Radar Vectors
Proposed Rule Proposed Issued Apr 15, 2019 149 pages Copy saved Nov 17, 2026

84 FR 15296 Streamlined Launch and Reentry Licensing Requirements

This rulemaking would streamline and increase flexibility in the FAA's commercial space launch and reentry regulations, and remove obsolete requirements. This action would consolidate and revise multiple regulatory parts and apply a single set of licensing and safety regulations across several types of operations and vehicles. The proposed rule would describe the requirements to obtain a vehicle operator license, the safety requirements, and the terms and conditions of a vehicle operator license.

This document is proposed. It may have been replaced. Check the official source before relying on it.

Official source ↗ Official PDF ↗

↑ Earlier in this document

I. Registration of Space Objects

Proposed § 450.217 would deviate from current Sec. § 417.19 and 431.85 by requiring the registration of foreign-owned space objects. The FAA would not require the licensee to determine the owner's nationality. The Department of State would use this information to ensure that other nations meet their obligations by registering their foreign objects. Proper registration of all objects owned by foreign entities would allow for the protection of the United States from liability associated with these objects.

Otherwise, the FAA would retain the same informational requirements. It would continue to require a licensee to submit information about the space object's international designator, the date and location of the mission, the general function of the space object, and the final orbital parameters. Additionally, proposed § 450.217 would retain current § 431.85's requirement that an operator notify the FAA when it removes a space object.

J. Public Safety Responsibility, Compliance With License, Records, Financial Responsibility, and Human Spaceflight Requirements

J. Public Safety Responsibility, Compliance With License, Records, Financial Responsibility, and Human Spaceflight Requirements

The FAA is not proposing any substantive changes to the requirements specified below. However, the agency is proposing to consolidate these requirements into the new, proposed part 450; clarify that the consolidated requirements apply to any licensed launch or reentry; and make other minor, clarifying edits. The following is a summary of the proposed changes: 1. Public Safety Responsibility and Compliance With License

The FAA would consolidate the public safety responsibility requirements in current Sec. § 417.7 and 431.71(a) into proposed § 450.201 (Public Safety Responsibility). Also, the FAA would move the compliance requirement in current § 431.71(b) to its own section, proposed § 450.203, Compliance with License. Although the location of these requirements would change, the requirements themselves would not change.

Therefore, proposed § 450.201 would provide that a licensee is responsible for ensuring public safety and safety of property during the conduct of a licensed launch or reentry. Proposed § 450.203 (Compliance with License) would require that a licensee conduct a licensed launch or reentry in accordance with representations made in its license application, the requirements of proposed part 450, subparts C and D, and the terms and conditions contained in the license.

The proposed requirement for a licensee to conduct a licensed launch or reentry in accordance with representations made in its license application is the same, in substance, to Sec. § 417.11(a) and 431.71(b). Section 417.11(a) states that a launch operator must conduct a licensed launch and carry out launch safety procedures in accordance with its application. Section 431.71(b) states that a licensee must conduct a licensed RLV mission and perform RLV safety procedures in accordance with representations made in its license application. The fact that representations made in a license application become binding on a licensee is discussed earlier in this preamble.

The proposed requirement for a licensee to conduct a licensed launch or reentry in accordance with the requirements of proposed part 450, subparts C and D, is the same, in substance, to § 417.1(b)(2)'s treatment of part 417 requirements. Section 417.1(b)(2) states that the safety requirements of part 417, subparts B through E, apply to all licensed launches of expendable launch vehicles. Part 431 does not have a similar requirement because application requirements and safety requirements are interlinked, leaving uncertain the actual safety requirements under a license. Note that in subpart C, the application requirement paragraphs do not apply once a license is issued, unless a licensee applies for a modification.

The proposed requirement for a licensee to conduct a licensed launch or reentry in accordance with the terms and conditions contained in the license is the same, in substance, to Sec. § 415.9(b) and 431.71(b). Section 415.9(b) states that a launch license authorizes a licensee to conduct a launch or launches subject to the licensee's compliance with terms and conditions contained in license orders accompanying the license. Section 431.71(b) states that a licensee's failure to comply with any license condition is sufficient basis for the revocation of a license or other appropriate enforcement action. The FAA includes terms and conditions in a license to address license-specific requirements. Under the proposal, a licensee's failure to act in accordance with these items would be sufficient basis to revoke a license, or some other appropriate enforcement action. 2. Financial Responsibility

The FAA would consolidate the current financial responsibility requirements in Sec. § 417.21 and 431.81 into proposed § 450.205 (Financial Responsibility Requirements). Although the location of the requirements would change, the requirements themselves would not change.

As such, the proposed regulation would require a licensee to comply with financial responsible requirements as required by part 440, and as specified in a license or license order. 3. Human Spaceflight

The FAA would consolidate the human spaceflight requirements in current Sec. § 415.8, 431.8, and 435.8 into proposed § 450.207 (Human Spaceflight Requirements). The proposal would require a licensee conducting a launch or reentry with a human being on board the vehicle to comply with human spaceflight requirements as required by part 460 of this chapter and as specified in a license or license order. Although the location of the requirements would change, the requirements themselves would not change. 4. Records

The FAA would consolidate the current record requirements in Sec. § 417.15(a) and (b) and 431.77(a) and (b) into proposed § 450.219(a) and (b). However, the FAA would replace the terms "launch accident" and "launch incident" in § 417.15(b) and the terms "launch accident," "reentry accident," "launch incident," and "reentry incident" in § 431.77(b) with "class 1 or class 2 mishap." As discussed in more detail earlier in this preamble, the FAA proposes to replace current part 401 definitions involving "accident," "incident," and "mishap" with specified mishap classes.

The proposed regulation would require an operator to maintain, for 3 years, all records, data, and other material necessary to verify that a launch or reentry is conducted in accordance with representations contained in the operator's application, the requirements of subparts C and D, and the terms and conditions contained in the license. To satisfy this requirement, the FAA expects an operator to keep a record of the actual conditions at the time of flight and any deviations outside of the flight commit criteria as specified in the current § 417.113(c). Similar to current requirements, in the event of a class 1 or class 2 mishap, an operator would be required to preserve all records related to the event until the completion of any Federal investigation (which could be greater than 3 years) and the FAA has notified the operator that the records need no longer be retained. The operator would need to make all records required to be maintained under the regulations available to Federal officials for inspection and copying.

K. Applicability

K. Applicability

1. General

Proposed § 450.1 (Applicability) would state that part 450 prescribes requirements for obtaining and maintaining a license to launch, reenter, or both launch and reenter, a launch or reentry vehicle. As discussed previously, proposed part 450 would consolidate licensing requirements currently covered in parts 415, 417, 431, and 435. 2. Grandfathering

1. General

Under proposed § 450.1(b), proposed part 450 would not apply to any launch or reentry that an operator elects to conduct pursuant to a license issued by the FAA or an application accepted by the FAA prior to the effective date of proposed part 450, with two exceptions. The proposed requirements for collision avoidance analysis (COLA) and asset protection would apply to all operators subject to the FAA's authority under 51 U.S.C. chapter 509 who are conducting launches after the effective date of the new regulations. The FAA would determine the applicability of proposed part 450 to an application for a license modification submitted after the effective date of the part on a case-by-case basis.

The proposed regulations are more performance based, and many of the current requirements would serve as a means of compliance to meet the proposed regulations. As a result, activities authorized under the existing regulations would be authorized under the proposed regulations. The FAA proposes to allow an operator to operate under the current regulations (specifically, parts 401, 415, 417, 431, and 435) when conducting a launch after the effective date of new part 450 provided it holds a license or has had a license application accepted prior to the effective date of this regulation. Pursuant to Space Policy Directive-3 \184\ (SPD-3), proposed § 450.169 and proposed appendix A to part 450 would align the COLA criteria with current common practice and provide better protection for inhabitable and active orbiting objects. Additionally, § 450.101 would require that the probability of loss of functionality for each critical asset must not exceed 1 x 10-\3\ to protect national assets. For that reason, the FAA is proposing that all operators would be required to comply with these two provisions on this rule's effective date.

\184\ Space Policy Directive-3, National Space Traffic Management Policy, 83 FR 28969 (June 21, 2018).

Because many of the current regulations would serve as a means of compliance for the proposed regulations, the FAA would review license modifications that applied the current regulations as means of demonstrating compliance with the proposed regulations. Additionally, an operator could use a means of compliance other than the current regulations to demonstrate compliance in a license modification request. The FAA would determine the applicability of proposed part 450 to an application for a license modification submitted after the effective date of the part on a case-by-case basis. The FAA does not anticipate that a vehicle operator would have any greater difficulty meeting the requirements under the proposed regulations than under the existing regulations. In fact, the FAA believes that the proposed regulations are more flexible because most allow for many different means of compliance.

An applicant for a renewal would be required to meet all the requirements of proposed part 450. The FAA anticipates that this would not be burdensome for operators seeking license renewals because there would be few, if any, additional application requirements that could not be fulfilled by reference to previously submitted information.

L. Equivalent Level of Safety

L. Equivalent Level of Safety

In addition to developing performance-based requirements, this proposal would preserve the equivalent-level-of-safety flexibility by relocating the provision to proposed § 450.37. Unlike using a means of compliance, which requires demonstration of compliance with a performance-based regulation, the ELOS provision would continue to allow an applicant to propose an alternative method to meet the safety intent of a current regulatory requirement. For example, § 450.117(d)(3) would require representative normal flight trajectory analysis outputs for each one second of flight. An applicant may wish to request an ELOS determination to the one-second interval, and the FAA would likely accept it if an alternative interval provides smooth and continuous individual PC contours.

To demonstrate equivalent level of safety, an operator would provide a clear and convincing demonstration, through technical rationale, that the proposed alternative approach provided a level of safety equivalent to the requirement it would replace. An ELOS determination means an approximately equal level of safety as determined by qualitative or quantitative means. Under § 450.37(b), an operator would not be able to use an ELOS determination to replace the public risk criteria set forth in § 450.101.

In 2018, the FAA issued a final rule that expanded the option to satisfy commercial space transportation requirements by demonstrating an equivalent level of safety in order to provide more choice to operators and reduce the number of waivers that must be prepared by industry and processed by the government.\185\ To utilize the option, operators are required to demonstrate that they are achieving a level of safety equivalent to any safety parameters specified in the regulations. The FAA evaluates every request for an alternative means of regulatory compliance under the ELOS provisions to ensure that the safety of the public, property, or any national security or foreign policy interest of the United States is maintained to be consistent with the requirements in 14 CFR chapter III. The FAA would preserve the process established in the 2018 rulemaking, and would include its ELOS determination as part of any license issued applying this provision.

\185\ Updates to Rulemaking and Waiver Procedures and Expansion of the Equivalent Level of Safety Option, Final Rule, 83 FR 28528 (June 20, 2018).

The FAA requests comment on the potential use of "safety cases" when demonstrating an equivalent level of safety under proposed § 450.37. A safety case is a structured argument, supported by a body of evidence that provides a compelling, comprehensive, and valid case that a system is safe, for a given application in a given environment.\186\ The ARC report (at p. 25) suggested that FAA review time could be minimize if applicant submittals were "structured as a reasonable safety case that the proposed actions are safe under all plausible scenarios." In fact, the ARC suggested "safety cases" could be useful options several times. With respect to the proposed regulation, a safety case would potentially show that certain requirements identified by the applicant, excluding the requirements of § 450.101, need not be complied with per se in order to demonstrate that an alternative approach provides an equivalent level of safety to the requirements identified by the applicant.

\186\ This Safety Case definition is from the U.K. Ministry of Defence (MOD) Standard 00-56, "Safety Management Requirements for Defence Systems."

A-P-T Research, Inc., under contract to the FAA, recommended the use of a safety case approach as an alternate path to securing a license.\187\ The FAA considered proposing a safety case approach to demonstrating an equivalent level of safety under proposed § 450.37 that would include a formal proposal process that must use a means of compliance accepted by the Administrator, unless the Administrator determines otherwise based on predicted public risks and consequences, or demonstrated reliability.

The formal proposal process would: (1) Facilitate an FAA audit of all risk management methods proposed for use, including a demonstration of how the proposed methods can demonstrate compliance with § 450.101; (2) implement all the recommended improvements from the audit or justify all deviations from the recommended improvements; (3) document the risk management methods used and the verification evidence to demonstrate compliance with § 450.101; (4) facilitate an audit by an FAA-approved third party of the risk management methods used and the verification evidence to demonstrate compliance with § 450.101; and (5) submit the results of the third party audit for FAA review and approval.

An applicant that sought to use this safety case approach would need to submit: (1) A description of their plan to facilitate an FAA audit of all risk management methods proposed for use, including a demonstration of how the proposed methods can demonstrate compliance with § 450.101; (2) a description of the improvements implemented based on the FAA audit and detailed justifications for any deviations from the FAA recommended improvements; (3) a description of the risk management methods used and the verification evidence to demonstrate compliance with § 450.101; (4) an agreement to facilitate an audit by an FAA-approved third party of the risk management methods used and the verification evidence to demonstrate compliance with § 450.101; and (5) a description of the results of the third party audit. The safety case approach recommended by APT included the use of a third party to review. The FAA sees potential complications, including liability considerations, when involving a third party in the licensing process.

The FAA seeks comments on the potential usefulness and challenges associated with a safety case approach, whether or not a third party would be involved.

\187\ A-P-T Research, Inc. "A New Path to Launch Licenses," Doc. No. CDSP-FL004-18-00402 (October 16, 2018).

Additional Technical Justification and Rationale

Additional Technical Justification and Rationale

The sections below provide detailed discussions of flight safety analyses and software safety. Additionally, this section discusses the numerous conforming changes the FAA proposes to the existing regulations in order to implement the proposed regulations.

A. Flight Safety Analyses

A. Flight Safety Analyses

As discussed earlier, for purposes of this proposed rule, an FSA consists of a set of quantitative analyses used to determine flight commit criteria, flight abort rules, flight hazard areas, and other mitigation measures, and to verify compliance with the public safety criteria in proposed § 450.101. The FAA proposes 15 sections for flight safety analysis, as discussed below. 1. Scope and Applicability

Proposed § 450.113 establishes the portions of flight for which an operator would be required to perform and document an FSA, and would describe the analyses required for each type of operation. The portion of flight governed by the public safety criteria is central to the scope of the FSA.

The current scope of FSA regulations is laid out in Sec. § 417.201 and 417.107(b) for ELVs. Specifically, § 417.107(b)(1) currently requires that FSAs quantify the collective risks from lift-off through orbital insertion for orbital launches and from lift-off to final impact for suborbital launches. Unfortunately, § 417.107(b)(2) does not clearly specify the portion of flight for which an FSA must quantify the individual risks. In practice, the FAA has reconciled this vagueness by requiring the same scope for both collective and individual risks: From lift-off through orbital insertion for orbital launches and from lift-off to final impact for suborbital launches.

It is also unclear in current regulations what portions of flight the FSA needs to cover for RLVs. Section 431.35(b)(1) simply states that the collective public risk limit applies to each proposed reentry, but does not speak specifically to beginning and end of the period of flight that an FSA must analyze. Reentry means to return or attempt to return, purposefully, a reentry vehicle from earth orbit or from outer space to Earth.\188\ Reentry includes activities conducted in Earth orbit or outer space to determine reentry readiness and that are critical to ensuring public health and safety and the safety of property during reentry flight. The definition also includes activities conducted on the ground after vehicle landing on Earth to ensure the vehicle does not pose a threat to public health and safety or the safety of property. In practice, the FAA has required public risk assessments to begin at the final health check prior to initiation of de-orbit burn and ending when flight stops, such as splashdown for a capsule.

\188\ 14 CFR 401.5.

Further, for both ELVs and RLVs, the current regulations do not expressly address the potential public safety hazards caused by the disposal of a launch vehicle stage or component from orbit. That is, Sec. § 417.107(b) and 431.35(b)(1), in addressing the public risk criteria, do not specifically address the disposal of launch vehicle stages or components. As discussed earlier, such vehicle disposals have become more common in recent years, reflecting the elevated priority put on orbital debris mitigation. The FAA explained in the 2016 final rule \189\ that when the FAA requires that the quantitative risk analysis account for the planned impact of a first stage (or any stage) jettisoned prior to orbital insertion, it includes accounting for stage impacts regardless of whether the actual impact occurs before or after orbital insertion.

\189\ Changing the Collective Risk Limits for Launches and Reentries and Clarifying the Risk Limit Used to Establish Hazard Areas for Ships and Aircraft, Final Rule. 81 FR 47017 (July 20, 2016).

For reentry, proposed Sec. § 450.101(b) and 450.113(a)(4) would clarify and reduce the period FSAs must analyze when quantifying the public risks posed by reentry operations. The proposal would clarify that post-flight operations are not included in the safety analyses necessary to quantify the public risks posed by reentry operations. In § 401.5, the FAA proposes to include a definition for deorbit that clarifies that deorbit begins with the final command to commit the vehicle to a perigee below 70 nautical miles, approximately 130 km, and ends when all vehicle components come to rest on the Earth.

Proposed § 450.113 replaces § 417.201 to clarify the scope and applicability of FSAs. In proposed § 450.113(a)(1), an operator would be required to perform and document an FSA for orbital launch, from lift-off through orbital insertion,\190\ including any component or stage landings. In proposed § 450.113(a)(2), an operator would be required to perform and document an FSA for suborbital launch, from lift-off through final impact. In proposed § 450.113(a)(3), the FAA clarifies the scope of disposal FSA that would be necessary to demonstrate compliance with the disposal safety criteria in proposed § 450.101(d). Specifically, for disposal, an FSA would span from the beginning of the deorbit burn through final impact.

\190\ The FAA proposes orbital insertion to mean the point at which a vehicle achieves a minimum 70-nautical mile perigee based on a computation that accounts for drag. This adopts the definition of orbital insertion in RCC 321-17 Standard.

Proposed § 450.113(a)(4) would require an operator to perform and document an FSA for reentry, from the beginning of the deorbit burn through landing. The proposal is consistent with current practice, but would clarify that post-landing activities are not included in the FSA.

Proposed § 450.113(a)(5) would explicitly address hybrid vehicles, which include air-launch rockets released from carrier aircraft such as the Pegasus rocket carried by a modified L-1011 airliner. The proposal would clarify that FSAs generally apply to hybrid vehicles, for all phases of flight unless the Administrator determines otherwise based on demonstrated reliability. Thus, the proposal would enable an operator of a hybrid vehicle with a high level of demonstrated reliability for the entire flight or for a phase of flight, to be exempt from performing some FSAs without seeking a waiver for the flight or phase of flight. Demonstrated reliability refers to statistically valid probability of failure estimates based on the outcomes of all previous flights of the vehicle or stage. For example, if an applicant seeks to operate a hybrid vehicle that features an air-launch rocket released from a carrier aircraft with minimal modification from the original design certified as a commercial transport aircraft, the FAA would find certain FSAs not applicable if empirical data sufficiently showed that the demonstrated reliability and estimated public risks of the system are equivalent to general aviation aircraft during a given phase of flight. Specifically, the FAA foresees that such an applicant could be exempt from some of the normal flight trajectory analysis requirements during the captive carry phases of flight if the applicant could demonstrate compliance with the public safety criteria in proposed § 450.101 without the benefit of some of the normal flight trajectory analysis outputs.

Proposed § 450.113(b) would identify the specific FSA actions applicable to all launch and reentry vehicles (in paragraph (b)(1)), a launch or reentry vehicle that relies on an FSS to comply with proposed § 450.101 (in paragraph (b)(2)), and launch of an unguided suborbital launch vehicle (in paragraph (b)(3)). 2. Flight Safety Analysis Methods

Proposed § 450.115 (Flight Safety Analysis Methods) would set the methodology requirements for FSAs. This section would replace the prescriptive requirements currently in § 417.203 and appendices A, B, C and I to part 417. Currently, § 417.203(a) requires that FSAs meet the requirements for methods of analysis contained in appendices A (section A417) and B (section B417) to part 417 for a launch vehicle flown with an FSS, and appendices B and C (section C417) for an unguided suborbital launch vehicle that uses a wind-weighting safety system. Specifically, section A417 provides prescriptive requirements on the FSA methodologies and products for a launch vehicle flown with an FSS. Section B417 provides prescriptive requirements on the FSA for hazard area analyses for ship and aircraft protection. Section C417 provides prescriptive requirements on the FSA methodologies and products for a launch vehicle flown with a wind weighting safety system.

Section 417.203(b) specifically lists the broad categories of approved methods of analysis while § 417.203(c) addresses requirements for alternate analysis methods. Section 417.203(c) currently requires that an alternate FSA method be based on accurate data and scientific principles, and is statistically valid. In practice, the FAA has evaluated the validity of an applicant's proposed methods by comparing the results to valid benchmarks such as data from mishaps, test, or validated high-fidelity methods. Section 417.203(e) requires that a launch operator demonstrate to the FAA compliance with the requirements of part 417, subpart C. In its application, a launch operator must include the analysis products required by parts 415, subpart F, 417, subpart A, and appendices A, B, C, and I, depending on whether the launch vehicle uses an FSS or a wind-weighting safety system.

Pursuant to § 431.35(c), the FSA for an RLV is required to account for any reasonably foreseeable hazardous event and safety-critical system failures during launch flight or reentry that could result in a casualty to the public. However, part 431 does not include requirements for the methods used to provide an FSA, thus providing no standards for evaluating an FSA's validity or level of fidelity. The part 431 license applications approved by the FAA included FSA methodologies and products comparable to those in 417 license applications.

Proposed § 450.115(a) sets the scope for FSA methods. This section would not materially change the scope of the FSA methods under current parts 417 and 431, which account for the risk to the public from hazards associated with normal and malfunctioning vehicle flight in accordance to § 417.205(a). However, proposed § 450.115(a) would add language currently not expressly provided in § 417.205(a) that would require an operator's FSA method to account for all reasonably foreseeable events and failure of safety-critical systems. This language is consistent with the current requirement in § 431.35(c) to account for any reasonably foreseeable hazardous event, and safety-critical system failures during launch flight or reentry that could result in a casualty to the public.

Proposed § 450.115(b) would establish the level of fidelity for FSAs. Specifically, it would require a level of fidelity sufficient to demonstrate that any risk to the public would satisfy the public risk criteria of proposed § 450.101, including the use of mitigations, accounting for all known sources of uncertainty, using a means of compliance accepted by the Administrator. It would also require that the analysis identify the dominant source of each type of public risk with a criterion in proposed § 450.101(a) or (b) in terms of phase of flight, source of hazard (such as toxic exposure, inert, or explosive debris), and vehicle response mode. Thus, this proposed rule would provide performance targets instead of the current part 417 approach that mandates a single level of fidelity equivalent to methods that comply with the extensive requirements given in the appendices of part 417.

The requirements in proposed § 450.115(b) would account for all known sources of uncertainty and identify the dominant sources of risk. The proposal would be consistent with the best practices of other regulatory agencies that use quantitative risk analyses as part of a risk management approach to ensure public safety. The Nuclear Regulatory Commission (NRC), which has a long history of performance-based regulations with quantitative risk analyses to ensure public safety, has a long-standing policy to ensure that the quantitative techniques used for regulatory decision-making take into account the potential uncertainties that exist so that an estimate can be made on the confidence level to be ascribed to the quantitative results.\191\ The NRC has also found that, through use of quantitative techniques, important uncertainties have been, and continue to be, brought into better focus and may even be reduced as compared to those that would remain with sole reliance on deterministic decision-making. The NRC found that direct lack of severe accident experience makes it necessary that proper attention be given not only to the range of uncertainty surrounding probabilistic estimates, but also to the phenomenology that most influences the uncertainties. In other words, the NRC found the need to identify the dominant sources of public risks and their uncertainties when using quantitative risk analyses to ensure public safety.\192\

\191\ Nuclear Regulatory Commission, Nuclear Regulatory Safety Policy Goals. 51 FR 28044 (August 21, 1986).

\192\ The Department of the Interior (DOI), Bureau of Reclamation, uses risk criteria for achieving public protection in dam safety decision-making in a manner consistent with this proposed rule. Specifically, the DOI uses mean values calculated from Monte Carlo or similar analyses that include explicit treatment of input uncertainty.

The FAA would require that operators use a means of compliance accepted by the Administrator for FSA methods. The FAA plans to publish a draft version of that AC concurrently with this NPRM. An important aspect of that AC is the use of approaches generally consistent with the consensus U.S. Government standards on launch and reentry risk assessments (e.g., RCC 321). The RCC 321 Standard (paragraph 2.4) recognizes that there is significant uncertainty in the computed risks of rocket launches and notes that confidence bounds of 90 percent describing the uncertainty in the computed risk can span multiple orders of magnitude. Thus, the consensus U.S. Government standards on launch and reentry risk assessments contains a policy statement that uncertainty cannot be ignored. The RCC 321 Supplement further concurred with several statements originally made by the NRC, including the following three: (1) The use of mean estimates does not, however, resolve the need to quantify (to the extent reasonable) and understand those important uncertainties involved in risk predictions; (2) sensitivity studies should be performed to determine those uncertainties most important to the probabilistic estimates; and (3) the results of sensitivity studies should be displayed showing, for example, the range of variation together with the underlying science or engineering assumptions that dominate this variation. Even so, the RCC went on to conclude that a formal uncertainty analysis may not be necessary under conditions where the best mean estimate of the public risk is low relative to the collective risk criterion.

For this rulemaking, the FAA considered adopting an approach to the treatment of uncertainty following RCC 321 Standard and Supplement. The FAA requests comment on whether this treatment of uncertainty is reasonable. Specifically, the FAA solicits input on the process whereby the uncertainty does not have to be considered if the computed risk is less than one-third of the primary aggregated collective risk criterion.\193\ Current Air Force practice is to include implementation of measures to improve risk analyses to reduce the level of uncertainty when the predicted risks exceed 3 x 10-5 EC. Examples of that could include refined input data or a higher-fidelity method for the risk computations.

\193\ The choice of one-third was consistent with the recommendation in AFSPCMAN 91-710 Vol.1, 1 July 2004. Attachment 5 states that if risk to all individuals from a single hazard exceeds an EC of 30 x 10-6, a range user may have to take additional measures to protect personnel and resources. Examples include to fix, correct, or improve existing non-compliances, improve risk analyses to reduce the level of uncertainty, require a day-of-launch risk analysis, or establish disaster aversion criteria.

Similarly, if the estimated risk level exceeds 3 x 10-5 EC, the RCC 321 Standard states that the range should compute the uncertainty to ensure that a launch is not allowed that would violate the criterion based on best estimates that account for uncertainty. There are published examples of uncertainty analyses for launch risks that explicitly account for uncertainties associated with the input data (e.g., the probability of failure associated with a given break-up state vector), and biases and uncertainties in key sub-models (e.g., the sub-model used to compute the PC given an impact with a given piece of debris on a specific structure type). However, the end effect of the RCC 321 Standard approach to uncertainty treatment is that a range or range user could continue operating under current practice, using their current tools without formal uncertainty quantification for missions with a collective risk no greater than 3 x 10-5 EC. Under the RCC approach, only missions that pose collective risks above 3 x 10-5 EC based on point estimates would be required to perform formal uncertainty quantification. The FAA requests comment on whether the current approaches to uncertainty treatment employed by the RCC or the Air Force are viable in the FAA's regulatory framework. The FAA further requests comments on any currently available approaches to address uncertainties in public risk assessments, including the approach identified in the draft means of compliance on uncertainty and level of fidelity in FSA methods.

Proposed § 450.115(b) would require that an operator account for all known sources of uncertainty in various FSAs. The FAA intends to ensure that FSA methods account for known sources of aleatory (random) uncertainties that are the result of inherently random processes. An example of aleatory uncertainty is the influence of prevailing weather conditions on the results of collective and individual risk analyses for launch or reentry. The true EC is often highly influenced by the prevailing weather conditions during the proposed operation. The uncertainty in the true EC due to weather conditions is substantial for a typical baseline risk analysis that accounts for the foreseeable weather conditions in a given month based upon historical data and assumes that an operation is equally feasible under any of those likely weather conditions given all the safety and mission assurance constraints.

For example, most vehicles would not attempt to fly through certain wind conditions due to the potential for the vehicle to break up or veer off-course, leading to a violation of safety or mission assurance constraints. The uncertainty in the true EC for a day-of-launch risk analysis is much smaller, but the uncertainty in any forecast or measured weather input data will still produce some uncertainty in the EC due to measurement errors and variability in the weather measurements and forecasts. There are several other potentially important sources of aleatory uncertainty in an EC analysis, and there are various valid approaches to account for these aleatory uncertainties. This proposed rule would require that aleatory uncertainties are accounted for, including known sources of randomness in critical input data.

These would include normal and malfunction trajectories, weather conditions, population and sheltering characteristics (e.g., between day and night), velocities induced during break-up, aerodynamic properties of the vehicle and debris, any yield from an explosive impact, and the amount of debris that burns up due to aero-thermal heating during re-entry.

Proposed § 450.115(c) would establish application requirements for methods of analysis. Specifically, the proposed rule would require that an applicant submit a description of the FSA methodology for each launch or reentry approved by the FAA, including identification of the scientific principles and statistical methods used, and all assumptions and their justifications. However, if the FAA determines that the range's FSA methods meets FAA safety requirements, then the operator would not be required to provide the FAA with a description of the FSA methodology. Also, an applicant would be required to include the rationale for the level of fidelity, the evidence for validation and verification required by proposed § 450.101(g), the extent that the benchmark conditions are comparable to the foreseeable conditions of the intended operations, and the extent the analyses accounted for risk mitigations. The FAA intends for assumptions to be justified using logic, historical flight experience data, relevant test data, and the results from physics-based simulations. 3. Trajectory Analysis for Normal Flight

The FAA proposes a single regulation governing an FSA for normal trajectories, applicable to all launch and reentry vehicles, in proposed § 450.117 (Trajectory Analysis for Normal Flight). The provision would distinguish between variability in the intended trajectory and uncertainties due to random sources of dispersion such as winds and vehicle performance. It would also clarify application requirements.

All the FSAs depend on some form of analysis of the trajectory under normal conditions, otherwise known as a normal trajectory. That is, one must first understand a vehicle's trajectory when it performs as intended and under normal conditions before one can determine the effects of malfunctions along its flight path.

Current regulations for normal trajectory analyses are found in Sec. § 417.207 and 431.35(d) and appendix A to part 417. Section 417.207 sets the current trajectory analysis requirements for ELVs. Section 417.207(a)(1) requires an analysis that establishes the limits of a launch vehicle's normal flight, as defined by the normal trajectory and potential three-sigma trajectory dispersions about the normal trajectory for any time after lift-off. Although this requirement is generally clear, the uncertainties the analysis must consider could be clearer. For example, the current requirement does not distinguish between inherently random uncertainties that could cause the actual trajectory to differ from the nominal trajectory, and variability in the known conditions immediately prior to the initiation of the operation (e.g., weather conditions at the time of the launch or the time into a launch window that the launch occurs for a rendezvous mission).

In terms of current RLV regulations in part 431, they describe flight trajectory analyses requirements in a single paragraph in § 431.35(d)(8). Specifically, the FAA requires that applicants provide flight trajectory analyses covering launch or ascent of the vehicle through orbital insertion and reentry or descent of the vehicle through landing, including its three-sigma dispersion. This regulation is silent as to the specific uncertainties for which the analysis must account. In practice, part 431 license applicants have provided normal trajectory data consistent with the part 417 regulations.

Proposed § 450.117 would retain the substantive normal trajectory analysis requirements currently in § 417.207 and the definitions of key terms such as "normal flight" and "normal trajectory." Proposed § 450.117(a)(1) would require a trajectory analysis that establishes the limits of a vehicles normal flight. The proposal would retain the requirement in § 417.207(a)(1) to establish a nominal trajectory where the vehicle performs as designed without any deviation due to winds, propulsion performance, or mass properties but would add clarity about the sources of uncertainty that a trajectory analysis must account for by distinguishing between variability and random uncertainty.

Specifically, the proposal would expressly require a trajectory analysis to establish two separate sets of trajectories to characterize distinct sources of uncertainty, including variability and random uncertainty. One set of normal trajectories in § 450.117(a)(1)(ii) would characterize the uncertainty during normal flight due to random deviations from ideal conditions, such as wind conditions, vehicle mass, and performance characteristics. Another set of normal trajectories in § 450.117(a)(1)(i) would characterize how the intended trajectory could vary due to conditions known prior to initiation of flight. An example of variability is how the intended trajectory would change due to different times for lift-off within a launch window that lasts several minutes for a mission with an orbital rendezvous as the primary objective. Another example of variability is how the intended trajectory would change due to wind conditions. In such cases, the nominal trajectory represents the most likely lift-off time. An FSA must distinguish between variability and random uncertainty in the normal trajectory in order to demonstrate that the criteria in proposed § 450.101 would be satisfied at any time the operator intends to initiate launch or re-entry flight.

Section 450.117(a)(2) would require a fuel exhaustion trajectory that produces instantaneous impact points with the greatest range for any given time after liftoff for any stage that has the potential to impact the Earth and does not burn to propellant depletion before a programmed thrust termination. This is the same as current § 417.207(a)(2). The FAA is unaware of any challenges with the current regulation regarding a fuel exhaustion trajectory.

For vehicles with an FSS, proposed § 450.117(a)(3) would establish a new requirement for trajectory data or parameters that describe the limits of a useful mission. The FAA proposes in § 401.5 to define the "limits of a useful mission" as the trajectory data or other parameters that describes the limits of a mission that can attain the primary objective, including but not limited to flight azimuth limits. Thus, the proposal would require an operator to establish the limits of a useful mission based on the values of trajectory parameters necessary to attain the primary mission objective, including flight azimuth limits. Note that the azimuth limit data is currently required by the Air Force in Air Force Space Command Manual (AFSPCMAN) 91-710 Vol. 2. The limits of a useful mission are essential input data for the flight safety limits analysis, and for an evaluation of whether a vehicle should be allowed to pass through a gate, as discussed later in this preamble.

Proposed § 450.117(b) would require a final trajectory analysis to use a six-degree of freedom trajectory model, and proposed § 450.117(c) would require a trajectory analysis to account for all wind effects, including profiles of winds that are no less severe than the worst wind conditions under which flight might be attempted, and for uncertainty in the wind conditions. These are similar to § 417.207(b) and (c), respectively.

Proposed § 450.117(d) would provide application requirements for trajectory analyses that address the proposed methodology, input data, and output data. In paragraph (d)(1), an applicant would be required to describe the methodology used to characterize normal flight and the limits of a useful mission, including the scientific principles and statistical methods used, all assumptions and their justifications, the rationale for the level of fidelity of the methods, and the evidence for validation and verification that would be required by proposed § 450.101(g). In paragraph (d)(2), the FAA proposes to require that the applicant describe the input data used in normal trajectory analyses and provides a list of the minimum input data an applicant must describe. In paragraph (d)(3), the FAA proposes to require that an applicant describe a representative normal trajectory analysis outputs (e.g., position, velocity, and vacuum instantaneous impact point) for each second of flight for (1) the nominal trajectory, (2) a fuel exhaustion trajectory under otherwise nominal conditions, (3) a set of trajectories that characterize variability in the intended trajectory based on conditions known prior to initiation of flight, (4) a set of trajectories that characterize how the actual trajectory could differ from the intended trajectory due to random uncertainties, and (5) a set of trajectories that characterize the limits of a useful mission as described in proposed § 450.117(a). The proposed application requirements provide regulatory clarity regarding the normal trajectory characterization necessary to ensure compliance with proposed § 450.101.

Note that in this proposed section, and other proposed flight safety analysis application requirements, the FAA requires representative data. This allows the FAA to evaluate an applicant's methodologies. Representative data should be the best, meaning the most realistic, data available given the intended flight parameters.

The applicant would also be required to submit additional products that allow the FAA to conduct an independent analysis, if requested by the Administrator. This same application requirement would also be in proposed Sec. § 450.119 through 450.141. At times, the FAA conducts independent flight safety analyses which usually require additional information than is normally required of an applicant. Instead of attempting to list out what is needed for every independent analysis, which is usually case-specific, the FAA proposes to simply state that more information may be necessary. The FAA's conduct of an independent analysis is usually reserved for new vehicle concepts, new analysis methods, or proposals that involve unique public safety issues. 4. Trajectory Analysis for Malfunction Flight

Proposed § 450.119 (Trajectory Analysis for Malfunction Flight) would consolidate trajectory analysis requirements for all launch and reentry vehicles. In consolidating, the FAA would also update its requirements to reflect advancements in trajectory analysis capabilities and clarify application requirements. A malfunction trajectory analysis is necessary to determine how far a vehicle can deviate from its normal flight path in case of a malfunction. This analysis helps determine impact points in case of a malfunction and is therefore a vital input for the analyses needed to demonstrate compliance with risk criteria. The FAA's current regulations covering trajectory analyses in case of malfunction are in § 417.209 (Malfunction turn analysis), appendix A to part 417, and § 431.35(d)(8).

Current § 417.209 sets forth the trajectory analysis requirements in case of a malfunction applicable to ELVs. Section 417.209(a)(1) requires a trajectory analysis to establish the launch vehicle's turning capability in the event of a malfunction during flight using a set of turn curves. Appendix A to part 417 (section A417.9) also provides more detailed and prescriptive requirements for analyzing "turn curves." Turn curve data offered a reasonable way to simulate failures that produce trajectory departures, particularly in response to thrust offsets when computational limitations made it impractical to perform six degrees of freedom (6-DOF) simulations of malfunction trajectories.

In the past, turn curves produced a reasonable way to model the classic cornus spiral behavior associated with a constant thrust offset or nozzle burn-through. Thus, § 417.209(b) requires a set of turn curves to establish the launch vehicle velocity vector turn angle from the nominal launch vehicle velocity vector, and to establish the vehicle velocity turn magnitude from the nominal velocity magnitude. There are two fundamental types of malfunction turn curves: (1) One that shows how the magnitude velocity changes during the turn; and (2) the other for the direction of the velocity. Given advancements in computational capabilities, the use of turn curves as mandated by the current regulations constitutes an outdated and unnecessarily simplified analysis technique. For instance, through current computational capabilities, particularly the prevalence of 6-DOF trajectory models, it is generally more efficient and more accurate for an applicant to provide sets of Monte Carlo trajectories that characterize a given type of malfunction, even for the thrust vector offsets and nozzle burn-through, than to provide turn curve data.

The current RLV regulations in part 431 do not explicitly address malfunction trajectory analyses. Section 431.35(d)(8) describes flight trajectory analysis requirements in a single paragraph. It requires that applicants provide flight trajectory analyses covering launch or ascent of the vehicle through orbital insertion and reentry or descent of the vehicle through landing, including its three-sigma dispersion. In practice, part 431 license applicants have provided malfunction trajectory analyses consistent with the part 417 regulations. However, the lack of clarity regarding the malfunction trajectory analysis requirements and ensuing discussions between the FAA and operators has resulted in inefficiencies and delays in the licensing process.

Proposed § 450.119 would consolidate all trajectory analysis requirements for a malfunctioning flight which would be applicable to any launch or reentry vehicle. Based on the noted advancements in computational capabilities that have rendered the current use of turn curves outdated and over simplistic, the FAA proposes to remove the § 417.209(b) requirements related to turn curves in favor of more modern Monte Carlo methods. Proposed § 450.119(b) would provide performance-based requirements regarding what a malfunction trajectory analysis must account for, including applicable times in flight and valid trajectory time intervals.

Specifically, the proposal would require the analysis to account for (1) all trajectory times during the thrusting phases or when the lift vector is controlled during flight, (2) the duration starting when a malfunction begins to cause each flight deviation throughout the thrusting phases of flight, and (3) trajectory time intervals between malfunction turn start times that are sufficient to establish flight safety limits, if any, and individual risk contours that are smooth and continuous. The proposal would retain in § 450.119(b)(4) the performance-based requirement currently in § 417.209(a)(3) to establish the relative probability of occurrence of each malfunction turn of which the vehicle is capable.

In proposed § 450.119(b)(5), the analysis would also have to account for the probability distribution of position and velocity of the vehicle when each malfunction will terminate due to vehicle breakup, along with the cause of termination and the state of the vehicle.\194\ Finally, in proposed § 450.119(b)(6), the analysis would establish the vehicle's flight behavior from the time when a malfunction begins to cause a flight deviation until ground impact or predicted structural failure, with trajectory time intervals that are sufficient to establish individual risk contours that are smooth and continuous.

\194\ The proposed § 450.119(b)(5) requirement would be equivalent to the § 417.209(a)(4) through (9) requirements. Under § 417.209, the FAA prescribed the use of "turn curves" that were a particular way to compute the position and velocity at the end of a malfunction trajectory.

Finally, proposed § 450.119(c) would provide application requirements for malfunction trajectory analyses that address the proposed methodology, input data, and output data. An applicant would be required to describe the methodology used to characterize malfunction flight including the same elements required for the normal trajectory analyses. The FAA proposes to require that an applicant describe the input data used in malfunction trajectory analyses and provides a list of the minimum data an applicant must describe. The FAA also proposes to require that an applicant describe representative malfunction trajectory analysis outputs (e.g., position, velocity, and vacuum instantaneous impact point) for each second of flight and for the probability of each trajectory that characterizes a type of malfunction flight. Finally, the FAA may also request additional products to conduct an independent analysis. These proposed application requirements are consistent or less burdensome than current requirements. 5. Debris Analysis

Proposed § 450.121 (Debris Analysis) would set the requirements for debris analysis by revising current requirements in § 417.211 (Debris analysis), accounting for part 431 practices not fully expressed in the regulatory language, consolidating requirements from § 417.107 (Flight Safety), and removing overly prescriptive and burdensome requirements from Appendix A to part 417.

Under § 417.211(a), a debris analysis must identify the inert, explosive, and other hazardous vehicle debris that results from normal and malfunctioning flight. Section 417.211(b) specifies that a debris analysis must account for various causes of a launch vehicle breakup. This analysis includes debris from any flight termination system activation, launch vehicle explosion, aerodynamic loads, inertial loads, atmospheric reentry heating, and impact of an intact vehicle. Section 417.211(c) asks for a list of debris fragments for each cause of breakup and any planned jettison of debris, launch vehicle components, or payload. Also, § 417.107(c) contains debris threshold requirements for debris analysis and appendix A to part 417 (section A417.11) provides detailed direction on the debris analysis constraints, debris models, and other debris analysis products.

Although part 431 does not expressly ask for a debris analysis, the FAA has deemed § 431.35(b) to require one, applying the same standards as those in part 417. However, this lack of regulatory specificity in part 431 has led to longer pre-application consultation periods as the FAA and operators worked to ascertain the applicable requirements.

Proposed § 450.121 would provide performance-based regulations regarding the level of fidelity required for key elements of a valid debris analysis. Proposed § 450.121(a) would include a debris analysis that characterizes the debris generated for each foreseeable vehicle response mode as a function of vehicle flight time, accounting for the effects of fuel burn and any configuration changes.

The FAA proposes to add the references to fuel burn and configuration changes that are absent from current part 417 because an operator's debris list will change over time with variations to the amount of available propellant and with the jettisoning of hardware.

Proposed § 450.121(b) would require that the debris analysis account for each foreseeable cause of vehicle breakup, including any breakup caused by an FSS activation or by impact of an intact vehicle. This proposal would include debris from a vehicle's jettisoned components and payloads because such debris could cause a casualty due to impact with an aircraft or waterborne vessel or could pose a toxic or fire hazard. This proposal is consistent with the ARC recommendation to develop a process for a debris catalogue. Foreseeable causes of vehicle breakup would include engine or motor explosion, or exceeding structural limits due to aerodynamic loads, inertial loads, or aerothermal heating.

Proposed § 450.121(c) is substantively the same as § 417.107(c). The section contains the debris thresholds requirements. It would adopt the references to inert, explosive, and other hazardous vehicle debris currently in § 417.211(a). The inert debris requirement would include all debris that could impact a human being with a mean expected kinetic energy at impact greater than or equal to 11 ft-lbs, or mean impact kinetic energy per unit area of 34 ft-lb/ in\2\. The required thresholds are well-established standards used by Federal launch ranges. In general, the 11 ft-lb requirement is the primary threshold for debris, whereas the 34 ft-lb/in\2\ is for penetrating injuries. This paragraph also would clarify the need to consider the effects of all inert debris on aircraft or waterborne vessels, or those that pose a toxic or fire hazard. The debris analysis would also be required to identify any explosive debris.

Proposed § 450.121(d) would provide the debris analysis application requirements. This paragraph would inherit, in a less detailed and prescriptive manner, the requirements in appendix A to part 417, section A417.11. It would expressly identify the information and data needed by the FAA to evaluate compliance with the regulatory requirements. Proposed § 450.121(d) would describe the level of fidelity required for the products of a debris analysis including (1) a description of the debris analysis methodology, including input data, assumptions, and justifications for the assumptions; (2) a description of all vehicle breakup modes and the development of debris lists; and (3) all debris fragment lists necessary to quantitatively describe the physical, aerodynamic, and harmful characteristics of each debris fragment or fragment class. Finally, as discussed earlier, the applicant would be required to provide additional products as requested by the FAA to conduct an independent analysis to ensure that public safety criteria are satisfied. 6. Flight Safety Limits Analysis

Proposed § 450.123 would set the requirements to identify uncontrolled areas and establish flight safety limits that define when an operator must initiate flight abort to (1) ensure compliance with the public safety criteria of proposed § 450.101 and (2) prevent debris capable of causing a casualty from impacting in uncontrolled areas if the vehicle is outside the limits of a useful mission.

Current § 417.213(a) requires that a flight safety limits analysis identify the location of populated or other protected areas and establish flight safety limits to define when an FSS must terminate a launch vehicle's flight to prevent hazardous impacts from reaching any protected area and ensure that the public risk criteria of § 417.107(b) are satisfied. Section 417.3 currently defines a flight safety limit as criteria to ensure a set of impact limit lines established for the flight of a launch vehicle flown with an FSS bound the area where debris with a ballistic coefficient of 3 psf or more is allowed to impact when an FSS functions. Thus, § 417.213(a) and the definition of flight safety limit require that any populated area be protected by flight safety limits from where the FSS must be activated. This requirement is not consistent with operations on Federal launch ranges that allow potential debris impact in populated areas inside the impact limit lines, as long as the individual and collective public risks remain within acceptable limits.

The requirements in § 417.213(b) are specific about potential contributors to the vehicle and debris dispersions for which the flight safety limits analysis must account including time delays, all wind effects, velocity imparted to vehicle fragments by breakup, all lift and drag forces on the malfunctioning vehicle and falling debris, all launch vehicle guidance and performance errors, all launch vehicle malfunction turn capabilities, and any uncertainty due to map errors and launch vehicle tracking errors.

Section 417.213(d) requires that the analysis establish designated impact limit lines to bound the area where debris with a ballistic coefficient of 3 psf is allowed to impact, assuming the FSS functions properly. In contrast, part 431 does not contain any express requirements for a flight safety limits analysis to set flight safety limits. That being said, part 431 license applicants have performed a flight safety limits analysis mirroring part 417 requirements in cases where an FSS was employed to satisfy the public risk criteria in § 431.35(b).

The FAA proposes to move the definition of "flight safety limit" from current § 417.3 to § 401.5 and update the definition to mean criteria to ensure that public safety is protected from the flight of a vehicle when an FSS functions properly. Thus, the proposal would remove any ballistic coefficient threshold from the definition of a flight safety limit. As previously discussed, the Air Force has permanently waived its previous requirement that embedded a specific ballistic coefficient threshold into the flight safety limits, and the FAA has also waived the corresponding requirement in § 417.213(d).\195\ When the FAA adopted the 3 psf ballistics coefficient standard (in 2006), the FAA recognized that ballistic coefficient is not well correlated with the probability of a casualty producing impact.\196\ Simply put, ballistic coefficient is an imperfect surrogate that was adopted based on past practice when computers were less capable than today.

\195\ 81 FR 1470 (January 12, 2016).

\196\ Licensing and Safety Requirements for Launch, NPRM. 67 FR 49464 (October 28, 2002).

In § 401.5, the proposal would also replace the term "protected area" with "uncontrolled area," defined as an area of land not controlled by a launch or reentry operator, a launch or reentry site operator, an adjacent site operator, or other entity by agreement. This change reflects the fact that all members of the public, even those in areas of land controlled by a launch operator, are protected to the extent that collective and individual public risk limits apply everywhere. Specifically, proposed § 450.123(a) would require protection of uncontrolled areas by flight safety limits and ensure compliance with the public safety criteria of proposed § 450.101, while controlled areas would be required to meet only the collective and individual risk requirements (also in accordance with proposed § 450.101).

The FAA intends to assess the need for flight safety limits to protect environmentally-sensitive areas in the environmental review process of proposed § 450.47. The FAA anticipates that not all environmentally-sensitive areas will need this protection. For example, current practice for launches from the Western Range protects a National Marine Sanctuary in the Pacific Ocean against planned impacts of jettisoned items, but not against debris from a flight abort.

Proposed § 450.123(a) would require an FSA to identify the location of uncontrolled areas and establish flight safety limits that would define when an operator must initiate flight abort to prevent debris capable of causing a casualty from impacting in uncontrolled areas if the vehicle is outside the limits of a useful mission, and to ensure compliance with the public safety criteria of proposed § 450.101. Given flight safety limits are only required to protect people in uncontrolled areas and not people in controlled areas, the proposal would reconcile the current inconsistency between the part 417 requirements versus the current practice at some Federal launch ranges that allows the public's exposure to debris hazards as long as the collective and individual risk criteria are met.

Proposed § 450.123(b) would require a flight safety limits analysis to identify flight safety limits for use in establishing flight abort rules. The flight safety limits would be required to account for temporal and geometric extents on the Earth's surface of any vehicle hazards resulting from any planned or unplanned event for all times during flight, and account for potential contributions to the debris impact dispersions. This is the same as § 417.213(b). Proposed § 450.123(b)(3) would add a requirement to design flight safety limits to avoid flight abort under conditions that result in increased collective risk to people in uncontrolled areas, compared to continued flight.

The proposed requirement is equivalent to the U.S. Government consensus standard that a conditional risk management process should be implemented to ensure that mission rules do not induce unacceptable consequences when they are implemented.\197\ In the flight safety context, a flight abort is a good example of a safety intervention intended to mitigate public risks, but that typically induces a conditional risk (e.g., a consequence associated with the debris event triggered by the flight abort). A flight safety limits analysis would ideally minimize all foreseeable consequences, not just those to people on the ground or to the extent necessary to meet the public safety criteria. For example, placing flight safety limits in areas where flight abort might place debris on a busy shipping lane or air corridor is not an ideal solution when other locations for the limits could meet the public safety criteria and consequence criteria, and still provide space for the vehicle to fly a useful mission.

Also, as a malfunctioning vehicle's debris footprint migrates towards a populated area, the consequence to people on the ground from a flight abort will increase from a low number and possibly reach the proposed consequence limit. The ideal location for a flight safety limit on such trajectory is not at the last location where an abort would still result in meeting the consequence criteria, which would presumably result in a consequence close to the limit, but at a location that minimizes the consequence. This proposed approach could result in flight safety limits that provide debris containment, or nearly so, while also allowing normal flight and flight within the limits of a useful mission without triggering an abort. In summary, the design of the flight safety limits and the associated flight safety rules would be required to avoid an increase in risk induced by a flight abort, compared to inaction or action at a different time. This is relevant to areas where debris containment is not possible, as discussed in greater length in the next section on proposed § 450.125.

\197\ RCC 321-10 at p. 2-7.

Proposed § 450.123(c) would require the flight safety analysis to include a gate analysis for an orbital launch, or any launch or reentry where one or more trajectories that represents a useful mission intersects a flight safety limit that provides containment of debris capable of causing a casualty. This is also discussed in more detail in the next section on gate analysis.

Proposed § 450.123(d) would provide flexibility to allow the computation of flight safety limits in real-time in lieu of computing flight safety limits preflight. This alternative would reduce the number of assumptions used in the flight safety limits analysis and allow for a computation that uses the best available data on the vehicle state. The proposal would allow the computation of flight safety limits in real-time to be performed on the ground or onboard the vehicle.

The FAA proposes to remove the requirement for a straight-up time analysis currently in § 417.215. A straight-up time analysis establishes when to terminate the flight of a vehicle that fails to pitch over, and thus flies straight up, to achieve debris containment. The straight-up time is not the only method of limiting the risks and consequences to the launch area in the case of a vehicle that flies a straight-up trajectory. Although the express provision is being removed in the proposed rule, the new performance-based analysis permitted under § 450.213 would allow the straight-up time approach to control the hazards from a straight-up flight, but its use would not be required.

Proposed § 450.123(e) lays out the application requirements for flight safety limits analyses. The FAA would require an applicant to submit: (1) A description of how each flight safety limit will be computed; (2) representative flight safety limits and associated parameters; (3) an indication of which flight abort rule from proposed § 450.165(c) is used in conjunction with each example flight safety limit; (4) a graphic depiction or series of depictions of representative flight safety limits, the launch or landing point, all uncontrolled area boundaries, and vacuum instantaneous impact point traces for the nominal trajectory, extents of normal flight, and limits of a useful mission trajectories; (5) if the requirement for flight abort is computed in real-time in lieu of precomputing flight safety limits, a description of how the real-time flight abort requirement is computed including references to public safety criteria of § 450.101; and (6) additional products requested by the FAA for an independent analysis when necessary to demonstrate compliance with risk criteria. The proposed application requirements are consistent with current practice under parts 417 and 431. 7. Gate Analysis

The FAA proposes § 450.125 to make regulations governing gate analyses more performance-based, flexible, and clear. This change would include revising the definition of "gate" and, as discussed earlier, adding a definition of the "limits of a useful mission." The proposal would also add an option to relax flight safety criteria without using a gate.

Current § 417.3 defines a "gate" as the portion of a flight safety limit boundary through which the tracking icon of a launch vehicle flown with an FSS may pass without flight termination. As discussed earlier, a gate is an opening in a flight safety limit through which a vehicle may fly, provided the vehicle meets certain pre-defined conditions such that the vehicle performance indicates an ability to continue safe flight. If the vehicle fails to meet the required conditions to pass a gate, then flight abort would occur at the flight safety limit. In other words, the gate would be closed.

The FAA has requirements for an overflight gate analysis in § 417.217 and appendix A, section A417.17, and for a hold-and-resume gate analysis in § 417.218. An overflight gate analysis determines whether a vehicle can overfly populated areas. This analysis requires a launch operator determine why it is safe to allow flight through a flight safety limit--the limit that protects populated or protected areas--without terminating a flight. This analysis accounts for the fact that it is potentially more dangerous to populated or protected areas to destroy a malfunctioning vehicle during certain portions of a launch than not to destroy it. In some circumstances, a destroyed vehicle may disperse debris over a wider area affecting more people than if the vehicle were to impact intact.

The primary purpose of flight safety limits and gates is to establish safe locations and conditions to abort the flight prior to the vehicle entering a region or condition where it may endanger populated or other protected areas if flight were to continue. From an operator's perspective, a gate should allow the vehicle to fly through a flight safety limit when the trajectory corresponds to a useful mission.\198\ Otherwise, a flight abort would be required for every flight that intersects with a flight safety limit even if the mission can still have a successful outcome. The optimal use of flight safety limits and gates would be to prevent vehicles that cannot achieve a useful mission from continuing flight, even when the flight is along a trajectory that crosses a gate.

\198\ As discussed earlier in this preamble, the FAA proposes in § 401.5 to define the "limits of a useful mission" as the trajectory data or other parameters that describes the limits of a mission that can attain the primary objective, including but not limited to flight azimuth limits.

The current gate regulations imply that gates are the only option when debris containment is not possible along a trajectory that represents a useful mission, whether it is normal or outside of the normal trajectory envelope. This requirement does not reflect current practice at the Federal launch ranges. Federal launch ranges sometimes relax flight safety limits to allow continued flight for these trajectories without the use of a gate, as long as the operations satisfies the collective risk criterion. Also, some Federal launch ranges do not currently require explicit identification of the conditional risk posed by a vehicle that flies on a trajectory within the normal trajectory envelope or the limits of a useful mission. The preflight risk due to such a trajectory is often small because the vehicle is not likely to deviate far from nominal. However, a gate or relaxed flight safety limit to allow flight on such a trajectory implies that the risk must be acceptable given that the vehicle does fly on such a trajectory. Such a failure to identify the conditional risk associated with such a trajectory as part of the gate analysis is inconsistent with the U.S. Government consensus standard (RCC 321-17 paragraph 2.3.6) that a conditional risk management process should be implemented to ensure that mission rules do not induce unacceptable levels of risk when they are implemented.

Although part 431 has no requirements related to gate analysis, the one orbital RLV operation licensed to date employed an FSS and performed a gate analysis.

The FAA's proposed § 450.125 would establish a single set of performance-based gate analysis requirements applicable to all launch and reentry vehicles. The gate analysis requirements in Sec. § 417.217 and 417.218 would be combined. Proposed § 450.125 would remove prescriptive requirements on the types of gates, standardize the requirements for establishing a gate, and open the possibility of relaxing flight safety limits. The FAA believes an operator should have the freedom to select risk mitigation methods that will present the best safety posture rather than prescribing certain strategies that may not be the best for all scenarios and vehicles. The FAA also proposes to revise the existing definition of "gate" in § 401.5 to replace the term "flight termination" with "flight abort" and to add language to reflect that the flight must remain within specified parameters to avoid flight abort.

Proposed § 450.125(a) would require a gate analysis for an orbital launch, or any launch or reentry where one or more trajectories that represents a useful mission intersects a flight safety limit that provides containment of debris capable of causing a casualty.

Proposed § 450.125(b) would set the gate analysis requirements. The FAA would require an analysis to establish a relaxation of flight safety limits to allow continued flight or a gate where a decision will be made to abort the launch or reentry, or allow continued flight. If a gate is established, the analysis should establish a measure of performance at the gate that would enable the flight abort crew or autonomous FSS to determine whether the vehicle is able to complete a useful mission, and abort the flight if it is not. Further, the analysis should establish accompanying flight abort rules. Finally, for an orbital launch, the analysis should establish a gate at the last opportunity to determine whether the vehicle's flight is in compliance with the flight abort rules and can make a useful mission, and abort the flight if not. This last requirement would achieve the goal of assuring that only missions that can be useful are allowed to proceed to orbit, thereby limiting the potential for space debris. In addition, when the vehicle performance does not demonstrate an ability to reach a minimum safe orbit (without an imminent random reentry), meaning it cannot pass the useful mission requirement, the regulation would require that flight abort occur.

In proposed § 450.125(c), the FAA would require the extents of any gate or relaxation of the flight safety limits to be based on normal trajectories, trajectories that may achieve a useful mission, collective risk, and consequence criteria. In proposed § 450.125(c)(1), the FAA proposes to require a gate or relaxation of flight safety limits anywhere a flight safety limit intersects with a normal trajectory if that trajectory would meet the individual and collective risk criteria of proposed § 450.101(a)(1) and (2) or (b)(1) and (2) when treated like a nominal trajectory with normal trajectory dispersions.\199\ Requiring all normal trajectories to be treated like a nominal trajectory with dispersions as input to a conditional risk analysis (given a sample normal trajectory) for the gate analysis would resolve the issue of an incomplete characterization of the conditional risk of a vehicle that flies through what was a flight safety limit while within the normal trajectory envelope.

\199\ The FAA would retain the definitions of "normal flight" and "normal trajectory" currently found in § 417.3.

Another requirement of the proposed gate analysis would be that the predicted average consequence from flight abort resulting from any reasonably foreseeable vehicle response mode, in any one-second period of flight, using any modified flight safety limits must not exceed 1 x 10-2 CEC. The goal of this requirement is to ensure that flight safety limits do not create an unacceptable consequence when used, since debris containment is no longer provided. A gate that does not have flight safety limits after the gate would not need to meet this consequence criterion since it would be placed at the same location as flight safety limits that do provide debris containment. Under the proposal, any intersections of flight safety limits with normal trajectories would result in flight safety limits that are relaxed enough to allow passage, or an open gate in the flight safety limit as long as there is enough data available to confirm that the vehicle is healthy (i.e., appears capable of reaching a minimum safe perigee).

Flight on normal trajectories must still meet the public safety criteria in proposed § 450.101, so this practice would ensure acceptable risks and use the best available data to confirm that a vehicle is unlikely to fail before being allowed to fly through a gate, if one is present. Whether flight safety limits would be relaxed enough to let a vehicle fly through that area, or be gated, is optional. A gate is preferred if it would reduce risk, given that there is sufficient information available to make a decision on whether the vehicle is sufficiently healthy to pass. This practice would align with the Federal launch range's current practice and meet the intent of the current requirement in § 417.107(a)(2).

In proposed § 450.125(c)(2), trajectories that are outside of normal flight but within the limits of a useful mission would be evaluated as potential normal trajectories. Proposed § 450.125(c)(2) would allow flight safety limits to be gated or relaxed where they intersect with any trajectory within the limits of a useful mission, if the trajectory would meet the individual and collective risk criteria of proposed § 450.101(a)(1) and (2) or (b)(1) and (2), assuming that the trajectory flown would be treated like a nominal trajectory with normal trajectory dispersions. The predicted average consequence from flight abort resulting from a failure in any one-second period of flight, using any modified flight safety limits, would be required to not exceed 1 x 10-2 CEC. The philosophy behind proposed § 450.125(c)(2) is to allow a non-normal flight to continue as long as the mission does not pose an unacceptable conditional risk given the present trajectory.

A good example of missions that fall into this category are missions that lift-off on an incorrect flight azimuth, usually due to a software input error, such as the Ariane 5 failure on January 25, 2018, during its 97th mission (VA241). Apart from the programming error, these vehicles may be healthy and are not expected to fail more frequently than a flight without the programming error, so these flights should be allowed to continue if they meet the individual and collective risk criteria on the present azimuth (unless the risk from planned debris impacts was unacceptable on the present flight azimuth). If they do not, such flights would be required to implement an abort. This proposal is consistent with the ARC's recommendation to expand part 431 to include flight abort rules that apply when the vehicle is performing outside of its profile and is unable to reach a useful orbit or survive, and needs to be terminated prior to overflight of a populated area.

Proposed § 450.125(d) would establish the application requirements for gate analyses. Specifically, the proposal would require an applicant to submit a description of the methodology used to establish each gate or relaxation of a flight safety limit; a description of the measure of performance used to determine whether a vehicle will be allowed to cross a gate without flight abort, the acceptable ranges of the measure of performance, and how these ranges were determined; a graphic depiction showing representative flight safety limits, any protected uncontrolled area overflight regions, and instantaneous impact point traces for the nominal trajectory, extents of normal flight, and limits of a useful mission trajectories; and any additional products requested by the FAA to conduct an independent analysis when necessary to ensure that public risk criteria are not exceeded. The proposed application requirements are consistent with current practice under parts 417 and 431. 8. Data Loss Flight Time and Planned Safe Flight State Analyses

The FAA proposes to consolidate and update data loss flight times and planned safe flight states requirements in proposed § 450.127 (Data Loss Flight Time and Planned Safe Flight State Analyses).

Data loss flight time analyses are used to establish when an operator must abort a flight following the loss of vehicle tracking information. In § 417.3, the FAA currently defines "data loss flight time" as the shortest elapsed thrusting time during which a launch vehicle flown with an FSS can move from its normal trajectory to a condition where it is possible for the launch vehicle to endanger the public. This definition is unclear as to what constitutes a condition where it is possible for the launch vehicle to endanger the public. Given the overall approach to impact limit lines in § 417.213(d) and the treatment of data loss flight times in appendix A to part 417, section A417.19, the FAA has interpreted the definition to mean any impact on a protected area with debris greater than 3 psf ballistic coefficient.

With this proposal, the FAA would move the definition of "data loss flight time" from current § 417.3 to § 401.5 and update the definition to mean the shortest elapsed thrusting or gliding time during which a vehicle flown with an FSS can move from its trajectory to a condition where it is possible for the vehicle to violate a flight safety limit. An important change in the definition would be the replacement of "move from its normal trajectory" with "move from its trajectory." Computing data loss flight times initialized using normal trajectories or nominal trajectories would both be acceptable means of compliance with the proposed regulation, since using the former should be more conservative. This resolves the issue of varying practices at different ranges and provides additional flexibility.

In § 417.219(a), the FAA requires a launch operator to establish data loss flight times and a planned safe flight state. In § 417.219(b), the FAA requires that thrust be considered as a means of moving a vehicle towards a protected area, but some vehicles can also glide a significant distance using lift. Further, § 417.219(b) requires the data loss flight time to be relative to reaching protected areas, not flight safety limits. The requirements in § 417.219(c) also include a method of establishing the planned safe flight state that includes the subjective phrase "the absence of a flight safety system would not significantly increase the accumulated risk from debris impacts." Data loss times are currently computed in different ways at Federal launch ranges, with some initializing the computation from the nominal trajectory and some from trajectories within the normal trajectory envelope, sometimes referred to as "dispersed" trajectories.

Part 431 has no requirements related to analysis to establish data loss flight times or planned safe flight state. However, the one orbital RLV operation licensed to date employed an FSS and established data loss flight times.

The FAA's proposed § 450.127(a) would require an FSA to establish data loss flight times and a planned safe flight state for each flight to establish each flight abort rule that applies when vehicle tracking data is not available for use by the flight abort crew or autonomous FSS. Substantively, this proposal is consistent with the current rule in § 417.219(a). However, the FAA's proposal would update language to account for autonomous FSS and the use of the term flight abort in place of flight termination.

Proposed § 450.127(b)(1) would retain the data loss flight time analysis requirements consistent with § 417.219, but with the addition of gliding flight as a means of moving a vehicle towards flight safety limits (in lieu of protected areas in accordance with § 417.219). The proposal would replace the subjective method of establishing the safe flight state with a more straightforward method of analyzing when the vehicle's state vector reaches a state where the vehicle is no longer required to have a flight safety system. This is to avoid aborting a flight due to loss of track data during a phase of flight in which track data is not required to ensure safe flight. Thus, the proposal would encourage operators to avoid a flight abort, which often correlates with creating debris, due to loss of track data when in an area where flight abort is not required to meet the regulations.

Proposed § 450.127(b)(2) would require data loss flight times to account for forces that may stop the vehicle before reaching a flight safety limit, such as aerodynamic forces that exceed the structural limits of the vehicle. When more conservative methods are used, such as assuming an instantaneous turn towards the nearest flight safety limit, data loss flight times can be underestimated in that a vehicle could not physically perform the turn without breaking up. Data loss flight times that are unrealistically low create the risk of an unnecessary abort (and thus, an unnecessary debris event) if track is lost, since track may return and allow flight to continue if the data loss flight times are greater.

Proposed § 450.127(b)(3) would allow the computation of data loss flight times in real-time in lieu of only computations made preflight. This proposal would allow for a computation using the last-known state vector of the vehicle before track was lost. Proposed § 450.127(b)(3) would allow the computation of data loss flight times to be performed on the ground or onboard the vehicle, depending on whether a traditional command destruct or autonomous flight safety system is used.

In proposed § 450.127(c), the requirements regarding the planned safe flight state would be consistent with those currently in § 417.219(c), only generalized to apply to reentry as well as launch. Proposed § 450.127(c)(1) would update the § 417.219(c)(1) requirement using new terminology without any change to the meaning.

Proposed § 450.127(d) lays out the application requirements for data loss flight time and planned safe flight state analyses. Specifically, the proposal would require an applicant to submit a description of the methodology used to determine data loss flight times; tabular data describing the data loss flight times from a representative mission; the safe flight state and methodology used to determine it; and any additional products requested by the FAA to conduct an independent analysis. 9. Time Delay Analysis

For ELVs, § 417.221(a) requires a time delay analysis that establishes the mean elapsed time between the violation of a flight termination rule and the time when the flight safety system is capable of terminating flight for use in establishing flight safety limits. Section 417.221(b) requires the analysis to determine a time delay distribution that accounts for the variance of all time delays for each potential failure scenario, a flight safety official's decision and reaction time, and flight termination hardware and software delays which includes all delays inherent in tracking systems, data processing systems, display systems, command control systems, and flight termination systems.

The FAA has also required time delay analyses for RLVs under the current regulatory scheme. Specifically, § 431.39(a) requires an RLV license applicant to submit contingency abort plans, if any, that ensure safe conduct of mission operations during nominal and non-nominal vehicle flight. In practice, a time delay analysis has been necessary to ensure safe conduct of an RLV that uses flight abort.

The FAA proposes to streamline the regulations governing the analysis of time delay in proposed § 450.129 (Time Delay Analysis). Proposed § 450.129(a) would use language identical to § 417.221(a), except that the term "terminating" would be replaced with the term "aborting." The proposal would replace the list of time delay contributions prescribed in § 417.221(b) with a performance-based requirement in proposed § 450.129(a), that the time delay analysis would be required to determine a time delay distribution that accounts for all foreseeable sources of delay.

Proposed § 450.129(b) would list application requirements. Specifically, the proposal would require an applicant to submit a description of the methodology used in the time delay analysis, a tabular listing of each time delay source and the total delay, with uncertainty, and any additional products the FAA would request to conduct an independent analysis. 10. Probability of Failure

Proposed § 450.131 (Probability of Failure Analysis) would cover probability of failure (POF) analysis requirements for all launch and reentry vehicles. The proposal would also make application requirements clearer and implement performance-based requirements to address allocation to flight times and vehicle response modes. The proposed POF performance requirements would allow an operator to employ alternative, potentially innovative methodologies so long as the results satisfy proposed requirements such as valid input data.

Current regulations covering POF analysis requirements for ELVs are found in § 417.224. Part 431 does not have requirements for a POF analysis. Even so, a POF analysis is necessary to demonstrate compliance with the public risk criteria set for RLV operations in § 431.35(b).

Section 417.224(a) requires that POF analyses use accurate data, scientific principles, and a method that is statistically or probabilistically valid. For vehicles with fewer than two flights, the POF must account for the outcome of all previous launches of vehicles developed and launched in similar circumstances. If a vehicle has more than two flights, the POF analysis must account for the outcomes of all previous flights of the vehicle in a statistically valid manner. Section 417.224(a) does not address the use of data on partial failures and anomalies, which is a shortcoming the FAA seeks to correct. Section 417.224(b) defines failure to mean when a launch vehicle does not complete any phase of normal flight, or when any anomalous condition exhibits the potential for a stage or its debris to impact the Earth or reenter the atmosphere during the mission, or any future mission, of similar launch vehicle capability. The paragraph makes clear a launch incident or accident also constitutes a failure. Finally, Section 417.224(c) explains that previous flights begin when the launch vehicle normally or inadvertently lifts off from a launch platform and that liftoff occurs with any motion of the launch vehicle with respect to the launch platform.

Although the § 417.224 definitions have generally served the FAA and the industry well, § 417.224 lacks requirements to address allocation to flight times and vehicle response modes (VRMs), even though these allocations are necessary to determine the public risks posed by various VRMs at various times in flight. Given POF is a primary factor in any risk computation, it is impossible for an applicant to demonstrate compliance with the quantitative public risk criteria without an analysis to determine the probability of any reasonably foreseeable outcome, such as an on-trajectory loss of thrust or a malfunction turn ending in aerodynamic break-up.

The FAA would retain the substantive § 417.224 POF analysis requirements in proposed § 450.131, including the definitions of key terms such as "failure" and "previous flight". However, the proposal would apply to all launch and reentry vehicles. In addition, it would clarify the data a POF analysis must use to establish a valid allocation to flight times and vehicle response modes.

Proposed § 450.131(a) would retain the same substantive requirements regarding the an operator's estimation of the POF for vehicles with fewer than two flights. However, for vehicles with two or more previous flights, the proposal would change the § 417.224(a) provision by requiring that the outcomes of all previous flights of the vehicle or vehicle stage account for data on partial failures and anomalies including Class 3 and Class 4 mishaps. Thus, the proposal would require an analysis to account for partial failures and anomalies. These changes should improve the credibility of POF analyses by giving due credit to stages that succeed even though a subsequent stage fails. For example, consider a vehicle launched two times, with a failure during the second stage on the first launch and no failures during the second launch. For the third launch, the proposal would allow a probability of failure analysis to account for the fact that the first stage flew twice without a failure, while the second stage flew twice with one failure.

Proposed § 450.131(b) would retain essentially the same definition of "failure" used in § 417.224(b), with changes using the proposed mishap terminology (Class 1 or Class 2) and to cover other vehicles beyond ELVs.

Proposed § 450.131(c) would retain essentially the same definition of "previous flight" for FSA purposes, with changes intended to encompass all launch and reentry vehicles, including cases where an operator uses a carrier aircraft. Thus, "previous flight" for the purposes of an FSA would cover the flight of a launch vehicle beginning when the vehicle normally or inadvertently lifts off from a launch platform. Liftoff would still occur with any motion of the launch vehicle with respect to the launch platform. The FAA would clarify that this would include a carrier aircraft as a launch platform, and would include any intentional or unintentional separation from the launch platform. In terms of a reentry vehicle, the flight of a reentry vehicle or deorbiting upper stage would begin when a vehicle attempts to initiate a deorbit.

Proposed § 450.131(d), titled "Allocation," would establish performance requirements to address POF allocation to flight times and VRMs. The proposal would require that a vehicle POF be distributed across flight times and vehicle response modes consistent with the data available from all previous flights of vehicles developed and launched or reentered in similar circumstances; and data from previous flights of vehicles, stages, or components developed and launched or reentered by the subject vehicle developer or operator. Such data may include previous experience involving similar vehicle, stage, or component design characteristics; development and integration processes, including the extent of integrated system testing; and level of experience of the vehicle operation and development team members. These requirements were not in § 417.224 or part 431. In this context, phases of flight would be defined by planned events affecting the vehicle configuration and its failure rate, such as ignition, first stage flight, stage separation, second stage ignition, second stage flight, payload fairing separation, etc. This proposal would require what is already necessary and thus done in current practice.

In proposed § 450.131(e), the FAA would require that a POF allocation account for significant differences in the observed failure rate and the conditional failure rate. The conditional failure rate represents the failure rate conditional on the vehicle or subsystem having survived, without a failure as defined earlier, to a given time in flight. The observed failure rate is the product of the conditional failure rate and the reliability function, which is commonly defined as the probability that the vehicle or subsystem has not failed prior to a given time in flight. For high reliability systems where the reliability function is close to one (by definition), the observed failure rate can be approximated as the conditional failure rate. If the overall vehicle or stage POF is below 10 percent (over the entire period of time corresponding to a phase of flight), then this simplified approach produces a relative error less than approximately 0.5 percent, which is generally not considered a significant difference. For lower reliability systems, this approximation does produce a significant difference between the observed failure rate and the conditional failure rate. Here again, the proposal would clarify what is already necessary and thus done in current practice.

Proposed § 450.131(e) would also require that a POF analysis use a constant conditional failure rate for each phase of flight, unless there is clear and convincing evidence of a different conditional failure rate for a particular vehicle, stage, or phase of flight. Thus, the proposal would require a POF analysis to assume that the conditional failure rate can be represented as a piece-wise constant function of time for each phase of flight, absent clear and convincing evidence to the contrary. The points that define transitions to a potentially different conditional failure rate must include staging events or other vehicle configuration changes, such as ignition of other engines or rocket motors. In some cases, the FAA anticipates that there will be sufficient evidence to justify a different failure rate, for example during a start-up or shut-down/burnout transient for a rocket motor compared to steady state operation of a stage, engine, or motor.

Proposed § 450.131(f) would lay out the FAA's application requirements for POF analyses that address the proposed methodology, assumptions and justification, input data, and output data. An applicant would also be required to provide a complete set of tabular data and graphs of the predicted failure rate and cumulative failure probability for each foreseeable VRM. The proposed requirements are consistent with current practice to the extent that any valid FSA must include the probability of failure assigned to each VRM as a function of time into flight. 11. Flight Hazard Areas

The FAA proposes to streamline its regulations on flight hazard area in proposed § 450.133, applicable to all launch and reentry vehicles. The FAA would codify its working definition of "flight hazard area" to mean any region of land, sea, or air that must be surveyed, publicized, controlled, or evacuated in order to protect the public health and safety and safety of property. An FSA would include a flight hazard area analysis to identify regions of land, sea, or air where an operation poses a potential hazard to the public. The proposal would reduce the size of the regions of land, sea, and air requiring hazard warnings from normal flight events and would reduce the size of regions requiring surveillance prior to initiating a commercial space transportation operation. These changes would be consistent with practices at Federal launch ranges.

The current FAA regulations most pertinent to flight hazard area analysis are found in Sec. § 417.107(b) (Flight safety) and 417.223 (Flight hazard analysis) for ELVs, and Sec. § 431.35(b) (Acceptable reusable launch vehicle mission risk) and 431.43(b) (Reusable launch vehicle mission operational requirements and restrictions) for RLVs. Both the ELV and RLV regulations require flight hazard areas to protect against hazards posed by vehicle malfunctions (e.g., an in-flight break-up) and normal flight events that create hazards (e.g., any planned jettison of debris, launch vehicle components, or vehicle stages).

The FAA currently sets requirements to warn of, or limit the operations of, ELVs and RLVs in regions where planned debris impacts are likely, for example, due to jettisoned stages. In § 417.223(b), the FAA currently requires flight hazard area analyses to establish ship and aircraft hazard area warnings to mariners and airman in regions that encompass the three-sigma impact dispersion area for each planned debris impact. Similar language appears in § 431.43(b), which states that a nominal landing location is suitable if the area of the predicted three-sigma dispersion of the vehicle impacts can be wholly contained within the designated location. In the 2000 final rule, the FAA explained that it intended the three-sigma to refer a location where the vehicle or stage landing would be contained 997 times out of 1000 attempts, or 99.7 percent probability of containment.\200\ Hence, these regulations used the term "three-sigma" to refer to a univariate Gaussian distribution,\201\ despite the fact that impact dispersions are bivariate, and not necessarily Gaussian. Notably, neither § 417.223 nor § 431.43 stipulate whether these warning areas must account for all debris or only debris capable of causing a casualty. There is evidence that the separation of large stages can liberate small fragments with a negligible probability of creating a casualty, depending on the nature of the exposed population. For example, people in aircraft are often more vulnerable than people on the ground because a fragment that impacts an aircraft has a much higher kinetic energy due to the velocity of the aircraft.

\200\ 65 FR 56618 (September 9, 2000), at 56629.

\201\ Gaussian distribution (also known as normal distribution) is a bell-shaped curve, and it is assumed that during any measurement values will follow a normal distribution with an equal number of measurements above and below the mean value.

Both the ELV and RLV regulations require public risk controls, such as evacuation or surveillance, to ensure that no individual member of the public is exposed to greater one-in-a-million (1 x 10-6) PC, irrespective of their location on land, sea, or air, to satisfy risk criterion in Sec. § 417.107(b) and 431.35(b). The part 417 regulations address the identification and surveillance of flight hazard areas explicitly in several sections, including Sec. § 417.111(b)(5), 417.121(f), and 417.223 as discussed below. Part 431 regulations do not expressly address flight hazard areas. However, the preamble to the 2000 final rule stated that the individual risk limit of 1 x 10-6 PC would dictate whether or not an area must be evacuated for launch or reentry activity along that trajectory to occur safely, and clarified that limit applied for any person not involved in the licensed activity. Hence, the current RLV regulations clearly intended the evacuation, and surveillance by inference, of any area where a person not involved in the licensed activity would otherwise experience more than 1 x 10-6 PC.

Continue reading ↓ 800 of 1213 passages shown