7.4 Jet engines: regulation and the design space

A commercial jet engine is a useful narrow example for examining regulation. Its compressor, combustor, turbine, nozzle, control systems, and supporting structures must work as an integrated system. Manufacturers balance thrust, fuel efficiency, weight, manufacturing cost, durability, maintainability, and reliability while protecting against fire, component failure, lightning, ice, hail, rain, dust, and bird ingestion.

Failure of a commercial jet engine can have dire consequences, most obviously for the passengers and flight crew. It is therefore exactly the kind of case in which government regulation might appear critical. At the same time, the consequences do not fall only on those aboard the aircraft. If an engine is shown to have been poorly designed or manufactured from a safety perspective, its manufacturer may face deaths and injuries attributed to its product, liability, loss of reputation and future sales, customer inspections, fleet groundings, and the cost of repairing or replacing engines already in service. Even without regulation, a jet-engine manufacturer therefore has exceptionally strong reasons to treat safety as essential.

The nature of jet-engine regulation

In the United States, aircraft engines are type-certified principally under 14 CFR Part 33. The European Union uses the corresponding CS-E certification specifications. Other rules apply to installation of the engine in an aircraft, production, maintenance, and continued airworthiness.

These requirements generally establish safety outcomes and demonstrations rather than prescribe a complete engine design. They address matters including structural integrity, vibration, overspeed, rotor containment, fire, endurance, icing, rain and hail, bird ingestion, control-system failure, and hazardous effects following component failure. For example, bird-ingestion requirements prescribe bird masses, impact conditions, engine power, and acceptable consequences. They do not prescribe the geometry of the compressor, the material used for every blade, or the control logic by which the manufacturer achieves the required result.

Certification therefore samples particular points in a much larger design and operating space. A successful test does not demonstrate safety under every combination of age, manufacturing variation, maintenance condition, atmospheric conditions, impact location, electrical failure, software state, and simultaneous fault. Some requirements also permit loss of engine thrust provided that the failure does not produce a hazardous engine effect. Safety of the aircraft may depend on containment, redundancy, the remaining engine, aircraft systems, and crew response rather than on every engine continuing to operate after every event.

Is there evidence that regulations cause manufacturers to produce safer engines?

The evidence required for certification does not prove that the manufacturer would have designed a less safe engine without the rule. Establishing that causal claim requires comparison with a credible unregulated alternative.

Accidents followed by new regulations are not sufficient evidence. An accident before a requirement existed shows that the previous combination of engineering, knowledge, incentives, and voluntary standards did not prevent that particular failure. It does not show that industry would have left the problem unresolved, nor that the later rule caused the solution. Likewise, an accident involving a certified engine does not show what the outcome would have been without certification.

Bird-ingestion data show an association, but do not provide evidence that regulation caused the improvement. Following accidents and analysis of events in commercial service, the FAA progressively increased the severity and scope of its bird-ingestion requirements. The accidents and service evidence were available to manufacturers independently of the later regulations. A later FAA and industry study compared engines certified under successive amendments and found that engines certified under more recent requirements had a lower proportion of ingestion events resulting in power loss than engines certified under the earliest requirements. The report itself notes the limited period of service data. Newer engines also differ in technology, materials, manufacturing, maintenance, and accumulated industry knowledge. Because regulatory amendments and those other changes occurred together, the comparison does not identify which of them caused the improvement.

Causal evidence could come from a comparison with otherwise similar engines or manufacturers exposed to the same safety information but not to the regulatory requirement. It could also come from an engineer or manufacturer identifying a specific safety improvement which was made because of a regulation and which would not otherwise have been made. Such testimony would be stronger if supported by contemporaneous design records, cost decisions, or internal analysis. It would also need to distinguish an improvement to the engine's safety from additional testing or documentation performed only to demonstrate conformance.

Is there evidence that manufacturers go beyond the regulations?

There is direct evidence that manufacturers sometimes test beyond an applicable minimum. The CFM56-5B used on US Airways Flight 1549 passed its applicable bird-ingestion certification tests, and the manufacturer also performed additional testing required by the French authority. The actual encounter involved birds larger and more numerous than the certification standard. This is evidence that its testing exceeded the applicable FAA minimum. Because the additional testing was required by another government authority, it is not evidence of action beyond government regulation generally. FAA analysis of US Airways Flight 1549.

Before the TACA Flight 110 event, the CFM56 had demonstrated approximately four times the water quantity required by the certification rule. The test method nevertheless failed to reproduce the critical way in which hail and water could concentrate in the engine core. The large nominal margin did not cover the relevant physical condition. Manufacturer and industry investigation produced design changes and revised test methods. Manufacturers began using the revised standard from approximately 1989, whereas the corresponding rule was issued in 1998. FAA analysis of TACA Flight 110.

The nine-year interval is evidence that manufacturers responded without waiting for formal enactment. The FAA account does not establish when manufacturers first anticipated the revised certification requirement or when certification authorities began applying it informally. The interval therefore does not establish that the response was independent of regulatory influence. These examples show conduct beyond some formally applicable written requirements, but do not show that regulation had no effect on the design decisions.

Is there evidence that regulations create unhelpful conformance costs?

For the proposed rain-and-hail ingestion requirements that became part of § 33.78, the FAA recorded historical industry estimates of incremental certification costs ranging from approximately $250,000 to more than $500,000 for a new engine model. One manufacturer estimated approximately $50,000 for a derivative engine. These figures are not estimates of the total cost of certifying an engine. The rule allowed multiple means of compliance, including analysis where appropriate, rather than requiring physical ingestion testing in every case. The FAA cost analysis attributes the estimated cost to compliance work such as engineering, analysis, test facilities, and physical testing. FAA inclement-weather rulemaking analysis.

For scale, the principal airplane rain-and-hail provision in § 33.78, together with the atmospheric specifications in Appendix B, accounts for only about 2% of Part 33 by word count.

In a separate production-certification rulemaking, industry participants objected to mandatory marking of individual parts on the ground that it would impose high costs without a verifiable safety benefit. The FAA revised its proposal to permit more flexible methods of identification. This does not measure the net overhead of engine certification, but it is a concrete example in which a proposed form of conformance was challenged for lacking a verifiable safety benefit and was changed before being imposed in that form. FAA production-certification rulemaking record.

A search of FAA engine-certification rulemaking records and aviation-safety reports from GAO, the Department of Transportation Inspector General, and NTSB found no analysis demonstrating that an existing jet-engine conformance requirement imposes costs without materially improving safety assurance.

Is there evidence that regulators lack sufficient expertise or knowledge?

In 1993, the United States General Accounting Office reported that manufacturers' commitment to safety and technical expertise had kept design-related safety problems to a minimum, while the FAA could not effectively oversee the certification process because its certification workforce had limited technical expertise, high turnover, and insufficient guidance, technical assistance, and training. GAO aircraft-certification review.

A 2005 GAO survey found that only about half of FAA safety inspectors believed, to a great or very great extent, that their technical knowledge was sufficient to perform their jobs properly. The corresponding estimate among aircraft-certification inspectors was 68%, with a 95% confidence interval from 51% to 85%. GAO technical-training review.

In 2020, GAO found that the FAA Office of Aviation Safety did not assess organisation-wide competency gaps among its inspectors and engineers on a recurring basis. GAO reported that this prevented the agency from obtaining an organisation-wide view of whether its workforce and training covered the competencies needed for safety oversight and new aviation technologies. GAO aviation-safety workforce review.

GAO has also reported that inconsistent interpretation was considered one of the most significant problems affecting aircraft certification. An FAA and industry committee identified unclear regulatory requirements and inadequate, nonstandard training in developing regulations, applying standards, and resolving disputes among the causes. GAO review of certification and approval processes.

The Department of Transportation Inspector General found that FAA guidance did not adequately address the integration of new technology into an existing aircraft model during certification of the Boeing 737 MAX. It also found that the FAA did not have a complete understanding of Boeing's MCAS safety assessments before the first accident and identified needs for additional expertise in systems engineering, human factors, and software. This finding concerns aircraft certification rather than jet-engine design, but demonstrates that the certification authority has not always possessed the knowledge or guidance needed to assess a manufacturer's technology. DOT Inspector General certification review.

There is also propulsion-specific evidence. Following US Airways Flight 1549, the National Transportation Safety Board concluded that the bird-ingestion certification requirements then in effect did not provide adequate safeguards against the ingestion threats faced by future large turbofan engines. NTSB accident report. More recently, GAO reported that representatives of electric-aircraft manufacturers and an aviation engineering firm encountered limited numbers of FAA staff with expertise in electric propulsion, limited standardisation in the certification process, and gaps in the standards applicable to novel engines. GAO electric-aircraft certification review.

These findings do not establish that every regulator lacks expertise or that every engine requirement is technically defective. They do provide independent evidence of insufficient specialist knowledge, outdated or incomplete guidance, unclear requirements, and inconsistent application within the institution responsible for certification.

Conclusion

The available evidence provides little basis for concluding either that regulation improves jet-engine safety beyond what manufacturers would otherwise achieve or that it has no safety benefit. Conformance costs are observable, while an incremental safety effect is not.