced Reactors” provides an alternative performance-based framework in the draft 10 CFR 73.100 for Part 53 licensees to license the physical security of a commercial nuclear power plant. In addition, DG-5076, “Guidance for Technology-Inclusive Requirements for Physical Protection of Licensed Activities at Commercial Nuclear Plants,” (Draft Regulatory Guide 5.97) was also released to provide guidance to meet the proposed 10 CFR 73.100. NRC-2024-0167, “Law Enforcement Response in Power Reactor Physical Protection Programs” is an interpretive rulemaking that would provide the justification for the use of offsite response for any (not just advanced) commercial nuclear power plants licensed under 10 CFR 50 or 52. These three rulemakings open the door to many performance-based methods of compliance for licensees. The proposed 10 CFR 73.100 provides the clearest and broadest possibility of allowing performance-based methods to meet the proposed regulation. The proposed guidance in DG-5076 did not provide a robust description on how to meet the proposed rule in a performance-based manner and often provided prescriptive expectations for redundancy in functionality. This report examines NRC guidance on the development of performance-based regulation and applies it to the proposed rulemakings above (predominately 10 CFR 73.100) to create a performance-based method to identify and maintain the physical security licensing basis, providing flexibility for licensees and transparency and traceability for regulators. 2.2. Performance-based Regulation NRC published guidance for the development of a performance-based alternative to prescriptive regulations in NUREG/BR-0303, “Guidance for Performance-Based Regulation.” The intent of NUREG/BR-0303 was to inform the NRC staff in a consistent manner to consider a performance-based framework and discuss the best practices and issues with developing performance-based regulations. NUREG/BR-0303 lay out a five-step process to provide information to define a performance-based approach: 1. Defining the regulatory issue and its context 2. Identifying the safety functions 3. Identifying safety margins 4. Selecting performance parameters and criteria 5. Formulating a performance-based alternative. Section 3 of this report will follow the above steps to define the information needed to define the performance-based framework that is ultimately captured in Section 3.5. An important aspect of performance-based regulation is the identification of performance parameters, indicators that can demonstrate the performance of the function and allow issues to be identified and corrective actions to be taken to reduce or eliminate delicacies. Figure 1 is from NUREG/BR-0303 and outlines how the objectives hierarchy relate to the fulfillment of a fundamental objective. In this case, we will consider it fulfillment of security functions. 3 Figure 1. NUREG/BR-0303 objectives hierarchy. The objectives hierarchy lays out the fundamental objective to be achieved that can then be met by several means through cornerstones. Cornerstones are defined by their key attributes, which must have performance elements (performance parameters) to measure their success by. NUREG/BR-0303 states in Appendix B.3, “Allocation of Performance” the following point: Generally, it is desirable to specify and monitor performance targets as high on the objectives hierarchy as possible, consistent with the viability guidelines. Allocating performance too far down on the hierarchy reduces licensee flexibility. Arriving at an implementation that maintains safety, while appropriately balancing licensee flexibility with the need for regulatory assurance of ongoing performance, will require some iteration with the allocation step. NUREG/BR-0303 provides greater detail of the means objective in determining the performance parameter in Figure 2. 4 Figure 2. Detailed decomposition of means objectives. With this background on performance-based regulation development, we can tackle the concept of performance-based physical security and the development of a framework to determine and maintain the licensing basis. 3. PERFORMANCE-BASED PHYSICAL SECURITY FRAMEWORK This framework follows the process described in NUREG/BR-0303 to define the regulatory issue, the fundamental objectives, performance parameters, margins, and ultimately describing the performance- based alternative. This framework was designed to demonstrate compliances with the draft proposed 10 CFR 73.100 and also provide a basis for performance-based alternatives to 10 CFR 73.55. 3.1. Defining the Regulatory Issue and Context Physical security requirements have traditionally been prescriptively determined, with a performance- based assessment (limited-scope or FOF exercises) to ensure compliance. The NRC security program regulations found in 10 CFR 73.55 provide minimum expectations for the implementation of a physical security plan and the methods of compliance. New commercial reactor facilities pose new deployment and operational paradigms, and the NRC has sought to provide flexibility to reactor designers through the development of performance-based physical security rulemakings. The proposed Part 53 and Alternative Physical Security Requirements for Advanced Reactors rulemakings provide alternative methods of compliance for protection against the DBT of radiological sabotage, without prescriptive requirements. The NRC-proposed methods that could be leveraged to meet the proposed rule in DG-5072 and DG-5076 did not define the specific framework to license and maintain compliance with the proposed rule language. Therefore, a need arises to frame physical protection program compliance within a performance-based security framework found in the proposed rulemaking in a holistic manner that can 5 justify initial licensing, support continued compliance under operation, while allowing flexibility by the operator in meeting the performance-based requirements. 3.2. Identifying the Security Functions When designing and defining security requirements it is important to identify the functions that makeup a physical protection system (PPS). A PPS consists of three key functions detection, which includes assessment and communication, delay, and response. PPSs have been described as a three-legged stool. For a PPS to be effective at mitigating any nuclear security event, the system must be equipped with detection, delay, and response. If the PPS is missing one of these three functions or one of these functions is inadequate, the PPS will fail at its intended function of mitigating nuclear security events. This is similar to a three-legged stool with a missing or broken leg. Figure 3. Functions of a PPS. The following sections will discuss and identify the functions of a PPS and the associated performance requirements that are used to evaluate the effectiveness of these functions. 3.2.1. Physical Protection System A PPS must be designed with security goals in mind to ensure that the PPS mitigates or prevents an undesired act. In nuclear security, the ultimate goal of a security system is to prevent the theft of nuclear material and to mitigate or prevent an act of sabotage that could lead to a radiological release. The descriptions of PPS functions will be described in the context of preventing a radiological release from a small modular reactor or microreactor. The figure below highlights the basic functions of a PPS. 6 Figure 4. PPS functions. It is important to note that the three functions of detection, delay, and response can be implemented with technology, people, or procedures. Often, a PPS is designed and implemented by using the three functions and a combination of technology, people, and procedures to achieve the desired outcome of the PPS. The three functions of a PPS are seen as subsystems of a PPS. A PPS is considered a complex system and requires a systems engineering approach. 3.2.2. Detection Intrusion detection systems (IDSs) are used for many purposes. IDSs consist of exterior intrusion detection sensors, interior intrusion detection sensors, alarm assessment, contraband detection, access controls, and alarm communication and display (AC&D) systems. There are many factors that can be evaluated in detection that can be used as performance indicators for a PPS. These factors include: • Probability of detection • Probability of assessment • Assessment time • Communication time. Additionally, many regulations and best practices have been in place to evaluate the performance of detection systems. These regulatory requirements will be referenced throughout this section. 3.2.3. Delay Delay barriers and systems play an important role in the overall effectiveness of a PPS. Delay barriers can be used to slow down an adversary to increase the overall adversary task time to complete an act of sabotage, and used to channel adversaries into advantageous locations that can improve the ability of the response force to neutralize the adversary. Therefore, delay barriers can be beneficial to both the probability of interruption and neutralization. Delay can be any barrier, distance, or element that an adversary must defeat, cross, or bypass to gain access to a system, structure, and component (SSC) location that if sabotaged could lead to a radiological release. An important note when evaluating delay times is that delay times can only be calculated as part 7 of the total adversary attack timeline if the delay barrier occurs after detection. For delay barriers to increase the adversary task time, the PPS must first be able to detect (sense and assess) that an adversary is attempting to attack the facility and attempt to cause a radiological release. Figure 5 below is a hypothetical example of how delay barriers are incorporated into the overall adversary task timeline. As can be seen from the figure below, each element, barrier, or space that the adversary must cross is calculated as part of the overall adversary task timeline. In the scenario below, fences, doors, moving between security layers, accessing the target, and sabotaging the target are incorporated into the overall adversary task timeline. Figure 5. Example of delay and adversary timelines. An important factor to consider in delay is that the delay is only effective after detection has occurred. If detection has not occurred, then it is not possible for the PPS and the PPS personnel to know that the adversary is attempting to penetrate the facility. Delay timelines are only accounted for in the overall PPS timeline when it occurs after detection. 3.2.4. Response The response force is the third critical function of a PPS. Response plays the critical role of the facility being able to interrupt and neutralize an adversary force before they complete their malicious act. The response force must interrupt the adversary and then neutralize the adversary for the system to be effective. Interruption is the successful arrival of the response force at an appropriate location with enough responders to stop the adversary. Neutralization is the ability of the response force after interruption to prevent completion of the adversary action. Neutralization is achieved when the response force kills, captures, or causes the adversary to flee before the adversary can complete the task. The overall effectiveness of the response force depends on many factors, including communication training, equipment, experience, adversary capabilities, tactics, knowledge, procedures, and use-of-force policies. The response force must have reliable and timely communication from a central alarm station (CAS) operator about a potential nuclear security event, a description of the cause of the security event, and 8 potential movements of the adversary that may have caused the alarm. Without communication from the CAS about an alarm generated at the facility, responders cannot be activated to interrupt the adversary force, and without interruption, the response force cannot neutralize the adversary force. Additional information, such as descriptions of adversaries from the CAS and movements of the adversaries, can help to improve the effectiveness of the response force to interrupt and neutralize the adversary forces. This communication also ensures that the response force arrives to the correct location to interrupt and neutralize adversaries. Once the response force is notified of alarms, the responders can then move to a location to interrupt the adversary before they complete their malicious act. This total time from response notification to arrival at the interruption location is a key factor in determining the effectiveness of the PPS. Once the responders arrive at the interruption point, there are additional factors that can influence the overall effectiveness of the response force to neutralize the adversary force. These measures include the equipment of the responders and the adversaries, knowledge of tactics, physical fitness, training, and use- of-force policies. The first factor is equipment used, if the response is less equipped and has fewer responders than the adversary, the responders may be less effective at neutralizing the adversary force. Knowledge of tactics plays a key role in the responders’ ability to execute contingency plans and response plans to nuclear security events to ensure the adversaries can be neutralized. Physical fitness and training play an important role in the response force’s ability to execute the contingency plans to respond to a nuclear security event. Training also ensures that the responders have the ability with their equipment to properly neutralize the adversary. The final factor that can influence the effectiveness of the response force is the use-of-force policies that are in place for the responders. Use-of-force for armed responders at nuclear facilities can be based on the legal authority of the responders, rules governing the conditions for the use-of-force, and the types of methods that can be implemented for use-of-force. Figure 6. Use-of-force factors. 9 3.3. Selecting Performance Parameters and Criteria Following NUREG/BR-0303, the selection of performance parameters that satisfy the viability guidelines and regulatory issue as described. NUREG/BR-0303 states: Generally, a performance-based regulatory alternative needs to: 1. allocate performance across relevant functions, systems, or barriers, in order to assess whether the target safety objectives are satisfied 2. then implement that allocation of performance which entails identifying the steps to be taken by licensees and/or NRC to make the performance allocation “come true” in practice …Part of implementation is confirmation of ongoing performance. The Advisory Committee on Reactor Safeguards recommends setting the performance criteria as high as a level as practicable to allow flexibility. The intent is that the performance parameters define the success of the system and identify opportunities for corrective action. Corrective action could be temporary (compensatory measures) until the performance meets the criteria, a change to the facility design or operational procedures to address the inadequacy of the current facility to meet the licensed performance parameters, or a license change justified through analysis to establish new performance parameter values. The following is a discussion of physical security hierarchy, and how performance parameters can be identified that tie to regulatory compliance. Hierarchy The regulatory framework guidance provided in NUREG/BR-0303 adapts well to the space for the proposed performance-based physical security licensing framework given in the 10 CFR Part 53 rulemaking § 73.100, “Technology-inclusive requirements for physical protection of licensed activities at commercial nuclear plants against radiological sabotage.” § 73.100(b)(1) provides the general performance objective as: The licensee must establish, implement, and maintain a physical protection program and a security organization, which will have as their objective to provide reasonable assurance that activities involving special nuclear material are not inimical to the common defense and security and do not constitute an unreasonable risk to the public health and safety. § 73.100(b)(2) defines that § 73.100(b)(1) is satisfied by the prevention of radiological sabotage and that capability must be maintained at all times, with defense in depth (DID), to achieve the performance requirements. § 73.100(b)(3) provides performance requirements for the physical protection program in (i) through (vi) for intrusion detection, intrusion assessment, security communication, security response vehicle bomb assaults, and access control portals. § 73.100(b)(4) through (12) provide other supporting or programmatic requirements for the security of the facility. 10 Thus, we can summarize the performance-based hierarchy as seen in Figure 7. Protect Public Health and Safety § 73.100(b)(1) Prevent Radiological Sabotage § 73.100(b)(2) Intrusion Security Intrusion Detection Security Response Bomb Assault Access Control Assessment Communication § 73.100(b)(3)(i) § 73.100(b)(3)(iv) § 73.100(b)(3)(iv) § 73.100(b)(3)(iv) § 73.100(b)(3)(ii) § 73.100(b)(3)(iii) Figure 7. 10 CFR 73.100 objective hierarchy. The hierarchy of the performance parameters relates to the fulfillment of the security program’s general performance objective and requirements, and the security functions identified in Section 3.2. Ultimately, the prevention of irreversible release of radionuclides demonstrates the prevention of radiological sabotage and is the goal of the security program. The performance parameters of individual security functions can be summed to meet the defined timeline and capability, demonstrating success in the prevention of radiological sabotage. 11 Demonstrating the prevention of radiological sabotage relates to the demonstration of capability in neutralizing the adversary by onsite response or the conclusion of the security event through the arrival and action of offsite response forces. Thus, the timeframe, as well as capability of these security functions, becomes the greatest points of interest; security elements can effectively meet their function within the required timeframe. Figure 8 demonstrates the security function parameters detection (including assessment and communication), delay, and response in the security event timeline. Detection Attack Detection Initiated Assessment Communication Initiated Adversary Physical Protection System Delay Adversary Disables Target(s) Irreversible Release Timing Response Communication Received Response Initiated Armed Response Engages End Security Event (neutralization or AIPT) Irreversible Release Timeline Site Security Timeline (neutralization or AIPT Margin Figure 8. Adversary and site security function timeline. The irreversible release timing defines the requirement for the site security timeline. The site security timeline consists of the functions of detection, delay, and response and seeks to demonstrate protection of the site until the site security bounding time is reached or the adversary is neutralized. The site security bounding time (defined as the AIPT in DG-5072) defines the time period that a site protective strategy must prevent radiological sabotage for (through passive delay features, active delay features, or neutralization by onsite response) until offsite response can be assumed to control the site. Ultimately the parameters for the performance-based security framework are identified as: • Irreversible Release Timing (a function of the reactor design) • Site Security Timeline (a function of the site protective strategy) - Detection Time - Delay Time - Response Time • Probability of Detection and Confidence (a function of detector selection and design) • Probability of Effectiveness (a probability of response force capability). 12 3.3.1. Irreversible Release Timing The parameter associated with the irreversible release timing will ultimately define the required performance of the PPS sum of parts associated with detection (including assessment), delay, and response (communication and interdiction). This parameter results from the target set analysis as described in DG-5071. Target sets are the site-specific combinations of equipment and actions whose sabotage by an adversary leads to an unacceptable offsite release. a The target set identification process will screen out those targets that do not meet the definition of a target set (could not result in a release in exceedance of the dose reference values), or those targets that are beyond the capability of the DBT adversary. The remaining list of target combinations are called “achievable targets” and are defined as those targets that must be considered in the protective strategy, with each target set having an irreversible release time. The irreversible release time for each target set is determined from the shortest time from the initial adversary action to the point that represents the last point that operator actions can reverse sabotage actions to prevent a significant release, and then that time walked back to the point operators need to initiate their actions. The initial irreversible release time is computed though the determination of adversary traversal times, breach times, and action times to complete the sabotage of all target elements, plus the progress of the sabotage event until the event is deemed irreversible. Operator actions to reverse a significant release may involve actions that prevent a release or mitigate the consequences below required dose consequence values. For sabotage events that are irreversible or unmitigable, the irreversible release timing would be the point of sabotage of the last target element of a target set. Initial irreversible release time determinations should not include any active components of the physical protection program (i.e., response) to provide a baseline determination for the screening of target sets against the site security bounding time. The physical protection program could implement earlier detection, delay elements, and response personnel to extend the irreversible release timeframe from the values determined from the target set analysis. Modifications to facility safety systems, structures, and components, or operational emergency procedures may also extend the irreversible release time. The neutralization time is the time associated with the detection, delay, and response of the physical protection program and represents the time from detection of an adversary to the conclusion of a security event through neutralization of adversary forces. This timeframe defines the site-only response to prevent radiological sabotage. Prevention of sabotage through delay and any iteration of offsite response can be defined though the AIPT. The AIPT time can be determined following guidance in DG-5072 to determine the timeframe that offsite response can conservatively be assumed to arrive at the licensed site and control the facility, allowing operator recovery actions to occur if necessary, as a bounding value for the entire site. The site protective strategy must be designed to protect the site until this occurs, either through delay and or neutralization of the adversary. Shorter individual target set and security event security bounding times can also be determined following DG-5072 to inform specific security event timelines. a. DG-5071 specifically defines a target set for non-LWRs and SMRs as “the minimum combination of equipment, operator actions, and/or structures that, if all are prevented from performing their intended safety function or prevented from being accomplished, barring extraordinary actions by plant operations, would likely result in a significant release of radionuclides from any source (e.g., a release to the environment exceeding that analyzed in the design basis accident licensing basis).” 13 Detection Attack Detection Initiated Assessment Communication Initiated Adversary Physical Protection System Delay Adversary Disables Target(s) Irreversible Release Timing Response Communication Received Response Initiated Armed Response Engages End Security Event (neutralization or AIPT) Irreversible Release Timeline Site Security Timeline (neutralization or AIPT) Margin Figure 9. Irreversible Release Timing vs. Site Security Timeline. Security events with irreversible release timing before the determined site security bounding time define those target sets that must be protected by the security force until the site security bounding time. The security functions must meet the event-specific security timeline associated with those security events with irreversible release timing less than the security bounding time or an event-specific security bounding time (as specified in DG-5072). The successful delay or neutralization of adversaries would “push” these events further in time (to the right) until they fall within the security bounding time (and are stopped through offsite response) or the adversary is neutralized, preventing the sabotage of the target set. The irreversible release timing provides the licensing basis for the facility (to prevent a large radiological release from any source) and the design objective, as the irreversible release timing can be met through a combination of: • Site security timeline reduction - Earlier detection - Earlier interruption and or neutralization - Earlier AIPT • Irreversible release timeline increase - Increase adversary delay to targets - Modify the reactor system to delay or prolong release - Modify the reactor system to increase adversary travel distance (though separation or number of target elements) - Increase ease of operator mitigating or recovery actions. 14 In this way, the irreversible release timing can be used as a design goal of the PPS to support traditional and security-by-design methods that ultimately demonstrate, for the licensing basis and during testing and inspection, high assurance of the protection of public health and safety. 3.3.2. Site Security Timeline The site security timeline is the site response to the detection of adversary action. This timeframe is composed of the elements of detection (including assessment and communication), delay, and response. The site protective strategy may rely on site passive and active delay features (to include response personnel) to the facilitate the response by the offsite response forces to arrive and secure the site. The site security timeline is concluded when the adversaries are neutralized or the security bounding time has elapsed and determines the end of the security event. This parameter is thus comprised by the sum of detection, delay, and response. These parameters can be treated and maintained separately. Treated separately, these values can be monitored closely for performance indicators. 3.3.2.1. Detect Detection is the process in which a physical sensor in the deployed environment first senses a change in the environment. Once this sensor identifies this change, a signal is sent through a communication network to the AC&D system that a CAS operator is staffing. Once the signal arrives, an alarm is reported to the CAS and should be audibly and visibly announced to the CAS operator. Once the CAS operator receives the alarm notification, they use video feeds, either live or recorded from closed-circuit television (CCTV) cameras. The video feeds then allow the CAS operator to determine whether the sensor has alarmed due to an adversary incursion or due to a nuisance or a false alarm. Once the assessment is made, the CAS operator communicates the cause of the alarm to the appropriate person in the security organization to respond to the alarm cause. The figure below shows the process that is used for alarms to be generated and assessed in the CAS. This process utilizes people, technology, and procedures to complete the detection process. Figure 10. PPS detection process. As a parameter, detection will include detection, assessment, and communication to the appropriate response forces. 3.3.2.2. Delay Note: All times listed in this section are purely hypothetical and should not be used for any official security analysis. The key performance metric for a delay barrier is the time it takes the adversary to bypass the area or barrier that is along the adversary pathway to the target location. Delay times include both the time it takes for the adversary to move across areas within the facility, the time it takes the adversary to breach a delay barrier, or the time it takes the adversary to bypass a barrier at the facility. 15 The first delay time addressed is the time it takes the adversary to cross areas of the facility (e.g., the protected area, across rooms). This is generally measured as a function of the weight an individual is carrying, the type of area they are crossing (e.g., grass, gravel, cement) and the total distance the individual is traversing. This delay time can be gathered in two fashions, 1) evaluation and testing of the DBT adversary across all areas that they may encounter at the facility; or 2) using delay databases that have been generated through the Department of Energy’s National Nuclear Security Administration. Once the time to cross an area has been determined, the time it takes can be used in the total calculation of the adversary timeline to complete the task of sabotage at the facility. Figure 11 shows a hypothetical example of running rates or movement rates throughout the facility. Information such as this could be used in adversary timeline analysis. 90 1—On paved/unpaved ground 2—With tools (1.07-m bolt cutters, pickax and shovel) 1 2 3 80 4 3—On sand 5 4—With weight (16 kg in toolbox) 5—With 2.4-m stepladder 6 6—With 10-m extension 70 ladder (2 men) 60 Distance (meters) 50 40 30 20 10 0 0 5 10 15 20 25 Time (seconds) Figure 11. Example of running rates. 16 Figure 12. Hypothetical sample of adversary attack tools. The second factor of delay measures is the time it takes an adversary to breach delay barriers to reach the target location and the time it takes the adversary to sabotage targets at the facility. The time it takes to breach any individual barrier that may be in place at the facility is dependent on the adversary tools and capabilities as well as the material and barrier that the adversary is attempting to breach.. Table 1 shows a hypothetical example of delay times based on the barrier and the tools that an adversary may have. 17 Table 1. Hypothetical breach delay times. No Hand Power Explosives Land Component Component (mm:ss) Equipment Tools Tools Vehicle Type Description (mm:ss) (mm:ss) (mm:ss) Stage 1 Stage 2 (mm:ss) Walls 60-cm reinforced Infinite Infinite 15:00 3:00 5:00 Infinite concrete wall 30-cm reinforced Infinite Infinite 14:00 2:00 0:54 N/A concrete wall 20-cm reinforced Infinite Infinite 10:00 2:00 0:00 N/A concrete wall Wood studs and 1:00 0:30 0:30 0:30 0:00 N/A sheetrock Doors 60-cm steel and Infinite Infinite 15:30 3:20 5:00 N/A concrete rolling door 30-cm steel and Infinite Infinite 10:40 2:40 0:54 N/A concrete rolling door 30-cm wood door Infinite Infinite 8:50 2:40 0:30 N/A with metal sheeting 10-cm wood door 0:05 with metal sheeting Infinite 5:00 3:00 1:18 0:00 large vehicle door 5-cm wood door Infinite 0:12 N/A 5-cm wood door with Infinite 0:12 0:12 0:12 0:00 N/A glass panel 0.75-cm steel plate Infinite 5:00 0:12 0:12 0:00 N/A door Class V or VI vault Infinite 8:00 0:30 0:30 0:00 N/A door Steel turnstile Infinite 1:12 1:00 1:00 0:00 N/A Other High-security Infinite 1:30 0:18 0:18 0:00 N/A Barriers padlock Concrete block 0:00 5:00 1:00 0:30 0:00 0:05 vehicle barrier 2.5-m chain-link 0:10 1:00 5:00 0:30 0:00 0:01 mesh fence Welded wire fabric 0:10 0:10 0:10 0:10 0:00 0:01 fence 2.5-m concrete panel 0:10 0:10 0:10 0:10 0:00 N/A wall Tempered glass 0:05 0:05 0:10 0:10 0:00 N/A window Electromagnetic 0:15 0:10 0:05 0:05 0:02 N/A strike lock 18 As can be seen in the table above, the adversary timeline can depend on the tools that the adversary is using to breach the barrier and the overall barrier material. This data is traditionally collected by testing the barrier over a statistically significant number of tests to determine the average time to defeat the barrier. It is important to note that when this data is collected, there are many important factors that may impact the time it takes the tester to defeat the barrier. These factors include but are not limited to: 1. The number of trials the tester has conducted on the same barrier 2. Weather and environmental conditions during the testing 3. As trials increase for the tester, the tester may have become familiar both physically and mentally with ways to defeat the barrier. Once these trials are conducted, the overall average time is then determined based on the overall distribution from the trials that are conducted to defeat the delay barriers. Another method for determining the time to breach a barrier can be gathered from databases, commercial standards or tests, or gathering data from professionals (e.g., firefighters) who use tools to breach through doors, windows, walls, etc. Another source of potential information that can be used to determine times to breach barriers is through professional standards. For example, ASTM International F2028-21 standard shows a method to test and evaluate doors for their resistance to penetration by individuals or groups of individuals. These standards can be used to assess whether barriers are resistant to specific breach methods and capabilities that may be used by an adversary. Delay barriers can also include those inherent into the targets that must be sabotaged that could lead to a radiological release. For example, when considering an adversary whose goal is to conduct sabotage of spent fuel inside of a dry-storage cask, the dry-storage cask also prevents a barrier that must be penetrated by the adversaries to cause a radiological release. The time it takes the adversary to cause a breach in the dry-storage cask is also considered part of the overall adversary task time that would be considered a successful adversary attack scenario. This applies for all potential targets and the time it takes to cause the sabotage of these targets. As a final example, if the adversary must completely shut a valve to cause a radiological release, the total time it takes the adversary to complete the valve shut off is also incorporated in the overall adversary attack timeline. It is important to note that the time required for breaching a barrier may depend on the time of breach that needs to be made by the adversary. For a breach to be successful by an adversary, the overall goal of the breach must be accomplished. For example, if the adversary must breach a barrier to ensure the adversaries can pass through the barrier, the adversary must create a breach that is passable for individuals. If a breach only requires a small penetration to cause a release or cause water to drain, then the breach may not be as large as the breach necessary for people to pass through. The final delay time that is calculated in the overall adversary attack timeline is the time it takes an adversary to bypass barriers at the facility. Bypassing a barrier considers an adversary that is not attempting to penetrate through a barrier but circumvent the barrier. An example of this is adversaries who face a chain-link fence at a facility. The adversaries may not attempt to breach or penetrate the fence, but may choose to climb the fence or tunnel underneath the fence. These times similarly can be collected through performance testing, databases, or standards. Facilities should gather these delay times for all areas and barriers at the facility that must be bypassed or penetrated by adversaries to reach their target location, and the time it takes to sabotage a target at the facility that could lead to a radiological release. Once all these times have been identified, they can be incorporated to develop the adversary task timeline and then used for comparison against the PPS timeline. Determining delay times plays an important role in evaluating the overall effectiveness of the PPS. 19 3.3.2.3. Response There are key performance parameters that are used to evaluate the effectiveness of the response force. These factors are also integral in evaluating the overall effectiveness of the PPS. Evaluation of the response force is used in both the probability of interruption and the probability of neutralization. Both are key factors in evaluating the overall effectiveness of the PPS. Unlike detection and delay, testing and collection of response force performance criteria is taken from conducting tests of the responders for a facility based on the protection strategy, targets to be protected, and contingency plans implemented to defend the facility against a DBT attack. Testing of the response force can be conducted through limited-scope performance tests (LSPT), time motion studies (TMS), alarm response assessment performance test (ARAPT), enhance limited-scope performance tests (ELSPT), and FOF exercises. Additionally, tests also include testing of responder proficiencies with weapons and the use of combat simulation tools. These methods can be utilized to evaluate the performance of the responders and within overall evaluations of the effectiveness of the systems. One of the first performance metrics utilized to evaluate the response force is communication time from the CAS to the responders, to respond to a nuclear security event at the facility. This time is used to determine how fast the response force is notified of a potential security event and begins to start the response ready time. Once the response force is communicated to about the cause of a security alarm at the facility, the response team must gather equipment (if needed) and begin other necessary steps for interrupting the adversary force. Depending on the strategy implemented, another time that must be evaluated is the time that it takes the response force to then deploy and arrive at a location where they can interrupt the adversary force. These above times for the response force can be evaluated by the facility using many of the testing methods described above. LSPTs can be utilized to determine specific times and proficiencies of responders for gathering and donning equipment needed in a security event. These tests can include proper loading of a weapon, proper methods for charging a weapon, etc. LSPTs are meant to test specific portions of the response force and response procedures. A TMS can be utilized to determine specific response times from a starting location to a finishing location. 3.3.2.4. AIPT The AIPT represents the time for an offsite response force to respond and secure the site against an adversary. We can treat the AIPT time as the end of the security event, similar to neutralization; as at that point, any operator actions may occur unimpeded. The AIPT methodology would apply for a private response force or local law enforcement. DG-5072 Appendix C proposes the framework to determine the site-specific AIPT for a facility. These steps would include: 1. Alarm assessment and communication time 2. Response time (travel and assembly) 3. Mission preparedness time (briefing, planning, and any rehearsals) 4. Mission execution time 5. Additional action time (operator actions). For this framework, step 5 would not be included within the AIPT determination, thus the AIPT would be a site-wide value associated with securing the facility. The operator actions associated with preventing or mitigating the release would be included in the target-set-specific irreversible release timing values. 20 3.3.3. Detection Probability and Confidence Probability of detection (PD) is the likelihood that an adversary action can be sensed and assessed. Probability of detection is a key performance metric for a PPS and is directly impactful to the probability of interruption. Probability of detection is determined by both the sensors probability of sensing (PS) and the likelihood that the alarm station operator can correctly assess the cause of the alarm (PA). The equation to calculate the probability of detection can be seen below. 𝑃𝑃𝐷𝐷 = 𝑃𝑃𝑆𝑆 𝑥𝑥 𝑃𝑃𝐴𝐴 Probability of sensing is a function of the sensor’s performance that is used in the operational environment. The probability of sensing is the likelihood that the sensor identifies a change in the physical environment based on the sensor characteristics and performance. It is important to note that while probability of detection is defined as a function of probability of sensing and probability of assessment, probability of assessment is generally assumed to be one. This assumption has been made on the basis that probability of assessment is extremely difficult to evaluate. The reason for this is the following factors play into a determining probability of assessment: • AC&D configuration • Recorded video feeds (auto call up or self-selected) • Pre and post-alarm recorded video • Integrated systems or separated systems • Human factors. These factors make probability of assessment extremely difficult to determine and site-specific. This has resulted in probability of assessment to be assumed as one, and therefore, probability of detection is often simplified to be taken as the probability of sensing for an individual sensor. The U.S. DOE requires sensors to have a 90 percent probability of detection with a 95 percent confidence. b In addition to this probability of sensing, the DOE requires that this be proven annually in accordance to performance testing standards. The NRC requires that sensors also have a 90 percent probability of detection at a 95 percent confidence level. c. The NRC Regulatory Guide (RG) 5.44 requires that sensors be operationally tested once every 7 days and that performance tests be conducted every 90- days. Once these testing requirements are known, each sensor must then be evaluated to determine whether it meets this 90 percent probability of detection at a 95 percent confidence level, which is also known as 90/95 testing. Generally, to determine whether a sensor meets this 90/95 rule, 30 tests are conducted across three different adversary defeat methods. These adversary defeat methods include walking, crawling, and running. However, sensors should be tested for all potential adversary defeat methods to include jumping. Once these 30 tests have been conducted, the probability of detection can be determined. The table below outlines the number of trials and the number of failures that allows the user to identify the probability of detection based on the confidence level. As can be seen from the table below, if tests are conducted on a sensor 30 times, per the given defeat method, and the sensor detects the tests 30 times, the probability of detection would be 91% at a 95% confidence level. However, if the sensor fails to detect once, the probability of detection drops to 85% at a 95% confidence level. Because of this, these types of performance tests are generally called a 30-for-30 test. b. DOE Order 473.1 A, Chapter II 4.h.4. https://www.directives.doe.gov/directives-documents/400-series/0473.1-BOrder- a/@@images/file c. NRC Regulatory Guide 5.44, “Perimeter Intrusion Alarm Systems.” https://www.nrc.gov/docs/ml0037/ml003740097.pdf 21 Table 2. Probabilities of detection for 90/95 tests. Trials Failures Confidence Levels 75% 80% 85% 90% 95% 30 0 95% 95% 94% 93% 91% 30 1 91% 90% 89% 88% 85% 30 2 87% 86% 85% 83% 80% 30 3 84% 82% 81% 79% 76% This process allows for the performance and adequacy of the sensor to be determined. These factors ultimately play into determining the probability of interruption and therefore the overall security system effectiveness. The probability of detection does not need to be associated with a single sensor but could be determined from the fractional contribution associated with several detection lines. The detection probability and confidence must occur before the licensee-defined critical detection point (CDP), which NUREG/CR-7145 defines as the point on the adversary’s path where path delay exceeds response force interruption time. Functionally, the CDP will allow time for response force interdiction and effective delay or neutralization (depending on protective strategy) and can be the parameter associated with the detection of an adversary of the PPS. As such, summed probabilities of lesser sensors can support functional compliance, as represented by: This may manifest as two sensors within an intrusion detection zone, each providing detection capability to the total performance, or if the CDP is located closer to the target, may include sensors within the intrusion detection zone and reactor building access doors, depending on the protective strategy. The use of the CDP on the adversary timeline would allow greater flexibility to the licensee in the design and operation of the PPS and allow a functional flexibility to be built into the license. Probability of detection is only one of the performance metrics that is used to measure the overall performance characteristics of a PPS. The other factors that are used to determine performance of intrusion detection sensors are time for alarm communication, time for assessment (TA), and time for communication (TC). The time for alarm communication is the time it takes for the sensor when activated to signal the cause of an alarm by audible or visual annunciation to the alarm station operator. The time for assessment is the time it takes for the alarm station operator to use CCTV cameras to assess the cause of the alarm as a nuisance, real, or false alarm. The time for communication is the time it takes for the alarm station operator to communicate the cause of an alarm to the responsible security organization. In this scenario, time for communication is the time it takes for the alarm station operator to communicate an alarm cause to the response team lead (RTL) who will then communicate to the response force how to respond to a nuclear security event based on developed contingency plans to the start of the adversary attack scenario. Because of the rapid growth and speed of modern security systems, the time for alarm communication is negligible and not used in evaluating the effectiveness of the PPS. These times are all determined by conducting tests to determine the speed in which the alarm station operator can assess alarm cause, conducting tests to determine communication time from the alarm station operator to the RTL and by testing the time it takes the RTL to communicate the plan to interrupt the adversary to the active responders at the facility. 3.3.4. Probability of Effectiveness When a PPS is being designed to prevent a high radiological consequence release, it is important to understand the performance of each function to determine the overall effectiveness at preventing an act of sabotage. This is done by analyzing the performance of each function during the design process. It is 22 important that during the design process, the PPS designers must think as an adversary would and ensure that the characteristics, attributes, and capabilities of the adversary and the DBT that the facility must protect against. The main reason is that one of the key elements in determining the effectiveness of PPS is to determine if the response force can interrupt the adversary from completing their tasks or objectives with a response force. This is often referred to as the probability of interruption (PI). The figure below shows a representative example of the comparison of adversary attack timelines with the site security timeline, labeled PPS time (i.e., the time it takes for the response force to interrupt an adversary force). Figure 13. Adversary vs. PPS timeline. To evaluate whether a security system is effective, the traditional approach has been to use system effectiveness (PE). System effectiveness is the probability that the PPS is able to defeat an adversary group before they are able to complete their intended malicious act—in this case, sabotage of the safety SSCs that could lead to a radiological release. System effectiveness is calculated by using both the probability of interruption and the probability of neutralization (PN). Probability of neutralization is determined by identifying the likelihood that the PPS is able to neutralize the adversary once the adversary is interrupted. Therefore, neutralization cannot occur without the interruption of the adversary force. System effectiveness can be calculated using the equation below. 𝑃𝑃𝐸𝐸 = 𝑃𝑃𝐼𝐼 𝑥𝑥 𝑃𝑃𝑁𝑁 System effectiveness is the total measure of the system’s ability to prevent an adversary from achieving sabotage that could lead to radiological release. The following sections will describe the security functions and how their performance basis can be determined using existing standards and methodologies that have been in use by the NRC, DOE and the International Atomic Energy Agency (IAEA). Total system effectiveness supports the justification for performance within the timeline. 3.3.4.1. Probability of Interruption The first step in identifying the overall security system effectiveness is to calculate the probability of interruption (PI). PI is the probability that the response force can arrive at a location in time to interrupt the adversary before the adversary is able to complete their task. The equation to calculate PI is shown below. In the equation above, PD1 is the probability of detection at the first sensing opportunity of the PPS. PDCDP is the probability of detection at the CDP. The steps used to determine probability of detection are listed below: 23 1. Create an adversary timeline: Document the sequence of actions taken by the adversaries to include delay times and probabilities of detection for sensors that the adversaries may cross 2. Create a response timeline: Document the timeline for the response team, including assessment time, communication time, response force deployment times, and response force transit times 3. Identify the CDP: Last opportunity for timely detection in which the response force has time to respond and arrive at a location before the adversary can complete their task. 4. Calculate PI: Use the above equation and the gathered data to determine the probability of interruption. The first step in the process is to create the adversary timeline. The data gathered from delay barriers and sensor performance testing plays the key role in determining these values to be used to identify the probability of interruption. Table 3. Sample adversary timeline with probabilities of detection. The second step in this process is to create a response timeline. Data gathered from evaluating CAS operators and the response force is essential to developing the PPS response timeline. Information such as assessment time, communication time, response ready time, response travel time, and response deployment are critical components in building the PPS response timeline. A hypothetical PPS response time is seen in the figure below. Figure 14. Hypothetical PPS response timeline. 24 The third step in this process is to identify the CDP. Figure 15 shows an example of how to determine the location of the CDP based on the hypothetical scenario throughout this document. Figure 15. Sample CDP identification. The CDP is determined by identifying the detection point in which the adversary task time is longer than the PPS response time from the final task in the adversary timeline. As shown in Figure 15, this occurs as the adversary is moving from the fence line of the facility to the outer door of the facility. Once this location is identified, the probability of interruption can be determined. In this example, the probability of detection is equal to 0.92. This analysis is crucial in determining the overall likelihood that the response force is able to respond in adequate time to interrupt an adversary force. 3.3.4.2. Probability of Neutralization Probability of neutralization (PN) is the likelihood that the response force is able to neutralize the adversary before their malicious act can be accomplished. Neutralization can occur when the adversary decides to give up their tasks and retreat from the facility, when the adversary is unable to complete their intended tasks, when the adversary is arrested by the response force, or when the response force kills or renders all adversaries unable to complete the malicious act of sabotage. There are many methods to determine probability of neutralization. The first method is a simple numerical table to compare probabilities of neutralization. For high-security applications, such as nuclear power facilities, this method is not recommended. The probability of neutralization is calculated by comparing the number of adversaries to the total number of responders involved in the engagement. There are many assumptions incorporated into a simple numerical table. These assumptions include but are not limited to: • Responders and adversaries are trained the same • Responders and adversaries have the same weapon systems (e.g., 5.56-mm ammunition) • Responders and adversaries have the same protective equipment (e.g., body armor) • Responders and adversaries engage each other from the same shooting posture (e.g., crouching, prone, standing, running) 25 • Use of Makarov chains, which assumes that the future state of a process depends only on the current state and not any previous states. Table 4 is a hypothetical simple numerical table that can be used to calculate the probability of neutralization and uses a Makarov chains and a 3-1 guiding principle. The 3-1 principle is a military strategy that requires the attacking force to outnumber the defending force by a ratio of at least 3-1 to have a reasonable chance of success. d This can be seen in the table below where the probability of neutralization is 4 percent when there are three adversaries engaging one adversary. As an example, if the response force is equipped with five responders against three adversaries, the probability of neutralization is estimated to be 89 percent. Table 4. Hypothetical neutralization analysis: simple numerical table. Table 5 shows how this kind of analysis could be evaluated comparing the number of armed responders with varying capabilities, and response time is determined based on a hypothetical group of three adversaries armed with handguns. Table 5. Hypothetical probability of neutralization determinations. Another method to evaluate the probability of neutralization is the use of combat simulation tools. These tools are used to simulate security scenarios that the facility may encounter. These combat simulation tools use set parameters such as DBT capabilities, responder capabilities, number of d. https://dupuyinstitute.org/2019/11/14/the-source-of-the-u-s-army-three-to-one-rule/ 26 responders, and developed security plans to evaluate the response to a security event. When using these tools the probability of neutralization can be calculated by a simple equation, as seen below. 𝑃𝑃𝑁𝑁 = In the equation above, the number of engagements must be a statistically significant number of simulations conducted in the combat simulation tool that is used. The process for using combat simulation tools is: 1. Modeling Adversary Behavior: - Adversary Profiles: Simulation tools often include predefined adversary profiles that describe potential attackers’ capabilities, tactics, and strategies. These profiles can be customized to reflect specific threats relevant to a facility. - Attack Scenarios: Users can create various attack scenarios based on the adversary profiles, which detail the sequence of actions an adversary might take to achieve their objectives (e.g., theft of nuclear material or sabotage). 2. PPS Modeling: - Facility Layout: The simulation tool incorporates a detailed representation of the nuclear facility, including its layout, security features (e.g., fences, cameras, guards), and critical assets (e.g., reactors, storage areas). - Security Measures: Users can input various security measures and protocols in place, such as access control, surveillance systems, and response force capabilities. 3. Simulation Engine: - Event Simulation: The core of the tool is a simulation engine that executes the modeled scenarios. It simulates the interactions between adversaries and the PPS, taking into account the timing, probabilities of detection, and response actions. - Stochastic Processes: Many tools use stochastic processes to account for the inherent uncertainties in both adversary behavior and security responses. This allows for the modeling of random events, such as detection failures or delays in response. 4. Data Analysis and Metrics: - Performance Metrics: The simulation generates various performance metrics, such as the probability of detection, probability of neutralization, and overall effectiveness of the PPS against the simulated attacks. - Statistical Analysis: Users can analyze the results statistically to understand the likelihood of success for different attack scenarios and the effectiveness of various security measures. 5. Visualization and Reporting: - Graphical User Interface (GUI): Most simulation tools feature a user-friendly GUI that allows users to visualize the facility layout, adversary movements, and security responses in real-time. - Reporting Tools: After running simulations, the tools often provide detailed reports summarizing the findings, including strengths and weaknesses of the PPS, recommended improvements, and potential vulnerabilities. Once the scenarios have been simulated, a statistically significant number of times the overall probability of neutralization can be determined. It is important that while the number of simulations must be statistically significant, the combat simulations should be based on identified and credible adversary attack scenarios that have been developed. Credible adversary attack scenarios should be developed within the bounds of the DBT and the capabilities that are present and can be used by the response force. 27 3.4. Identifying the Security Margins A necessary component in a performance-based framework is margin in performance to consequences. Fundamentally, a standard in physical security is the planning and application of compensatory measures, a baseline application of DID, and a margin to consequence. In this framework, this margin can also be manifested and quantified in two additional ways: (1) DID of the PPS and (2) timeline margin within the site security timeline to consequence. 3.4.1. Compensatory Measures One of the many factors considered in the design of a PPS is the proper design and placement of compensatory measures into the PPS. Compensatory measures are additional security measures to include technologies, procedures, and policies that are put in place to mitigate vulnerabilities due to primary PPS component failures. According to 10 CFR 73.55, licensees must: • Implement compensatory measures upon discovery or suspicion that any access control device may have been compromised e • Implement compensatory measures that ensure the effectiveness of onsite physical protection program when there is a failure or degraded operations of security-related components f • Identify criteria and measures to compensate for degraded or inoperable equipment, systems and components; compensatory measures must provide a level of protection that is equivalent to the protection that was provided by the degraded or inoperable equipment, systems or components; compensatory measures must be implemented within specific timeframes necessary. g Additionally, DG-5076 describes that compensatory measures should be applied for IDSs and intrusion assessment systems. Additionally, DG-5076 states: • “The licensee shall establish and implement timely compensatory measures for degraded or inoperable security SSCs to meet the requirements in 10 CFR 73.100. Compensatory measures shall provide a level of protection that is equivalent to the protection that was provided before the degradation or inoperability of the security systems, equipment, or components.” • “The licensee shall document processes and procedures and maintain records for implementing the corrective actions; compensatory measures; and maintenance, inspection, testing, and calibration of security SSCs.” When designing a PPS, compensatory measures are a key factor in the overall designing and planning for a PPS. Compensatory measures can be implemented with additional technologies or through the use of onsite security personnel. For example, when a sensor at the perimeter of a facility is taken offline for maintenance, loses power, or fails for any reason, a compensatory measure is put in place which could include: • Using an additional line of sensors that performs at the same probability of detection as the first line of sensors. (In application, this looks like two lines of detection along the perimeter if regulatory requirements only specify one line of detection). • Placing a member of the security team (i.e., armed security officer at the location) to attack as a form of detection at the location. • Using CCTV camera or pan-tilt-zoom cameras overlooking the down sensor location with a member from the alarm station to constantly watch the area. e. 10 CFR 73.55 (g)(6)(i)(c) f. 10 CFR 73.55 (n)(1)(v) g. 10 CFR 73.55 (o) 28 Compensatory measures must also exist for the failure of assessment components. These compensatory measures traditionally require the use of personnel as assessment in locations where cameras are lost. Additionally, compensatory measures are designed for delay, response, and communication within a PPS. The following are potential security measures that can be put in place for delay, response, and communication: • Delay - Movable barriers that can be placed when a barrier fails - Using an armed responder in the place of an open door - Using large vehicles in place of active or passive vehicle barriers - Using movable vehicle barriers in the place of active or passive vehicle barriers • Response - Multiple responders with fields-of-fire over areas around the perimeter and within the nuclear power facility - Use of offsite responders or local law enforcement to supplement the onsite responders • Communications - Multiple lines of communication that are tested frequently between the CAS, onsite responders, offsite response force, the backup alarm station, and local law enforcement agencies. Compensatory measures are used to ensure that the security system maintains its effectiveness when the primary components, as listed in the security plan, fail to operate. However, to verify that compensatory measures are effective, they must be tested on a frequent basis to ensure that they can be implemented in a timely manner and retain their effectiveness over time. Once a primary component or feature of the PPS fails, the compensatory plans should be implemented immediately and remain in place until the primary component is available and is confirmed to perform and operate correctly and effectively. 3.4.2. Defense-in-Depth Analysis The proposed 10 CFR 73.100 provides some requirements pertaining to DID: 73.100 (b)(2)(ii): Provide defense in depth in achieving performance requirements through the integration of engineered systems, administrative controls, and management measures. 73.100 (b)(3)(iv) Security Response: The physical protection program must be designed to provide timely security response to interdict and neutralize adversary attacks up to and including the design basis threat of radiological sabotage. The physical protection program must be designed to provide layers of security response, with each layer assuring that a single failure does not result in the loss of capability to neutralize the design basis threat adversary. Structures, systems, and components relied on for delay functions must be designed to allow for timely security responses to adversary attacks with adequate defense in depth. h Thus, the proposed requirements indicate that detection, delay, and response elements exhibit DID, and DID for security has not been well defined. Taking the above language at its word, the offsite response to any site could be interpreted to meet the requirements of response DID to neutralize the adversary. However, a robust understanding of DID needs to be mutually understood to appropriately scale the performance-based framework to prevent overly conservative assumptions of DID equating to redundancy. h. Underline added for emphasis 29 It would be the most appropriate to provide a flexible framework for licensees to choose their own methods to demonstrate DID. Below are several proposed example methods that would meet the intent of DID within the security space for elements of the PPS. Licensees could apply one or a combination of the following methods to demonstrate DID for the elements of the PPS. The intent is that elements involved in the demonstration of the site security timeline have DID, which is consistent with the proposed rulemaking § 73.100(b)(2)(ii) and (b)(3)(iv). 3.4.2.1. Redundancy Method The most conservative method to demonstrate DID is redundancy. Though this may be impractical for some aspects of the PPS (e.g., a redundant armed responder at every post), a licensee may demonstrate DID through redundancy for some elements of the PPS. Redundant detection and assessment hardware, alarm stations, locks, or delay barriers may factor within the planned PPS for other reasons (e.g., easy compensatory measure recovery, operational program execution) and may be credited for their contribution to DID. This simple method is not performance-based but provides a simple way to demonstrate compliance. 3.4.2.2. Sensitivity Analysis Method Current methods of demonstration of DID within security use a sensitivity analysis. Individual elements of the PPS are deleted, and the impact of their loss is realized through modeling and simulation, tabletop exercises (TTXs), or limited-performance testing. The PPS is tested to see if it is still effective without that PPS element. This method of DID demonstration is also conservative, as loss of PPS elements would be addressed through compensatory measures, and security should not consider random failures during security events. 3.4.2.3. Risk-Informed and Performance-Based DID Adequacy Method Within this performance-based framework, sensitivity analysis can be quantified and provides a method to risk-inform DID determination. Sensitivity analysis could highlight those elements of the PPS whose loss contributes the greatest negative impact to the security of the site. In the framework proposed here, the impact could be quantified by changes to the site security timeline. In this way, the loss of those PPS elements would be quantified, and the application of DID elements could be risk-informed according to the time contributions. Those PPS elements that contribute longer extensions to the site security timeline could have DID applied to reduce the impact of the PPS element. In this way, security could be aligned with the methods endorsed by NRC in RG 1.233 of the Nuclear Energy Institute (NEI) 18-04 Rev. 1, Licensing Modernization Project (LMP) methodology. The LMP defines DID adequacy criteria, as demonstrated in Figure 16 from NEI 18-04 Rev. 1, Figure 5-2, based on plant capability DID, programmatic DID, and a risk-informed and performance-based evaluation. 30 Figure 16. NEI-18-04 Rev. 1 framework for establishing did adequacy. This DID adequacy can be adapted for the performance-based physical security framework. Plant Capability DID Plant capability DID can be equated to the normal security program system effectiveness and the robust expectation of performance to protect against the DBT of radiological sabotage. The standard established for the execution of a security program for a commercial nuclear power reactor, with oversight from the regulatory, and evaluated through limited-performance testing and FOF exercises provides a strong justification for plant capability. Risk-Informed and Performance-Based Evaluation of DID The risk-informed and performance-based evaluation of DID can be accomplished through a risk- informed performance-based security sensitivity analysis. The sensitivity analysis can result in a quantified contribution to the extension of security timelines that allows the licensee to risk-inform that application of DID and addition of DID barriers. NEI 18-04 Rev. 1 uses the individual risk values for the LMP DID evaluation. It should be noted that the LMP does not seek absolute risk avoidance but instead a reasonable evaluation of risk within the analysis and the application of additional SSCs and programmatic application to keep risk within performance targets. Similarly, instead of individual event risk values, as used in NEI 18-04 Rev. 1, the security sensitivity analysis would evaluate the impact of loss on the site security timeline and the reasonable extension of that timeline. If the analysis determines an unacceptable extension of the timeline, then additional security SSCs or programmatic elements should be applied to reduce the resulting DID site security timeline to within acceptable bounds. For the security sensitivity analysis, security SSCs can be grouped into PPS elements: • Loss of each detection line (both for exterior and interior detection zones) - Assumed 0% detection in that zone with no notice of loss or penetration to CAS or Secondary Alarm Station (SAS) 31 • Loss of each assessment system - With a detection, the associated assessment for that detection zone is unavailable • Loss of fidelity of each security gate or security door within a security barrier - Along the path of adversary travel • Loss of each armed responder • Loss of each communication system. Evaluating an individual PPS element limits the analysis to integrated systems that serve individual functions within the PPS. The loss of those PPS elements would represent a bounding analysis for that entire system segment. Each PPS element should be analyzed unless that PPS element demonstrates DID in another manner. Each PPS element would be analyzed with a bounding security event (or events) that most challenges the loss of the security elements and timeline (e.g., adversaries traverse that lost detection line, through that open doorway for the event with the shortest timeline to irreversible release time). The additional time associated with the loss of a PPS element should be within 10% of the bounding event irreversible release timeframe (this would apply only for the DID analysis). This part of the DID analysis would not consider partial failures of PPS elements or the application of compensatory measures. In this way, the risk-informed and performance-based evaluation of DID portion of Figure 16 for security would be conservative. The identification of additional PPS elements as the result of the risk-informed and performance- based evaluation of DID would be identified to reduce the DID security timeline to less than the irreversible release timeframe and result in DID adequacy. Programmatic DID Programmatic DID is described in NEI 18-04 Rev. 1 that it “provides a means to incorporate special treatment i while designing, manufacturing, constructing, operating, maintaining, testing, and inspecting the plant and the associated processes to ensure there is reasonable assurance that the predicted performance can be achieved throughout the lifetime of the plant.” Programmatic DID scales treatment (e.g., inspections, testing) to the risk associated with the component. Security programs have previously applied conservative prescriptive requirements to the testing requirements associated with security SSCs. Under the performance-based framework, performance-based testing frequencies and success criteria could be used by the licensee to propose alternative risk-informed and performance-based criteria based on the risk contribution according to the sensitivity analysis. Those security elements with DID security timelines greater than the irreversible release timeline should have additional programmatic requirements to ensure functionality. This may be fulfilled by increased maintenance, testing, for SSCs, or higher performance requirements or training for programmatic PPS components. Programmatic DID can also be a function of the application of compensatory measures. Capability for security is already regarded as high, given the expectation of preventing radiological sabotage, and performance of detection, delay, and response elements. The security program is required to consider random failures and deficiencies and apply temporary measures to meet the security function in a degraded condition. Compensatory measures are applied to any function of the security program in a degraded condition and allows flexible temporary resolution of these issues. The security program applies compensatory measures in a DID manner to fulfill the function of reasonable assurance against radiological sabotage. i. According to RG 1.201, “…special treatment refers to those requirements that provide increased assurance beyond normal industrial practices that structures, systems, and components (SSCs) perform their design-basis functions.” 32 3.4.3. Timeline Margin Following the explanation of irreversible release timing provided in Section 3.3.1, the concept of security margin is given here. Given the security timeline shown in Figure 17 where there is a time delay between the target sabotage event and the irreversible release timing, the security margin can be defined as the time difference between when security event ends and the irreversible release timing. The larger this margin is, the more confidence can be had that an undesirable release can be prevented considering all the uncertainties in the attack and response scenarios. In a single attack scenario, the margin is a singular value. However, if this scenario is repeated multiple times accounting for uncertainties in the attack speed, detection probability, delay period, communication and assessment, and neutralization probability, the margin will have an uncertainty band in which some of the margin may be negative indicating the failure of physical protection. The security margin is a useful concept in the risk-informed performance-based security framework because it informs licensees on various flexible ways to meet the required timeframe of the irreversible release timing. For example, licensees may increase margin and therefore protection confidence by: 1. Delaying the irreversible release timing. This may be done by utilizing preventive safety measures after a component is sabotaged to slow down the progression of process events that leads to irreversible damage, for example by derating reactor power or putting it in a hot standby mode, thereby shifting irreversible release timing to the right in Figure 11. 2. Improving security response capabilities to shift the ending of security event to the left of Figure 11. This may be done by refining strategies and positioning for a faster neutralization thereby shortening the armed engagement period, reducing response team travel time for a faster deployment that also helps with strategic positioning prior to engagement, reducing intrusion assessment time, and/or sensor positioning to allow earlier detections. These various methods can be implemented in combination with each other and be tested using computer modeling and simulations for many number of times to ensure a sufficient security margin is achieved. A goal for the security program is a percent of margin compared to the irrevesible release timeline of at least 10%. Such a goal provides confidence in the site protective strategy for protecting against the DBT of radiological sabotage. An example would be a irreversible release timing of 60 min. The site protective strategy results in neutralization of the adverary at 40 min, giving a margin of 20 min, or 33% (20 min/60 min). 33 Detection Attack Detection Initiated Assessment Communication Initiated Adversary Physical Protection System Delay Adversary Disables Target(s) Irreversible Release Timing Response Communication Received Response Initiated Armed Response Engages End Security Event (neutralization or AIPT) Irreversible Release Timeline Site Security Timeline (neutralization or AIPT) Margin Figure 17. Security timeline highlighting the time margin. 3.5. Formulating a Performance-Based Alternative The following is a demonstration of a performance-based method of compliance consistent with the above approach. This method is applicable for the determination of the performance-based licensing of PPS detection, delay, and response, either as individual elements or for an entire PPS. 3.5.1. Target Sets The licensee will determine the target sets through the target set determination process, as described in RG 5.81 Rev. 1 or DG-5071 as Rev. 2 when published. Allowed operator actions will be applied to determine the list of achievable target sets. This will include the determination of the timeline associated with each target set. Target sets that are deemed the most desirable and those with the shortest timelines would likely be the best candidates to evaluate for the licensing basis. It is noted that all achievable targets would fall within the requirements for protection and may be selected for demonstration during an audit or inspection action by the regulator. 3.5.2. Design Physical Protection System Detection and Communication The PPS design of the detection, assessment, and communication will support determining the communication to offsite response organizations. Once the detection and communication timeline is established, the AIPT can be determined. Iterations on the design of SSCs and procedures can lead to updates to the AIPT and should be captured through the PPS design. 34 3.5.3. Offsite Response Timeline The site detection and communication and offsite response timeline will need to be determined to establish the site AIPT, as well as any event-specific security bounding time to support PPS design. The AIPT would be determined according to the guidance given in DG-5072. 3.5.4. Licensing Basis Security Scenarios The initial PPS design can be started early in the design phase, before full target set characterization consistent with RG 5.81. Initial delay values (not crediting any programmatic response) can, with the target set timelines, determine the initial values for the irreversible release timing. The design of the PPS using the Design Evaluation and Process Outline (DEPO) methodology will then update the irreversible release timing associated with the target sets with each iteration of the PPS design. The licensing basis security scenarios can be determined from the irreversible release timing from each individual target set and then down-selected given the following considerations: • Shortest target set time to release (as determined as part of the target set identification process). • Most desirable target sets. • Directional attacks from different sides of the facility in manner best for adversary (e.g., stealth, shortest path, greatest cover or concealment). This would inform bounding values for adversary traversal times through passive security barriers and task times. • Attack methods that employ variations to the DBT, to include number, function, and methods of adversaries, as well as the use of applicable vehicles and explosives, cyberattacks, and insider actions. This would inform bounding values for adversary traversal times through passive security barriers and task times. Ten should be the minimum number of representative irreversible release timeline scenarios to form the licensing basis security scenarios. The scenarios should be representative of different DBT adversary attributes, targets within the facility, and methods. This number of representative security events should be increased when accounting for: • Larger facilities (i.e., 300 MWe to 1 GWe +) that have a greater number of adversary pathways • Multimodule facilities • More than two radiological sabotage targets The resulting list of licensing basis security scenarios and the associated irreversible release times will be representative of the PPS response for the purposes of licensing. As part of the PPS design, the parameters and programmatic requirements will be determined for the detection probability and confidence. The initial maintenance and testing requirements should be determined according to manufacturing requirements and supported by available operational data. As site-specific operational data is gained, site maintenance and testing requirements should be updated to support the licensed performance parameters associated with detection probability and confidence as necessary through the lifetime of the facility, as well as the training and qualification requirements to support probability of neutralization. 3.5.5. Validation Validation of the PPS is an upfront and ongoing task that must be completed by a licensee during the licensing process and operation of the facility. As defined in the DEPO methodology, there are many processes and steps that are used in validating a PPS. This includes, but is not limited to, component testing, TTX, modeling and simulation, LSPTs or ARAPT, FOF exercises, and finally mitigating 35 potential vulnerabilities or failures through corrective action programs. These processes should work together so that components, subsystems, and the entire PPS are evaluated for effectiveness Whole Component Subsystem System Testing Testing Testing Figure 18. Process for system evaluations. 3.5.5.1. Component Testing Component testing is testing of individual security technologies, personnel, or procedures that are used within a PPS. Generally, testing of personnel and procedures is considered LSPTs as well. During the design and subsequent operation of a facility acceptance testing and performance testing, respectively, should be conducted. When considering licensing of a nuclear facility, acceptance testing plays a key role in determining whether the security technologies described within the security plan meet the effectiveness of regulatory requirements and perform at a level that ensures the PPS will operate at its designed effectiveness level. Acceptance testing considers evaluating the components of a PPS for the relevant functions the component is to perform. For example, exterior sensors should be evaluated for their overall probability of detection for the DBT, the nuisance alarm rate, and false alarm rate. These tests will confirm that the component can operate within the regulatory requirements and ensure that when the component is placed into the operational PPS it will function as stated. Once the equipment is installed, these acceptance tests are conducted to ensure regulatory compliance and compliance with the security plan developed for the facility. For guidance on conducting these tests, please refer to Section 3.2.2 and 3.2.3. Additionally, a component level test may also include evaluation of the response force’s ability to specifically engage adversaries during a security event. These tests can be conducted by using weapon platforms that the response force can use within its system and test the overall portability of hit (i.e., the probability that the responder can hit the target at a certain range over a specified number of rounds being shot at the target). These tests can be used to ensure that the responder training is adequate to engage the adversary force and that the responders can successfully engage an adversary at a given range within the contingency plan that is developed by the site. 36 3.5.5.2. Limited-Scope Performance Testing LSPTs can be used to evaluate people and procedures used within the PPS. An LSPT can be used to evaluate individual components of a process or the process itself that is being used as part of the PPS or the PPS strategy. LSPTs can be beneficial in evaluating portions of the response time, the process for conducting contraband searches, the process for conducting vehicle searches, and processes for allowing personnel and vehicles into the facility once they have been searched. An LSPT may require an armed responder to pick up their weapon and load the ammunition magazine into the rifle. This LSPT confirms that the armed responder is knowledgeable about loading their weapon and the time is recorded to gather the total task time that is used in the overall response timeline to respond to a nuclear security event. Additionally, these tests can be used to evaluate the ability of an armed security officer to use a handheld metal detector to conduct contraband searches of individuals as they enter the facility. Another category of this evaluation is ELSPTs that are used to test multiple parts of a plan, procedure, and technologies and encapsulates subsystem testing of the PPS. An example of an ELSPT could encompass activating a sensor in the field, ensuring the sensor alarms in the CAS, ensuring the CAS operator assesses the cause of the alarm, ensuring the CAS operator communicates the cause of the alarm, and finally timing and ensuring accuracy of these alarm causes. Another example is using a radio dispatch from the RTL to the response force, the response force getting their equipment ready, moving to respond to the radio callout, and timing and ensuring adequacy in every step of the process. These tests are used to ensure that the times used to implement a response to a nuclear security event and ensure all functions of the response to a security event are within the specified time limits to implement an effective response. 3.5.5.3. Path Analysis Path analysis tools are used to evaluate the total adversary attack timeline and the response timeline and determine the probability of interruption for an identified adversary attack scenario. Path analysis tools use much of the information that is collected through component level tests and LSPTs. This information is used to identify whether ample time exists for the response force to interrupt an adversary force before they can achieve their malicious act. Path analysis software tools can be used to identify the probability of interruption for various adversary attack vectors such as the most vulnerable path (the path with the lowest probability of interruption), the fastest path (the path with the shortest over all adversary task timeline), and the stealthiest path (the path where the adversary has the lowest cumulative probability of detection). By using path analysis tools, a facility can ensure that its probability of interruption meets design standards and the response force can adequately interrupt the adversary force. These identified adversary pathways can also be used to inform other validation and evaluation tools, such as TTXs, combat simulations, and FOF exercises. 3.5.5.4. Tabletop Exercises and Combat Simulations TTXs can be used to evaluate and validate various areas of a PPS. A TTX can be used to evaluate the site’s response strategies, communications, and procedures against identified and outlined adversary attack scenarios. TTX tools exist within three-dimensional software tools that can be used to record a TTX or can be carried out using facility blueprints and maps on paper. 37 Figure 19. Screenshot of Scribe3D TTXs recorder tool. j These tools are used to assess how the PPS will respond to a security event at a facility. A TTX generally starts with a pre-identified adversary attack scenario that may have been developed or informed by the results of path analysis or developed by a mock adversary team. It is important that the adversary attack scenario developed for a TTX be consistent with the DBT and any other regulatory bounds for the adversary team. The TTX is conducted in a turn-based fashion, allowing the adversary team and response team to make moves according to the adversary attack scenario and the contingency plans developed for the facility. Because a TTX is conducted using these setup adversary attack scenarios and the contingency plans developed for the facility, these exercises can be used to evaluate the overall effectiveness of the developed security plans and contingency plans that have been implemented. These exercises also can be used to evaluate the communications that are used during a security event and how the communication works between groups at the facility. TTXs can be a valuable way for the facility to evaluate that its security and contingency plans are effective against an adversary attack scenario. A TTX can be very beneficial when the facility’s data is used in the TTX. This includes adversary task times, probabilities of detection, and response force probabilities of the hit for the weapon platforms they are using. By conducting these component level tests and subsystem level tests, the data can be used to ensure accuracy in a TTX. Another tool that can be used to evaluate or validate a PPS is combat simulation tools. Combat simulation tools can be used to simulate adversary attack scenarios and the PPS response to those adversary attack scenarios and include simulated combat engagements between the adversary team and the response team. Combat simulation tools can ultimately be used to calculate the probability of neutralization for a PPS. Combat simulation tools utilize the data that is generated from continuous testing as well as human behavior modeling to determine the outcome of a particular security event that occurs at the site. The combat simulation tool is used to evaluate the developed adversary attack scenarios in a Monte-Carlo fashion over a statistically significant total number of scenarios to determine the overall probability of neutralization. Combat simulation tools can also be used to evaluate the effectiveness of a contingency plan and the security plan for the facility to ensure these plans are adequate at mitigating and defending the facility against an adversary attack. A listing of current NRC technical letter reports on modeling and simulation tools can be found in Appendix A. j. https://modsimtools.sandia.gov/scribe3d/ 38 3.5.5.5. Force-on-Force Exercises One of the final and ultimate evaluations that should be conducted by a facility are FOF exercises. A FOF exercise is a realistic scenario-based drill with a trained mock adversary force that attempts to attack the facility according to a developed adversary attack scenario and the response of the PPS to defend against the adversary attack scenario. These exercises use simulated weapons and simulated explosives for both the response force and adversary force. These FOF exercises are meant to evaluate the facility’s security plan and contingency plans in live action and determine whether the plans, procedures, technologies, and personnel used to implement these plans are effective. FOF exercises are recommended to be the final evaluation tool that is used to ensure the system is effective at mitigating adversary attack scenarios. FOF exercises can be very costly to implement at nuclear facilities due to the need for a response team to participate in the exercise, a response team that is needed to actively defend the facility against adversary attacks at the nuclear facility, controllers for the FOF, and evaluators for the FOF. Because of this, it is recommended that FOFs be conducted based on other types of analysis that are conducted. For example, during the evaluation process discussed above, the use of TTXs and combat simulation tools may show some adversary attack scenarios that the system has a low probability of neutralization to defend against, an adversary pathway with a significantly low probability of interruption, or that the tools created a new adversary attack scenario that the facility has not evaluated. These scenarios, if utilized in an FOF, can help ensure that the PPS and its associated plans and measures are effective at protecting the facility against the most concerning adversary attacks that the facility may face. Using this process and methodology for a structured approach to testing and data collection can ensure that an FOF encapsulates all the known information about the facility and protective strategies to evaluate the effectiveness of the security system. 3.5.5.6. Corrective Action One important factor of the evaluation and validation process for a PPS is the compensatory measure and corrective action process that is put in place to mitigate vulnerabilities or deficiencies within the PPS. A corrective action is implemented to mitigate a vulnerability or deficiency in a process, procedure, technology, or plan for the site. It is important to note the difference between compensatory measures and corrective actions that are taken based on the vulnerability or deficiency that is seen through the evaluation and validation process. An example of this process is when a microwave sensor fails to ensure the required probability of detection at the required confidence level during a performance test. An immediate compensatory measure should be put in place to mitigate any potential vulnerabilities due to the failure of this sensor. The compensatory measure could include placing an armed security officer at the location of the microwave, or the site could task a pan-tilt-zoom or fixed camera to the location where an alarm station operator could be used to backfill for the detection. While these measures may be implemented by the facility to counter a potential vulnerability temporarily, the facility must correct the deficiency to ensure the effectiveness of the PPS. This requires the facility to implement a corrective action to permanently fix the deficiency. In this example, the corrective action is to conduct maintenance or replace the microwave sensor in the field. Compensatory measures and corrective actions can also be implemented for personnel and procedures that are used within the security plan and contingency plan. An example is when an armed security officer fails to conduct a vehicle search correctly when a LSPT is conducted. A compensatory measure for this action may be to replace the armed security officer that failed the test with another qualified officer that has not failed the test. The corrective action for this example may be to conduct training with the armed security officer and conduct another evaluation after the training to ensure the officer correctly conducts the vehicle search. It is important that the facility implements compensatory measures and corrective actions in a timely manner once identified. Vulnerabilities and deficiencies can be identified through any of the above- 39 mentioned evaluation and validation steps. By implementing compensatory measures and corrective actions in a timely manner, the facility can ensure it adequately mitigates vulnerabilities to the facility. 3.5.6. Security Margin To support the parameters determined for the PPS, the margin of the PPS is characterized through DID and timeline margin. These analyses support the determination of acceptability of the proposed design to regulatory requirements associated with performance and DID. 3.5.6.1. Defense-in-Depth Analysis The DID of the designed PPS should be selected by the licensee for each component of the PPS. The three methods of DID demonstration may be applied by a licensee to different elements of the PPS to provide a holistic view of PPS DID. For example, some elements (e.g., alarm stations, communication) may be best addressed through functional redundancy. These SSCs and programmatic elements may have redundancy built into the design and operation of the facility. Some licensees may choose to have full detection redundancy to address possible issues with compensatory measure implementation. Some security program elements may best demonstrate DID through a sensitivity analysis. This may be the best method for passive delay features or response elements. The performance-based method of DID demonstration may take additional resources to justify design decisions but provides a pathway to demonstrate the least conservative approach for compliance. The application of programmatic DID can provide methods to apply special treatments to those security program elements identified as the most sensitive in the DID analysis and provide a performance-based method to apply increase performance-applications. 3.5.6.2. Timeline Margin For each of the irreversible release timelines compared against the site security timeline, the margin should be characterized. Lesser margin can be licensed into the facility to provide operational flexibility in meeting the required timeframe to prevent radiological sabotage. 3.5.7. Documentation It would be necessary for the licensee to document the analyses that support the parameter determinations. Such analyses should be included within a technical report, available for regulatory inspection as part of a license determination audit, and maintained at the licensed site for inspection throughout the license. The parameters of performance (bounding irreversible release timings, site security timelines, detection probability and confidence, and probability of effectiveness) should be recorded within the site security plan as commitments of the license. In this way, the licensee will have flexibility to meet the licensed performance parameters. This would allow smaller changes to occur without regulatory notification, changes that impact descriptions within the security plan to easily occur under the 10 CFR 50.54(p) or the proposed § 53.1565(d)(4), framework, and changes that impact the performance of the security plan to require approval under a § 50.90 or the proposed § 53.1510license change process. 4. CONCLUSION The NRC has developed proposed physical security performance-based rulemakings that provide the opportunity for a new method of demonstrating compliance for nuclear security at commercial power reactors. The performance-based rulemakings will allow for greater flexibility by a licensee to meet the high-level goals of protection of the public health and safety from the consequences associated with the DBT of radiological sabotage. The performance-based security licensing basis determination framework proposed here provides a clear and traceable method to demonstrate and maintain compliance while 40 allowing operational flexibility. This framework also provides a method to demonstrate the justification for alternative methods of compliance with the prescriptive requirements currently found in § 73.55. 41 5. REFERENCES A. Evans, A. Neef, B. Stromberg, C. Evans. “U.S. Domestic Microreactor Security by Design.” SAND2021-13779 R. Garcia, M. L, “The Design and Evaluation of Physical Protection Systems,” Second Edition, Boston: Butterworth-Heinemann, 2008. Nuclear Energy Institute, “Risk-Informed Performance-Based Technology Inclusive Guidance for Non- Light Water Reactor Licensing Basis Development,” NEI 18-04 Rev. 1, August 2019. Nuclear Regulatory Commission, “Alternative Physical Security Requirements for Advanced Reactors,” NRC-2017-0227-0038 (August 5, 2024). Nuclear Regulatory Commission, “Law Enforcement Response in Power Reactor Physical Protection Programs,” NRC-2024-0167-0001 (December 9, 2024). Nuclear Regulatory Commission, “Part 53—Risk Informed, Technology-Inclusive Regulatory Framework for Advanced Reactors,” NRC-2019-0062 -0012 (October 30, 2024). Nuclear Regulatory Commission, “Part 73 – Physical Protection of Plants and Materials,” page last reviewed/updated August 14, 2024, accessed August 28, 2025, https://www.nrc.gov/reading-rm/doc- collections/cfr/part073/full-text.html. Nuclear Regulatory Commission, DG-5071, “Target Set Identification and Development for Nuclear Power Reactors,” Draft RG 5.81 Rev. 2, July 2024. ML22021B529 (OUO-SRI) Nuclear Regulatory Commission, DG-5072, “Guidance for Alternative Physical Security Requirements for Small Modular Reactors and Non-Light-Water Reactors,” Draft RG 5.90, July 2024. ML20041E037 Nuclear Regulatory Commission, NUREG/BR-0303, “Guidance for Performance-Based Regulation,” December 2002. ML023610315 Nuclear Regulatory Commission, RG 5.81 Rev. 1, “Target Set Identification and Development for Nuclear Power Reactors,” December 2019. ML19253C754 (OUO-SRI) 42 Appendix A Publicly Available NRC Reports on Modeling and Simulation Tools Nuclear Regulatory Commission, “Preliminary Assessment of Physical Protection Modeling and Simulation Tools,” TLR-RES/DE/REB-2023-09, September 2009. ML23346A027 Nuclear Regulatory Commission, “Integration of Safety, Security, and Safeguards During Design and Operations—A Technical Assessment and Regulatory Considerations for Advanced Reactor and Advanced Fuel Fabrication Facilities,” TLR-RES/DE/REB-2024-16, SAND2024-12397R, September 2024. ML24275A075 Nuclear Regulatory Commission, “Publicly Accessible Reports Assessment and Demonstration of Physical Security Modeling and Simulation Tools,” TLR–RES/DE/REB–2025–06, March 2025. ML25090A245 1