By: Mark H Goodrich – Copyright 2013
Everyone has heard about the astronaut who related that his thoughts before launch were centered on the fact that he was sitting atop a vehicle with all component parts having been designed and built by the lowest bidder. Pilots smile at the story, but few realize the aircraft they spent their lives flying were similarly designed and built for the most part to the lowest standards allowed within the certification regulations – technically described as “minimum standards” – and sometimes to even less.
The concept of establishing minimum standards was, for some 60 years, a principal pillar beneath the regulations defining requirements for certification of aircraft. From the start, it was seen by government officials that establishing and maintaining precise requirements for the broad spectrum of aviation products was an enormous task. Initially, it would have required experienced design engineers to delve into the intricacies of structure, materials, aerodynamics, performance and systems for each type and model of aircraft. Next, those details would have required drafting with specificity into an airworthiness code, including an ongoing effort over time to ensure the regulations were continually updated to reflect advancing states-of-the art. Finally, regulators would have been required to maintain a large agency staff of engineers knowledgeable and experienced enough to oversee, inspect and ensure conformance of aviation products with those regulatory standards. This daunting scenario was analyzed as beyond the technical, economic and administrative capabilities of a government agency.
Instead, it was believed that manufacturers would be in the best position to make decisions about standards for intended applications, with a natural incentive to improve standards as allowed by advancements in the technological states-of-the-art, and a willingness to meet the public responsibilities with which they were entrusted. It was assumed that industry would see the application of standards aligned with the intended use of each product, and the evolution of elevated standards over time, to be in its long-term interests through lower insurance rates, higher customer satisfaction and improved safety in operation. It was reasoned that manufacturers would know that the same minimum standards are never logically applicable across a broad range of products from gliders to jets.
The use of a “minimum standards protocol” limited the necessary participation of the regulator to defining only the minimums below which no product could be certified, with reliance placed upon manufacturers to generally design above that minimum standard, using technological and economic feasibility analyses to define currently reasonable standards for an intended use profile. For example, even though two airplanes might be certified under transport category regulations, the standards applied to one intended for operation at low altitudes and speeds over a use pattern measured in hundreds of hours a year, might be quite different than those to one that would operate at high altitude and speeds close to mach over an annual use pattern measured in thousands of hours. Further, advancing standards as allowed by improved materials, processes and techniques would not be artificially restricted by the regulations, as they would define only minimums, exceeded by substantially all products.
To establish the framework for the intended methodology, substantive regulations from 1940 forward stated that products would be required to meet at least the minimum standards set forth. Accompanying interpretive regulations said that “minimum standards do not constitute the optimum to which the regulated are expected to strive” in the design, materials and manufacture of aviation products. This language was set forth in various parts of the aviation regulations for almost 60 years.
In addition to design and manufacturing, it was also determined that testing to ensure conformance with reasonable standards, and acceptance for certification as an agent for government, would be delegated to the manufacturers. The regulator would do little or no actual inspection. The regulatory language said, “A statement of conformance… [from the manufacturer] will be accepted as sufficient indication that the applicable requirements have been fulfilled”. Few seemed to grasp that government was thereby authorizing the fox to legally oversee the chicken coop. Although originally termed the Delegation Option Authorization (DOA), the regulator’s executives became so embarrassed by repeated references to its acronym in the wake of misconduct by industry, that the functions of DOA were morphed into a delegation system renamed Organization Designation Authorization (ODA). Despite the change in form, the problems with its substance remained the same.
While the concept of minimum standards sounded reasonable in the abstract, it naively over-estimated the level of responsibility that industry would bring to the process. Indeed, manufacturers used mostly the minimum standards as design targets from the start.
After 1946, the rate at which new designs, materials and manufacturing processes were developed began to accelerate. Research – mostly from the military sector – provided data upon which significant improvements could be based. And yet, the civilian sector lagged, cherry-picking improvements that might increase payloads or performance, and those that might reduce manufacturing costs, but eschewing those related to safety, ergonomics, crashworthiness and maintainability for continuing airworthiness. Having successfully lobbied against legislation allowing the government to order a recall of defective aviation products, industry saw little risk in the failures of its products to perform in service, as the resulting economic burdens fell to aircraft owners and operators.
Testing individual units – even those mass-produced to a “common” standard – always results in some scoring above and some below the design target. This is referred to within the engineering community as “test data scatter”. When the minimum standard of the regulations is the design target, the natural result is that a percentage of individual units test below that minimum. In practice, management often responded to test data scatter by directing engineering to average the data. The obvious rhetorical is, “If four units test below the minimum standard, but five test above it, can one reasonably argue that those scoring below the minimum are actually above it in regulatory terms due to averaging?” The obvious answer from the standpoint of legality is “No”. Every unit must test above the minimum standard. Further, quality control and assurance must reasonably predict that every unit manufactured will meet or exceed at least minimum standards. In addition, good engineering practice will seldom support design and manufacure to only “minimum standards”, technical legality notwithstanding.
In practice, many aviation products have been certified despite that significant percentages of those units tested below the minimum standard. The argument of industry that averaging data is a legitimate engineering practice is even more absurd when the standard at issue is a materials limitation, because it implies that jury-rigging test results will somehow suspend the laws of physics. I investigated a series of engine failures following what can best be described as “metallurgical meltdown”. After months of fighting the legal barricades of the manufacturer, I discovered that 35% of its tests resulted in temperatures exceeding the limitations of the engine metallurgy by hundreds of degrees. Yet, the engine was self-certified as compliant “on the average”. As a predictable consequence, many subsequently failed in service, some at critical points. Airplanes caught fire, crash landings resulted, people were killed and injured, and airplanes were destroyed. The resulting airworthiness directive grounded over 700 aircraft, and imposed millions of dollars in expenses upon the owners and operators to purchase, from the manufacturer, a minimum standards bandage that could never do more than partially fix a fundamentally flawed design.
Using minimum standards as design targets overlooks their general applicability across a category and class of aircraft. Proper engineering practice requires that a particular design be set by the specifics of the intended application first, and then verified as meeting or exceeding the minimum standards in the subsequent testing phases.
Even when insufficiencies were revealed by testing, there were procedures by which waivers or deviations could be sought. One commonly used mechanism allowed a manufacturer to argue that, although a particular element failed to meet even the minimum standards, it nonetheless provided an “equivalent level of safety”. Using this regulatory side-step, many aviation products have been certified to standards below the regulatory minimums, and its use has not been restricted to minor components. Wing spars, engine pylons, fuel tanks and other fundamental components that were designed to a minimum standard and thereafter tested below it, have nonetheless been granted equivalent level of safety status, usually in the wake of administrative appeals and pressure on the regulatory agency from politicians and industry, pleading that it would be too expensive, hurt small investors in the company stock and cost jobs, if a redesign was required at a late stage in the project.
The failures of industry to advance the standards in practice over time also became ingrained in the process. As but one example, the minimum standard for the strength of passenger restraint systems was set in 1952 at 1,575 pounds, reflecting a longitudinal acceleration load of 9 g for a 175 pound occupant, and mimicking the minimum standard for structural strength of the cabin. That military research of the 1950s validated the limits of human survivability in the range of 60 g over a 200 millisecond pulse was ignored by industry while stifling arguments by universities and research facilities for an increase in standards to levels where passengers would not be needlessly killed in survivable crash landings or aborted takeoffs. Indeed, while making 40 g seats and restraint systems for agricultural aircraft, manufacturers simultaneously argued that such standards were beyond a feasible state-of-the-art, leaving the 9 g longitudinal standard in place even today, 60 years later.
When highway traffic safety officials mandated a 10,000 pound requirement for restraint systems in automobiles, webbing manufacturers ceased making the weaker material used by aircraft manufacturers. The response of some aircraft builders was to downgrade the attachment hardware, in order to ensure that the restraint systems in their aircraft would continue to fail at 1,575 pounds. When challenged, industry argued that to admit a higher standard was feasible created a potential for liability, despite that the rules of evidence in federal courts provided that undertaking voluntary remedial measures to improve product safety could not be used against a manufacturer in legal proceedings. In short, industry was instinctively reacting to the tribal lore passed between and among executives, rather than objectively considering the safety interest of those who rode in their products, the facts and the law. Long hounded by criticism that it was not heeding the regulatory requirement for use of appropriate product-specific standards above the minimums, including the elevation of standards over time, industry began to pressure political appointees of the regulator, resulting in a “back room deal” and cancellation of the order by which the concept of being expected to exceed minimum standards was established. In essence, that removed the ability to legally challenge certification of a product to only minimum standards where higher standards were feasible and appropriate for the intended use. Industry was no longer required to design and build aircraft to a reasonable standard, but rather only to whatever standard it could negotiate with the assistance of its political allies. As with so much else in the age of the corporatocracy, economic and political power became more important than the laws of physics, standards of sound engineering practice, and interests of public safety.
As a consequence, the regulator now carries the burden to try and bring certification requirements up to a reasonable state-of-the-art, defying the logic by which it was originally determined that government could not effectively or economically maintain the regulations to a state of currency regarding issues of design, materials and manufacture. Absent both the legislative authority and budget to hire and train the necessary engineering talent, the agency has found itself with ever less ability to maintain certification standards.
Because the bulk of testing and certification of their own products has long been delegated to the manufacturers, the regulator would now find it hard to obtain the political leverage necessary to reacquire those functions, even if it had the engineering talent and budget to do so. Indeed, the resulting political firestorm from such an effort would create a glow on the horizon. Industry employs thousands whose jobs are creating log jams in the processes of rulemaking, and the time required to successfully promulgate even a watered-down version of a regulation is measured not in months or years, but decades.
One of the most disturbing trends within the aviation manufacturing industry has been the increasing influence of corporate power to skew the regulatory process. It has always been the case that favorable treatment from a regulator could be garnered through promises of employment following retirement from government service. But, the anti-regulatory political environment of the 1980s spawned a whole new level of conspiratorial malfeasance between the legislature, agencies of government and industry. Mergers and acquisitions brought even small manufacturers into worldwide corporate conglomerates with enormous financial resources, able to unleash armies of lawyers and lobbyists into the halls of the national legislature and the regulatory agencies of government, and with historically unparalleled capacity through contributions to influence the votes of individual legislators and control the agendas of national political parties.
Individuals being considered as prospective appointees for senior positions within regulatory agencies were suddenly compelled, as a condition precedent to formal appointment, to first circulate among major corporations and trade associations, kissing rings and promising to continue the laissez faire approach to oversight. Corporate legal departments – often renamed with outlandish hypocrisy using such terms as “product integrity counsel” – became repositories to hide the very design and testing documentation relied upon by their company for self-certification of its own products, thus allowing manufacturers to argue that those documents were shielded from discovery under a sham of an apparent attorney-client privilege, in addition to their long-standing arguments that such documents were properly withheld from public scrutiny as trade secrets.
All of the above helped set the stage for corporations – the management of which were becoming ever more remote from their products as acquisitions and mergers increased their size and bureaucratic inefficiencies, and hired in any event as balance sheet manipulators expected to increase share value by any means possible – to increasingly use financial and political leverage to obtain favorable treatment from government in the operation of their businesses. As a final and ultimate insult to the integrity of the certification process, the political right packed the federal courts with business-friendly ideologues in judicial robes, appointed for life and without oversight of any kind. The “fix” was in.
In response to specific issues arising from high-profile crashes, the regulator was able to use emergency rulemaking, usually through “special federal aviation regulations”, to work around some of the obstruction from industry, but this increasingly resulted in an internally inconsistent hodge-podge within the regulatory structure. To try and fill gaps created, the concept of “special conditions” was used to prescribe special standards regarding an issue when the regulations did not contain “adequate or appropriate safety standards”. But this has also been a piecemeal, inefficient and broadly ineffective alternative to a logical and well-constructed system of regulation.
Complicating the entire scenario has been a contemporaneous agency effort to computerize its system of orders, handbooks and directives. The initial combining of materials into a common database revealed a broad range of internal inconsistencies across the panoply of agency policies and procedures. As long-issued documentation was cancelled, gaps were the natural result, creating what even agency employees admit is an uneven system of “policy by memorandum”. The true facts are that no one outside the carpeted hallway on the top floor of the headquarters building has any knowledge regarding most agency policies, or how to obtain the relevant memoranda even when they exist. Predictability and equality of treatment have been lost. Policy decisions are often made sua sponte by low-level agents with little knowledge and less experience. Management by crisis is the order of the day.
Industry continues to use the often insufficient minimum standards of certification regulations as design targets and, now as a matter of course, resorts to multiple requests for deviations and equivalent level of safety determinations when the resulting designs fall short of meeting even those minimums. Corporate managers, who often see the end of their responsibility as short term results on a balance sheet, find it increasingly easy to pressure engineers, inspectors, and quality assurance personnel to place continued employment above what they know to be professional malfeasance. The wake of large corporations is littered with the terminated and blacklisted careers of those who would not yield professionalism to company pressure. On the other side of that same ledger, managements are increasingly populated by those willing to be “team players”, even at the sacrifice of their personal and professional integrity, placing corporate profits above public responsibilities.
An unfortunate consequence of an increasingly concentrated manufacturing industry worldwide is that regulating authorities are ever more subject to the political influence inherent when a handful of manufacturers have enormous financial power over national economies, international balances of payments, and employment. When large employers and political contributors apply pressure, it becomes all too expedient for regulators to bend with the prevailing political winds, and allow certification of products that do not meet sound engineering practices, reasonable standards or even the letter of the minimum standards within the airworthiness code.
The concept of using a minimum standards protocol was not the curse – it was not only valid, but the best vehicle for maintenance of a strong certification system. What it needed and lacked was a willingness by the regulator to enforce the application of reasonable standards above the minimums as required, including the imposition of significant penalties for those who falsified test data and otherwise placed profits above safety. In the absence of a strong oversight function, corporations pushed their perceived interests to the maximum extent possible. In the final analysis, implosion of the aircraft certification system was the result of weak regulatory managers being more interested in their resumes and post-government employment than in meeting their public responsibilities to stand firm for integrity and safety against the forces of industry and politics. As a result, the agency is now riddled with a self-inflicted, advancing cancer. The winner in the struggle of the corporatocracy versus public interest has been nothing more or less than short-term industry profits. The losers are many. In violating the proscription against serving two masters, the regulator has appeased moneyed and political interests, but breached its public duty to maintain the integrity of the aircraft certification system.
Mark H. Goodrich – Copyright © 2013
The Curse of “Minimum Standards” was first published in the November 2013 Issue (Vol 10 No 3) of Position Report magazine.