Regulation – John Vincent

Qualitative and Quantitative

Do I need to define these two? Yes, because they are not words that automatically come up in everyday conversation. Not even in technical conversations.

A qualitative approach to an argument deals with words, meanings, and experiences. Not so much opinions but defensible objective judgements based on best available information.

A quantitative approach to an argument deals with analysis, numbers and statistics. A scientific approach rooted in believable evidence that is reliable, repeatable, rational deduction.

In the aviation safety world, we use both. It’s my view that one is not subordinate to the other. However, the persuasiveness of one is greater than the other. At the end of the day, persuasiveness matters. If a key decision is to be predicated on an argument, then how it is constructed and presented matters a lot.

Many is the time when I have sat in a room listening to a person make a case in as elegant and enthusiastic a way that they can, but the results have been less than impressive. I’ve made the mistake myself. Coming prepared with a barrage load of sound evidence accompanied by a flaky presentation that misses the mark.

That is why the order needs to be thus. Establish a qualitative argument supported by quantitative information. Not the other way around.

In the field of aircraft system safety assessment this is vital. Probabilistic analysis is a powerful tool. Relating numbers to human experience is not so easy. So, detailed charts and logic diagrams sprinkled with numbers are unlikely to be enough on their own.

Stating an objective is met by handing over a pile of calculations is half the job. However impartial and truthful the numbers may be, arguments are not made until they are explained.

Many is the time when I have lifted a large report from my in-tray, studied it and concluded that there’s something missing in that gap between the QED and the statement of compliance.

Maybe this an element of safety regulatory psychology. It’s difficult to approve something if it becomes difficult to explain to others what’s been done. That evidence trail needs to be explainable. And easily explainable to a wide audience.

This doesn’t change with digitalisation. Replacing large piles of paper with large files of results puts the reader in the same position. Impenetrable evidence needs understandable explanation. Presenting a qualitative argument supported by quantitative information is a good way to go.

What’s next?

Brexit hasn’t gone away. Hushing up the debate doesn’t solve a problem. It makes it worse. Trying to run a marathon whilst carrying a weighty load of baggage on our backs is never going to get easier. Knowing that those weights can only get heavier makes it even worse.

If I remember rightly, which I do, those promoting this failed enterprise talked of a Singapore-on-Thames[1] like Britain that would stroll the world like a colossus. Free of entanglement with our neighbours. We’d be able to command cheap goods in the shops for all. Costs would go down and standards of living would go up. A mighty English lion would roar, and the rest of the world would tremble. The deception was criminal. It still is. What’s been delivered is the exact opposite. It’ tragic.

Now, we have a Prime Minister (PM) who must go out of his way to defend the Union. The United Kingdom (UK). Making passionate speeches about the world his parents inhabited. Kindling nostalgic emotions to emphasis the government’s commitment to holding our Union together.

If the Brexit vote was motivated by a public concern over immigration and freedom of movement, then what the UK has as a result is significantly worse than what we had during our membership of the European Union (EU). I know these subjects remain of great concern, but the direction of travel post-Brexit is 180 degrees from where we need to be.

Listening to industry in the UK amongst the issue that make them very unhappy, Brexit is never far from the top. The uncertainty and red tape that persists makes trading difficult. Finding, training, and retaining the skilled workforce they need has got harder. Costs keep rising as UK influence over world affairs weakens. Brexit is like a weighty millstone around our necks.

It would be nice to think that the advocates of Brexit have a solution to the problem that they caused. However, painful as it maybe, if you listen to them all they do is carp. Point the figure of blame. Then propose even more Brexity type ideas that mean digging an even bigger hole for us.

There’s hope. At least hope in rhetorical terms. This week, the PM has been floating a set of possibilities[2]. It’s characteristically vague. I don’t criticise him for the caution, given the volatility the political landscape and the financial boost his opponents have gained.

A decade ago, I continue to think that the Leavers did not win the Brexit vote. It was lost by the Remainers. The Leave campaign deployed their arguments with passion and marketing knowhow. Remainers, of which I was a guilty party, majored on logic and rational debate. Sticking to the facts. Sadly, being as attractive as an economics lecture by the gloomiest professor in town. Besuited business highbrows boring the pants off the public.

That’s the dilemma. We know, not only by facts, but now by experience that Brexit was a massive mistake. We also know that pure facts alone do not cut through however conclusive the arguments may be. Charts and statistics have their place only when they move the heart.

I hear this with industry people I talk to. It goes like this; surely the public can see that we need to reengage with Europe. Especially with the conflicts that are raging around the globe. Those words that – we are stronger together – have never been truer.

If there is another UK referendum soon, then the case for being an EU member must be an enthusiastic and visibly patriotic one. No castigating former Leave voters. Such a national campaign must be forward looking, positive and inspiring. An emotional anthem to unite us around the cause. A real world approach to the real world.

[1] https://www.culs.org.uk/per-incuriam/brexit-could-singapore-on-thames-become-reality

[2] https://www.bbc.co.uk/news/articles/cmlyewr0lnj3o

Servicing the future

Extrapolation is a dodgy game. That is to take a set of figures from past performance and then projecting them into the future. All too often, even in the world of artificial intelligence, a linear projection is deemed sufficient. Having a mass of collected data and then plotting a line that nicely fits somewhere in the middle. Assuming that there’s a value that is atypical of the accumulated data set.

It’s well accepted that predicting the future is fraught with assumptions. Real life is multidimensional and full of combinations and permutations of possibilities that are off the charts. However, everyday there is enough stability to make planning a useful activity.

There’s the phenomenon of the step change. Overnight a new law may be applicable. In the morning a new system may be switched on. In the evening, a major crash may stop transport links for hours. Events, both predicable and unpredictable throw a spanner in the works.

Trying to predict future workforce needs based on current facts and historic trends is something every major industry must do. I say “must” accepting some struggling and pathological organisations survive on a day-to-day basis.

So, if I start with a 100 people and expect a growth of 5% each year, it’s not unreasonable to think that I might need 105 people next year. It’s surprising how often this basic calculation is presented. Same with the predicable response from a chief executive who would prefer an improvement in productivity. Deliver the same output with the same, or less, people.

Predicting one year ahead is a must. Predicting 10-years ahead is nice. However, in certain industries where, say the products have long lives or the demand for services isn’t going away a long-term projection has meaning.

The search for productivity goes on. Digitisation has made inroads into the paper-based world of a couple of decades ago. I remember when I first held a digital camera in the office. In the mid-1990s it was a whizzy bit of kit to be loaned out on rare occasions.

Artificial intelligence, and all the technology wrapped around it, are reshaping business strategy, risk assessments and workforce planning. Everyone is seeking a competitive advantage.

This morning when I woke up, I was introduced to a new word. It was a BBC radio broadcast called: Rethink: manufacturing or services?[1] It doesn’t exactly roll off the tong. It’s servitization.

The premise is that the line between organisations manufacturing products and those who provide services isn’t a ridged one. Progressively, whatever the product, train, plane or automobile, they use masses of data not only to make but to operate. The data flows rapidly backwards and forwards between maker and operator. And lots of it too.

So, if I had to estimate the number of service engineers that would be needed in the next 10-years, I’d better be aware that their jobs are going to change dramatically. Radically.

However, overall will the numbers of people needed by an industry decrease or not? Or is it just that the nature of work will change for roughly the same overall numbers? That’s not so easy as a straight-line graph. Maybe, as productivity increases so it becomes possible to take on new tasks that previously were not viable. The performance of a product continuously improves.

I’d buy a service and not so much a product. I’ll buy 100,000 miles, not a car.

[1] https://www.bbc.co.uk/programmes/m00325dd

Be fair

Here’s an example of why the public interest needs protection. It’s easy to assume that an App in common usage is comprehensive, unbiased, and trustworthy. But is it?

Sitting in a theatre there’s music playing as people fill the auditorium. Sounds a bit fancy. I’m talking about a local arts centre on a Saturday night. Playing was a compilation of songs and comic tunes few of which I knew. This kept the audience entertained as they shuffled in.

It’s the digital age. One tap on an App on my phone and the result is usually a whirring symbol whilst the App does its analysis to eventually show me what was playing. The whirring symbol kept going. This is hard, so the App excused itself. Then it drew a blank.

The next approach was to pick what seemed to be a theme in the song and look it up on a search engine. There’s the answer. A children’s show theme.

That made me think. Did the App not know the answer or did the App not bother to have reference to that type of song? I went down this line of thinking because what the App normally does is “helpfully” show where a song it has found can be purchased (streamed).

What interest could the App’s algorithm have if there’s nothing available to buy at the end of the process? Any presumption of coverage, fairness and objectivity is merely something I projected on the App’s internal workings. It’s not necessarily there.

The wider point is that placing trust on Apps that are free to download and for the most part produce useful results, as if by magic, maybe not be quite what they seem. Afterall nobody reads the small print – do they?

Some Apps compose data that is extracted from reputable sources. These become popular because of the love of convenience. Within bounds, the information provided can be trusted. This is the general assumption I may make about a freely available weather App.

A key factor here is trust. However, most often that trust only comes about by a regular “try it and see” experience. There’s no mark of quality. There’s no organisation looking after the public interest. Now, if you say to me – that’s what Ofcom[1] does in the UK, I’ll have a little laugh. That regulatory may keep a watching eye on the worst offenders as far as lawbreaking is concerned, but they are not much help in saying this App is a gold star, and this one is equivalent to a dodgy doorstep salesman.

For Apps with mechanical algorithms that chunter out information this may not be a problem. Even if their display includes advertisements and enticements to win a holiday.

What comes next with AI is more concerning. The commercial interests of a smart music identifier are evident. Making recommendations is not to my detriment. Although purposefully fishing in a restricted pool of information could be.

With general purpose Apps, which are a front to AI systems, there’s the potential for manipulation. Where commercial interests are the primary motivator there does need to be someone looking after the public interest. My interests. Yes, I am in favour of regulation.

[1] https://www.ofcom.org.uk/

Unraveling Future Risks

Much talk of the end of times doesn’t make for the end of times[1]. It’s true to say that commentators have taken time on their keyboards to ponder on the apocalypse. It isn’t the first time the subject has occupied the public mind. I could say that the four horsemen of the apocalypse have been around since biblical times. They have, and now there’s a couple more of them riding around the globe.

In my lifetime, the atomic age, the jet age, being wiped out by bombs and a following nuclear winter was brightly coloured there on the chart of probability. Far too high up that chart. Then the Cold War ended and a period of moderate sensibility seemed to exist. That didn’t last long. Chewing over the masses of accumulated planetary climate data created another headache. This time, but at a slower pace, climate change became one of the horsemen. No doubt the popularity of apocalyptic science fiction and science reality comes from a deep insecurity. A realisation that the evolution of space and time is way beyond human control.

Playing my vinyl that’s been sitting in a box for too many years, I’m listening to one of my favourite David Bowie tracks. “Five Years[2]” is the ultimate apocalyptic disaster song. It asks the listener to imagine the News coming over that we’ve only 5-years left on Earth. What would happen in such a situation? Would we all scream and run around like chickens with no heads?

Now I must ask, is that a good lead into one or two words about Artificial Intelligence (AI)? Those working on this technology seem to be getting nervous about that technology which they have cherished, developed and promoted. Let’s first say that AI is not new. When I graduated, in the 1980s, it was one of those esoteric subjects that was only for those wanting to do a PhD in an obscure part of the world. Not more than 20-years ago, given that aviation research was part of my remit, academics lobbied me because they thought a complex neural network should be able to predict the next major accident in aviation. That is fed with enough safety data. They may have been ahead of their time.

What has surprised a lot of people is the speed with which industry has gone from academic research to live digital systems that exhibit astonishing abilities. Access wide open too.

I don’t know if there’s a Moore’s Law[3] for AI. There doesn’t need to be one. The observation is that every week there’s News of even more powerful technology being made available. The disruptors are disrupting themselves before the rest of us have a clue what’s happening. As David Bowie would sing: “my brain hurts a lot.” Here’s what I think is problematic.

One. Concentrated Ownership. This innovation, and its potential power are in the hands of few people and States. Specialised expertise and infrastructure are prized intellectual property.

Two. Forecasting Dilemma. Projections are fuzzy, to say the least. Who can be believed? A simple curve on a graph doesn’t cut it. Step transitions are more than likely.

Three. Sleeping Sentinels. Regulatory lag is normal. Rules and regulators take time to mature. It’s just that, in this case, this lag may be disastrous. And some prefer to sleep.

Four. Legacy Thinking. Imagining AI’s future to be a version of past progress is woefully inadequate. It’s a new game when machines exceed human intelligence.

I do not believe that the world will end in five years but it maybe unrecognisable from today’s world. AI will create as many problems as it solves. Surely the powers that be need to step back and think about how the next decade will pan out.

[1] https://www.theguardian.com/us-news/2026/sep/02/jd-vance-god-podcast-end-times

[2] https://youtu.be/8gPSGrpIlkc

[3] https://online.hbs.edu/blog/post/moores-law

Money vs. Integrity in Elections

Money and politics. There’s a combination that goes together like clouds and rain. It’s difficult to conceive of one without the other. Money doesn’t guarantee success, but it does tip the balance in the favour of those who have it. If a political party is out of sync with the voters, then piling on the pounds will barely tip the scales. Same is true if a candidate’s credibility is in the pan. What money does in the run up to elections is summed up in three points.

One. It is used to influence the climate of opinion. Journalists are a fickle bunch. Raining issues down on them and pushing hard gets lots of coverage. Having a team of fresh thinkers drafting speeches and doing research (hard graft) means arguments can be placed for maximum impact. Volunteers can be superb. Trouble is there’s never enough of them to go around.

Two. Professionalising campaigns. British politics, at a local level, is renowned for its amateurism. In so many ways that’s a wonderful situation. Petty reveries, few ancient but dedicated leafleteers, committee rooms with dry rot, out of date records of voting intentions, a dream, never fulfilled, that a bigwig’s visit is just around the corner, all that and more. Money does afford the recruitment of motivated campaign managers and organisers who know how to get a message out and follow-up on voters’ concerns.

Three. Pure swagger. Saying that a new candidate will never win here, is not from around here or is an accidental choice of a struggling local committee, they can all be pushed aside. Money allows even the most desperate, disorganised, dishevelled individual to be bigged-up. Confident candidates, with adoring supporters run the full distance, whatever the inevitable knocks. Money is great for maintaining momentum and papering over the cracks.

It’s interesting to note that none of the above means that a constituency gets a better Members of Parliament at the end of the process. However, if all the candidates on the ballot have access to some of the above then local voters do get a wider choice.

On the basis that no political party will ever win all the seats in the House of Commons, targeting has become the number one strategy. Only expend resources where they may best produce positive results. That can be 300-400 parliamentary constituencies, all with their idiosyncratic issues and 70,000 voters poised to put a cross in a box. A third of them will not bother with the entire process.

So, what’s the problem with money? Well, that depends what you hold in your heart about liberal democracy. Afterall the ideals of democracy are played out in a fraction of the 200 or so, sovereign countries across the globe.

My view is that all citizens should be eligible to put themselves forward for election to a political post. The pool should be as wide as possible. Naturally, there are a few exceptions, jail being one of the more notable. In standing for election there should be a reasonable level playing field. Ideally, it’s effort, ideas and talent that should win the day. Not piles of cash or undue influence.

For the voters, it’s come as you are. I lament when hearing of candidates trying to change minds on the doorstep with nothing but a metaphorical pea shooter and tendency to take the moral high ground. There are difficult issues to address. Where vast amounts of money can be destructive is where they used to shift the climate of opinion onto subject that are far away from the ones that really count. The major topics of education, health, employment, tax, and spending should never be marginalised. I’d even add transport to that list.

Vast sums thrown about by wealthy people distort that level playing field. That’s not right.

Digital Age Dilemma

There was a time when most of the political parties in the UK would object to mass surveillance and the State having a watchful eye on every one of its citizens. The influence of the Magna Carta, that 800-year-old document, was a keystone of thinking in this country. Although it was aimed at the privileged when first placed in front of a Monarch, it embraced the idea of liberty. It so upset the establishment that a Pope denounced it.

Ideas endure and those contained in the Magna Carta went on to inspire the framers of the constitutions and legislation in many countries. It doesn’t sound so strange in modern times that a person should be able to go about their lives without an overbearing State impinging on their liberty. Well, at least in theory.

I remember the intense debates that took place in the mid-1990s. The subject being the introduction of identity cards in the UK. Technical arguments were made of how such a measure would simplify lives in the coming digital age. Those agreements paid little attention to the erosion of fundamental rights that the centralising personal data on citizens would bring.

What’s happened now, through a process of stealth, is that people have given away vast qualities of data to international commercial entities without batting an eyelid. The legal device that has been used is a contract that few people ever read. The End User License Agreement (EULA). This detailed fine print addresses data collection and aims to protect software providers from liability.

I suspect that Conservative politicians have observed how little attention people are paying to the erosion of their rights that they feel emboldened to abandon any commitments to liberty. For some time now the Conservative mantra of has been to leave the European Convention on Human Rights (ECHR). This is based on a gamble that most people will not have read the document. It’s a bit like the snatch and grab that the typical EULA has become.

Now, the Conservative Party has doubled down on a shift to the right of politics by inviting the prospect of mass surveillance. There’s a reasonable argument for the police to excurse powers to delve into the digital world when investigating crime. In fact, how can they not do so. However, this should be done so when there’s reasonable basis for doing so. Not everyone is a suspect of criminal activity. The law of the land still upholds the idea that guilt must be proven.

Computational tools are getting ever more powerful. I could joke that the number of articles on AI would reach to the Moon and back, but I wouldn’t be far wrong. One of the powerful software tools that’s accumulating uses is that of digital facial recognition. We’ve advanced from the days of inky fingerprints, although such prints are still valuable identifiers.

Where do we have a problem with liberty? The framers of many longstanding constitutions didn’t have to think to hard about what liberty meant in a world devoid of digital cameras. Now, any citizen’s image is going to be caught hundreds of times through the day. Maybe thousands of times. That instant image can be present on storage devices across the globe.

Who owns that image of me or you? Can it be used by anyone? Tricky questions given phones, traffic cameras, doorbells and home and business security cameras are subject to EULAs. On a superficial basis, requiring police forces to extract digital images and run them through a national database can sound like a good crime fighting idea.

The problem is where is the line drawn? Mass surveillance? Where everyone is an immediate suspect until proven not guilty? Not guilty by virtue of the calculation of an algorithm. Widespread digital camera coverage does put this in prospect. Images stored all over the world. Do we need to reframe the Magna Carta for the AI age? Now, there’s a thought.

Future of Aviation: Enhancing Safety and Resilience

The aim is clear. A safe, secure, sustainable and resilient international aviation system. I’d add to that list an ambition to continuously improve.

And that’s in the face of the challenge that was recognised 30-years ago. With an aviation system that knows how to archives a high level of safety, even incremental improvement is challenging. Factor in projected air traffic growth that’s unabating. What we know is that that continuously improving the global accident rate requires active interventions.

Since here I can only highlight the key priorities, I’ll choose three aspects of active intervention. Back to what I said earlier in my talk. Aviation is still climbing a maturity ladder. Digitisation has facilitated a growth in proactive work. Initiatives and plans identify actions aimed at solving known aviation safety problems. But the next step remains some way off. It’s that ability to anticipate safety problems before they occur. To predict.

We have the notion of safety intelligence. There are Big Data projects that are gathering large quantities of aviation safety data. Today, what might be called prognostics continues to depend on expert judgement. That’s not negative, as such. In fact, that augmented by objective and trustworthy safety intelligence is a route to the next step in maturity.

In my view, here’s the three categories that must be addressed now and, in the next decade.

One is a foundation stone of aviation safety work. It’s the capability and will to react in a timely manner to events. Not just accidents and serious incidents but any event that elevates aviation safety risks above acceptable levels.

Top operational risk that are likely to need immediate actions are those involving loss of control and those concerning runway incidents.

To take timely corrective action the results of investigation need to be readily available. As pointed out by the industry, across the globe, there’s much scope for improvement here. Timeliness matters.

My next category is that of systemic risks. System wide safety risks. People, organisations, processes, procedures, the human factor and human performance issues.

Much is being done to ensure the effective implementation of safety management systems (SMS). The four pillars of SMS have proven to work – safety policy, risk management, safety assurance and safety promotion.

The acceptance, elimination or mitigation of risks is not an act that begins on one day and stops on the next. Digitation can help but there’s always the dangers of the needle in a haystack or even problems sitting in plain sight.

What’s often weak is the communication of what’s discovered. Believable, understandable straightforward communication to decision makers is vital.

Thirdly, what’s ahead of us is an order of magnitude more complex than what’s gone before. And a rate of change that mindboggling.

I’m calling this emerging aviation safety risks. The question arises; are we ready for advanced leaps in technology? Hydrogen, hybrid propulsion, machine learning, quantum computing, complex airspace networks and robotics. Humans as executive manager whilst retaining control only in emergencies.

It’s a famous quote; data is the new oil. We are ushering in extremely large and complex datasets that will be essential to the workings of automated and autonomous systems. Could data be the new quicksand? This is a huge issue in dense airspace where crewed and uncrewed aircraft must share the airspace.

Another emerging aviation safety risks, that industry and regulators are starting to address is the fact that safety and security have become inseparable. In the past, these disciplines were addressed as silos. That’s no longer viable.

Another emerging aviation safety risk is that of the workforce. Aviation has attracted dedicated professionals who acquire experience and train to a high level of competence. A safety culture has been embedded. In this respect the fundamentals of aviation safety remain constant.

The rapid growth of technology is a two-edged sword. Now, there are many opportunities for future generations, but aviation may not be at the top of the list of choices, as it has been. There’s a serious aviation safety risk if serious investments are not made, people are not motivated and able to gain the competencies needed by a rapidly changing industry.

To sum up. Everything I’ve spoken of can only be addressed in partnerships. In the past, the aviation industry and its regulators have proven themselves to be imaginative, resilient and forward looking.

The challenges ahead, dare I say, are even greater. Levels of integration, interdependency and rapid technology adoption are off the charts. Yet, I believe we can continually improve global aviation safety. We can step up.

POST: Farnborough International Air Show 2026. 20-24 July 2026.

Current State of Aviation Safety: Data-Driven

The present day. Today. There will be Press Releases galore at this air show. I expect ambitious and upbeat predictions. More orders. More growth. More developing markets. Demands for more skilled people, more performance, more pressure for the early adoption of rapidly changing technologies.

My subject is safety. Where are we with aviation safety in the present day? There’s a sound legacy. Afterall aviation has a long tail. What we have now, and if I look at the Farnborough Airshow catalogue of ten years ago[1], is digitisation. The ambition to be data-driven. A continuing drive to exploit data of all kinds. Extracting information from ever increasing sources in aviation.

This desire for actionable safety intelligence is not new. On my desk I have a coaster celebrating the UK CAA’s MOR Scheme[2]. That’s been alive for 50 years (1976-2026). Occurrence data has been, and is being, used to construct all sorts of safety performance indicators. These indicators can be useful but it’s not always so. They are fuel for safety management systems.

I used the word “actionable” with purpose. Today, making the results of safety data analysis useable by front-line actors and responsible managers has been a challenge we wrestle. I’ve often heard it said – that’s great but what do I do about it? Too many “greens” on a chart and complacency can set in. Too many “reds” and everyone goes into panic mode.

I have been fortunate to work with a number of safety initiative. One called the Commercial Aviation Safety Team (CAST) in the US. To match its work, on this continent, I set up a European version; ECAST in 2006. The focus was on data-drive analysis. There’re safety teams for Rotorcraft and General Aviation too.

These safety initiatives came about a reactive way. Reporting in 1997, the Gore Commission, was a reaction to dilemmas that is still with us. What to do in the wake of catastrophic accidents. And more pertinent to this talk; what to do when the global accident rate curve looks flat at the same time as predictions are for a significant growth in global air traffic. Reading this report now, the predictions were horrifying. Predictions of one large aeroplane fatal accident every week by now. Luckily, commercial air transport didn’t suffer that fate.

I should NOT say “luckily”. Current safety performance is down to an enormous amount of work done by thousands of aviation professionals over a sustained period. Not only this but shocks, whether they be financial crisis, volcanic eruptions or global pandemics, have tested this industry severely.

Taking on board what was learned over the decades, safety management has developed and matured. Incorporated in aviation’s safety regulatory framework, it’s now practiced across the whole industry. Almost. We don’t just react. We now have Safety Plans. Lists of actionable tasks aimed at continuous safety improvement. On the scale mentioned earlier – we are proactive.

Let me stop. This is not a utopia. I’d say at least half the global aviation industry is struggling with the need to be proactive. So, the road we are travelling on remains a long one. We don’t just need the capability to do better; we need to maintain the will to do better.

[1] 50^th Farnborough Air Show in 2016.

[2] https://www.caa.co.uk/about-us/make-a-report-or-complaint/report-something/mor/occurrence-reporting/

The Evolution of Aviation Safety: A Historical Overview

A quick tour of the history of aviation safety. In a couple of minutes.

As powered aircraft took to the skies it quickly became apparent that, of all the means of transport, aviation was less forgiving than others. Moving at speed in four dimensions, with the necessity to take-off and land safely, has inherent risks. By no means does this mean that flying is an dangerous activity. What it does necessitate is an exercise of sound engineering, preparation, and proficiency. When these are missing, and in cases of misfortune, accidents and serious incidents happen.

In the days between the world wars, aviation moved from the military and a circus-like amusement to a viable means of public transport. Progressively, more passengers had the opportunity to experience the wonders of flight.

Today, it’s not the early days of flight I want to focus on, but we do owe the engineers and aviators of that time a great debt. Much was learned as aeronautics matured. Since the end of the second world war the production, promotion and application of standards has embedded what had been learned. This has established the way that international civil aviation works.

One simple expression of the stages we’ve been through is Prof Patrick Hudson’s[1] model. This model describes five distinct levels of cultural maturity. The steps are pathological, reactive, calculative, proactive and generative.

This model sits alongside steps in the development of technology that have been a hallmark of aviation. Different categorisations exist. AIRBUS[2] has one that’s easily understood in terms of generations of civil aircraft.

First are the “classics” of the 1950s and 60s. Next, the second generation starts down the road of more electronics. More safety critical systems. Third generation starts flying in the early 1980s when automation becomes the norm (early fly-by-wire, glass cockpits, flight management systems). The fourth generation introduces safety systems, which address the major causes of fatal accidents, namely Controlled Flight Into Terrain (CFIT) and Loss Of Control In-Flight (LOC-I). That’s the early 1990s. When I made that move from industry (design and production) to an aviation safety regulator.

Now, I’d say we are well into fifth generation aircraft, with composite structures and integrated modular avionics. In fact, we are well into speculation about what the sixth generation may look like. Machine learning, single pilot operations, hybrid powerplants, aircraft as a node on a network.

Because aviation is a learning industry, and a conservative one, the expectation is of progressive improvement in maturity throughout the aviation system and innovative technology applied to enhance safety.

This is not to say that the potential for error, missteps and recklessness don’t exist. They do. I expect continuity in a way that takes full advantage of advanced methods without scarifying the legacy of an astonishing good global aviation safety achievement.

Later, to cover a lot of ground in a short time, I’m going to condense all the above into three distinct categories. But before that, I’ll talk about the present and a vision of the future.

POST: UK Farnborough Air Show on Thursday 23rd July 2026. Time: 13:00-13:15 BST. Location: Hall 3 Stage. Session title Where Next for Aviation Safety?

[1] https://www.sciencedirect.com/science/article/abs/pii/S0925753507000227?via%3Dihub

[2] https://d10x.airbus.com/generation-of-jets/