This piece in the aviation trade paper Flight International talks about a United Airlines Boeing 777 that flew for 193 minutes on a single engine when on a non-stop flight from Auckland to Los Angeles. This is the furthest that an airliner has ever flown on a single engine after failure of the second engine.

Of course, all airliners are designed to do this if necessary, but one thing that is really interesting are the regulatory aspects, and how they relate to the various battles between Boeing and Airbus for supremacy in the airliner market.

Until 1980, virtually all long haul flights were flown with four engined aircraft (such as the DC-8, 707, and 747) or three engined aircraft (such as the L-1011, and DC-10) with twin engined aircraft being used only for relatively short haul flights, usually over land, where there were plenty of airports that the aircraft could fly to in an emergency. Three or four engined aircraft could at least stay in the sky on one engine if two engines failed. However, in 1979 the Boeing 767 went into service. Long range versions of the 767 had the range to cross the Atlantic, and various airlines wanted to use them for this purpose. There were fears about what might happen if both engines failed in separate incidents.

(There are of course circumstances in which more than one engine can fail in a single incident, in which case having extra engines may not help. There have been incidents of airliners landing successfully with no engines, the most famous being the "Gimli Glider" incident involving an Air Canada Boeing 767 in 1983. The plane's fuel gauge was out of order and the mechanics refueling the plane had to measure the fuel manually. Unaware that all the documentation for the relatively new type of aircraft gave fuel capacity in kilograms of kerosene rather than the more traditional pounds of kerosene, the plane was refueled with an inadequate fuel load, and the aircraft ran out of fuel in flight. The pilots were able to glide the plane to a nearby disused airforce base, and land it without any loss of life and without too much damage. (There was an auto racing club having a meeting on the runway at the time, which was both good and bad. Bad because they had to run out of the way, but good because they had a lot of emergency equipment on hand and were able to rush to the plane with fire extinguishers immediately after the landing). There has also been at least one incident of a Boeing 747 losing power from all engines after flying into the ash of a volcanic eruption, but in this case the engines were able to be restarted after flying out of the ash. But I digress).

The American FAA resolved this problem by creating something called ETOPS approval. (Extended Twin OPerationS). The performance of a particular type of aircraft with a particular airline was monitored by the FAA. If the airline and aircraft flew a certain number of trouble free hours flying with the particular aircraft type, then they could initially be certified for 60 minutes ETOPS. This meant that The plane was allowed to fly on routes from which it was possible to reach an emergency alternate airport within 60 minutes flying with one engine. Over the years, continued good safety, aircraft and airlines could be certified for 120 minute ETOPS and eventually 180 minute ETOPS.

European regulators have always been reluctant to go along with these American initiatives. It may be that they were and are genuinely concerned with safety, or it may be for very cynical reasons (In 1980, Europe's Airbus consortium had no twin engined aircraft with anything like the range of the 767). In any event, the number of transatlantic flights flown by twin engined aircraft has steadily increased over the years, to the point where it is not terribly common to cross the Atlantic on a four engined aircraft when flying on a European airline and it is virtually unheard of when flying on a US airline. Transatlantic routes are dominated by 767s and 777s.

In the 1980s, both Airbus and Boeing developed new types of aircraft to replace DC-10s, L-1011s, and early generation 747s on long haul routes. Airbus developed two aircraft with a lot of commonality, the twin engined A330 for relatively short haul routes and the four engined A340 for longer haul routes. On the other hand, Boeing developed a single twin engined type, the 777, for routes of all length. (Boeing developed its aircraft later, when engines were more powerful. Airbus probably couldn't have built a long range two engined aircraft at the time it designed the A330/A340). The longer range versions of these aircraft were designed with trans-Pacific routes in mind, and Boeing wanted the 777 approved for 120 minute and 180 minute ETOPS with a much streamlined approval process. Airbus, which did not face this problem with the four engined A340, was responsible for a little bit of a scare campaign about the safety of 180 minute ETOPS, and there was a little more ill will between regulators over this. (Part of the issue is that the two engined 777 manages better fuel consumption then the four engined A340, so Airbus was countering a commercial disadvantage somewhat). But, again, Boeing got its way, and the 777 went into production with the 180 minute ETOPS. Airbus continued to make noises, but the 777 sold very well, and the A340 less well. (Airbus has done much better with the medium range A330-200, however).

And one pretty much has to concede that Boeing and the US regulators got this one right. There has never been an incident in which one engine has failed, and an aircraft has then gone on to lose the other engine separately, despite the fact that there are now hundreds of ETOPS flights flown per day. ETOPS are now used for trans-Pacific flights, as is demonstrated by the article I linked to.
And it seems that airlines push things to the limit, given that an aircraft with 180 minute ETOPS approval in this case had to fly for 193 minutes after losing one engine. (This does not necessarily imply anything was done wrong, as weather conditions could have changed or something like that, although I am sure they are looking into it).

As the article says in the end, the US FAA is now trying to change the basis of regulation, so that certification of aircraft has nothing to do with the number of engines. To me, this seems reasonable. The risk due to flying long distances over water with only two engines has been shown by experience to be miniscule. Although it may be that there will one day be an incident caused by the loss of two engines separately, this risk is small compared to other risks that are taken in everyday operations.

However, we still have the differences between Europe and America. European regulators are again reluctant to go along with the American desire to change the rules.

Interestingly, Boeing and Airbus have different ideas concerning the future of long distance air travel (which means flights across the Pacific, or flights between Europe and Asia, mainly). Boeing believes that travel on very large aircraft between hub cities like Los Angeles, Tokyo, and Hong Kong, is going to become less important and point to point services (perhaps something like Manila-Las Vegas) on smaller aircraft will become more important. This is very similar to what happened across the Atlantic after the invention of the 767. Boeing therefore is selling its 777 and a new longer range, twin engined 767 sized aircraft that is in development to serve this market. Airbus on the other hand believes that the hub airports will become more crowded and therefore that bigger aircraft will be needed, and is therefore selling its new, giant, four engined A380 as the solution to this problem.

It may or may not be that regulation here is being influenced by commercial concerns. I suspect that regulators in both sides of the Atlantic believe that their position is driven by safety concerns, and that in both cases safety actually is the paramount concern. But still, there is a pattern.