One chunk of a recalling of the path civil aviation has taken in the last 40-years is called: Safety Management Systems (SMS). It’s a method or set of methods that didn’t arrive fully formed. It can easy be assumed that a guru with a long white beard stormed out of his quiet hermitage to declare a eureka moment. No such thing happened.
Through every part of my engineering design career the importance of reliability and quality systems was evident. Codified, procedural and often tedious. Some say the quality movement had its origins in the world of the 1960s moonshot and the advent of nuclear weapons. I don’t think there’s a single spring from which the thinking flows.
That said, there are notable minds that shaped the development of standardised quality systems. Acknowledging that the Deming Cycle[1] is core component doesn’t take too much of a leap. It’s a simple idea for capturing the idea of continuous improvement. Aerospace design and production organisations adopted this method readily.
Those first steps were all about the Q word, Quality. How to deliver a product that reliably worked to specification. At the time the S word, Safety wasn’t spoken of in the same way. There had been an underlying presumption that quality success led to safety success. However, this was not entirely true. An aerospace product can leave a factory 100% compliant with a pile of requirements, specifications and tests only to subsequently reveal failing and weaknesses in operational service.
In the saddest of cases those failing and weaknesses were discovered because of formal accident or incident investigation. In civil aviation these are conducted independently. Worldwide accident investigators and aircraft operators often detected a lack of learning from past events. This situation stimulated activities aimed at accident prevention.
In 1984, the International Civil Aviation Organization (ICAO) published the first edition of its Accident Prevention Manual. This document introduced concepts and methods aimed at accident prevention. It was a pick and mix of initiatives and processes gleamed from the best-known practices of the time.
One of the jobs I had on joining the UK Civil Aviation Authority (CAA) Safety Regulation Group (SRG) was to work with the ICAO secretariate on an update to the Accident Prevention Manual (Doc 9422). The UK CAA has long been an advocate and early adopter of occurrence reporting and flight monitoring. Both were seen as key means to prevent aviation accidents.
It was envisaged that a second edition of the manual would be available in 2001. That didn’t happen. Instead, ICAO decided to harmonise information available on safety and put that into one manual. At that point safety information was scattered around the various ICAO Annexes. Thus, the content of the Accident Prevention Manual was consolidated into the Safety Management Manual (SMM) (Doc 9859). This new document was first published in 2006.
There’s much more to say since the above is merely a quick snapshot.
[1] https://deming.org/explore/pdsa/
John Vincent 
It’s good to remind ourselves, as you are doing here, how we got to where we are now, because safety is hugely better now than it was in 1984, and has improved measurably further since this century dawned. It’s important to identify what we are doing right, so we can keep on doing it, and adding to it.
As you point out here, “there’s much more to say”. As a journalist reporting on operational safety since 1980, I have watched many factors that influenced safety improvement, but perhaps the single most influential factor was the arrival of personal computers which enabled accident/incident data collation and analysis, because that in turn gave us the ability to develop “data-driven” policy.
The problem with the old days when all the paper reports were stored in filing cabinets was that the data was all there, but there was masses of it, and it was not instantly retrievable nor visible. To use the old cliche, we “couldn’t see the forest for the trees”.
As soon as we could see clearly what had been going on, the FAA set up its Commercial Aviation Safety Team (CAST), Europe set up equivalents, and a set of priorities for safety action were developed because the greatest risks were suddenly clear. I remember, on a more local level, how British Airways developed its own BASIS, (BA Safety Information System), which ensured all its incident reports went into PCs and came out as usable data.
There’s still much more to say! And it’s a continuing task to keep on the safety rails. Boeing with its 737 Max crashes in 2018 and 2019 proves that even the best can go off the rails if they don’t keep their eyes on the ball.