From the 1960s onwards, the chemical and oil industries developed and used a number of new safety techniques which, in time, became second nature to those who applied them. They included the use of QRA for deciding priorities, Hazop and audits for identifying problems, inherently safer design for avoiding hazards, and more thorough investigation of incidents for identifying underlying causes. However, it has not yet become second nature to remember the accidents of the past and the actions needed to prevent them happening again.

I joined industry in 1944 and moved to production in 1952. Then, and for at least 15 years afterwards, safety was a non-technical subject that could be left to arts graduates and elderly foremen. There was concern that people should not be hurt—great attention was paid to the lost-time accident rate—but there was no realization, that it was a subject worthy of systematic study by experienced technologists.

This view changed at the end of the 1960s. A new generation of plants had been built, operating at higher temperatures and pressures and containing larger inventories of hazardous chemicals; the result was a series of fires and explosions and a worsening fatal accident rate. Figure 1 shows the situation in ICI, at the time the UK, s largest chemical company. Other companies experienced a similar state of affairs.

As a result in 1968, I was appointed one of the company's first technical safety advisers, an unusual appointment at the time for someone with my experience, and if the reason for my appointment had not been so obvious I would have wondered what I had done wrong. I and my colleagues tried to apply the same sort of systematic thinking to safety that we applied in our other professional work. We developed some new concepts and techniques and adopted others. A common feature of our ides, realized only in restrospect, was that they consisted of more than mere problem-solving techniques. Once people had got used to these new concepts and used them a few times, they began to look at a whole range of problems in a different way.