Sample Fleet Safety Policy: Fill & Download for Free

That's aircraft type specific.On an Airbus A330 and A340 the CVR or cockpit voice recorder saves at least the last two hours of sounds and conversations from the cockpit.The DFDR or Digital Flight Data Recorder records more than 1,200 parameters on the latest 330s with varying sampling rates, depending on the parameter; some are sampled once per second, some once every tenth of a second for example. The DFDR keeps at least 25 hours of data before it’s overwritten.CVR and DFDR are located in the back of the aircraft in one crash proof box: the black box; which is actually a bright orange box so it will be easily identifiable in case of a crash.On an Airbus A350 over 2,500 parameters are recorded on the DFDR. I assume an A380 will be similar to that. An older Boeing 747–400 records about 500 parameters.There’s another recorder used for monitoring flights. On the Airbus it’s called the QAR or Quick Access Recorder. This one can piggyback information from the DFDR or directly from avionics and can record even more if the airline desires. QARs are easily accessible in the avionics bay, under the cockpit and on the A350 for example can send data wireless from aircraft to the airline when parked at a gate, eliminating the need for engineering to regularly replace the discs.QARs are not crashproof. Their function is to allow the airline to improve safety by monitoring flight data on a regular basis without the need for laborious black box removals or without relying on crew’s self reporting.Every cycle about 250 GB of information goes to the safety department in the airline I fly for. (I forgot if that was per airplane or per fleet of roughly 100 airplanes; the number just stuck from one of the briefings I had about it. Also I believe “cycle” implied per month here, but I could stand corrected.)That information is run through computers, and excesses and events show up when they exceed certain limitations set by airline policy. If some airports stand out by a particularly high event rate of something, say high rate of descend in final approach, that airport could be included in the next simulator training package, for example. If some events start creeping in more than the airline likes, then similarly it could be a topic for the next round of training. In other words: the statistics from the QAR allow the airline to identify trends and proactively counter issues or steer towards safer operations.Of course if pilots decide to go rogue on the airline's policy, they will be caught too and can expect to be invited for coffee and cake. I heard there rarely is any coffee nor cake offered by the fleet office then. Therefore pilots in almost all cases proactively self report instances they think may be borderline or over.

Within a word, but most Americans have never heard the story.The day after Donald Trump announced his intent to break the Intermediate-Range Nuclear Forces Treaty, signed by Reagan and Gorbachev in 1987, is the 56th anniversary of President John F. Kennedy’s television announcement to the world that US intelligence agencies had discovered Soviet intermediate range (and short range) nuclear ballistic missile sites under construction in Cuba. The news shocked the world when JFK said “the presence of these large, long-range, and clearly offensive weapons of sudden mass destruction -- constitutes an explicit threat to the peace and security of all the Americas…missiles in Cuba add to an already clear and present danger .” He said the US Navy would quarantine all ships carrying offensive weapons to Cuba.JFK assembled an executive committee (EXCOMM) to guide his actions, but they were not aware that the Soviet Union had installed 46,000 combat troops with 100 tactical nuclear weapons (in addition to the 60 nuclear missiles) in Cuba and sent four submarines armed with nuclear torpedoes to establish a base in Mariel Bay. Overnight, the table was set to escalate the world into nuclear WWIII. This part of the story is widely known, but the rest of the story, based on research for my book, will unfold on Quora over the next five days.After JFK’s speech, Defense Condition (DEFCON) 3 went into effect, but the U.S. had no plan in place because U.S. intelligence believed that the Soviets would never install nuclear missiles in Cuba. EXCOMM considered six courses of action,1. No action.2. Diplomacy to pressure the Soviet Union to remove the missiles.3. Warning to Castro of the grave danger to Cuba.4. Blockade: U.S. Navy quarantine of missiles destined for Cuba.5. Air strike: U.S. Air Force attack against missile sites.6. Invasion: Full invasion of Cuba including overthrow of Castro.The Joint Chiefs of Staff (JCS) unanimously supported a full-scale invasion of Cuba because they erroneously believed that the Soviet Union would not battle the U.S. over Cuba. They were unaware of the 46,000 combat troops with 100 tactical nuclear weapons had been deployed to Cuba for that purpose. Kennedy was skeptical, reasoning that Khrushchev would respond, if not in Cuba, then in Berlin. Kennedy signed the naval quarantine proclamation.The US Navy received CIA intelligence reports that Soviet submarines, possibly carrying nuclear missiles for Cuba, would arrive within a week. The intelligence was faulty. Four Soviet submarines had already arrived in the Sargasso Sea carrying nuclear torpedoes, not missiles, and Moscow assigned them to dispersed combat sectors near the quarantine line. Robert Kennedy later reported that "the President ordered the Navy to give highest priority to tracking the submarines and to put into effect the greatest possible safety measures to protect our own aircraft carriers and other vessels.” The US Navy deployed the largest antisubmarine warfare (ASW) fleet ever assembled in one ocean to the Sargasso Sea.The US Government promulgated a document entitled “Submarine Surfacing and Identification Procedures” to the US Navy, the Soviet Government, and the rest of the world as an open Notice to Mariners. The protocol called for the dropping of five grenade-sized practice depth charges (PDCs) as a signal for a submarine to surface. The Soviet submarine captains never received the notice.Life on the Soviet subs became a living hell when they reached the subtropical Caribbean climate. Three of the four subs had sustained damage from their transit through Hurricane Ella, and the heavily insulated boats designed for Arctic operations became steaming hell holes with temperatures over 120 degrees in the torpedo room and 140 degrees in the engine room. The fresh water distillation plants became so inefficient that they could produce only one cup of water per man per day. As a result, the crews suffered from heat exhaustion, dehydration, and horrible skin rashes with no ability to clean themselves during periods of profuse sweating.By dawn on October 24, 1962, the world held its breath as the last twenty-two of the one hundred Soviet merchantmen disguised as humanitarian transports, but stuffed with hidden troops, missiles, warheads, and other offensive weaponry, steamed toward Cuba. The US fleet was in position along the Walnut Line, a 500-mile arc around Cuba, and they searched for Soviet submarines from the Bermuda Triangle to Newfoundland, Canada. US Secretary of Defense McNamara intended to "put pressure on the submarine, move it out of that area by pressure, by the pressure of potential destruction," but he acknowledge the “very dangerous situation.” Kennedy responded, "We don't want to have the first thing we attack as a Russian submarine." Soviet Premier Khrushchev cautioned that Soviet submarines will attack any American ship that stops a Soviet ship.While the world focused on surface ship confrontations, the US Navy stepped up its submarine search efforts, and three hours after the quarantine went into effect, the first positive submarine siting was reported as contact C-18, the Soviet Foxtrot class B-130 struggling to remain operational on three damaged diesel engines. The crippled sub escaped. Additional submarine sightings continued throughout the remainder of the day, leaving the US Navy uncertain as to how many Soviet submarines had entered the theater. Assistant Secretary of Defense Arthur Sylvester announced that military action would be directed at Soviet submarines that refused to surface, and asserted that the US Navy would board submarines that were forced to surface. The Soviet submarine captains were under orders to remain covert and undetected.More submarine sightings occurred for the next two days, and the US Navy attempted to maintain contact with submarines by air-dropping explosive echo-ranging devices called Julies and sonobuoy listening devices called Jezebels. When US destroyers approached submerged submarine positions, they dropped practice depth charges in groups of five as prescribed by the US submarine signaling protocol, unaware that the sonic efficiency of sea water and steel hulls rendered the underwater explosions indistinguishable from a full-blown attack on the subs.October 27, 1962 — later designated as Black Saturday by the Kennedy White House — shaped up to be a dangerous and horrifying day in terms of a nuclear apocalypse. The tension of impending disaster encircled the globe and tightened as the doomsday moment crept forward. The highest peacetime defense readiness conditions prevailed on both sides, one step shy of war. The slightest provocation could tip the fragile balance beyond the point of no return. The US Strategic Air Command targeted fifty-five hundred bundles of nuclear wrath at the Red Menace, with nearly three thousand nuclear warheads aimed at targets in the Soviet Union, awaiting the launch command. The Pentagon’s Single Integrated Operational Plan (SIOP), the nuclear order of battle, called for launching the complete nuclear weapon inventory against a thousand sites.The US tested two nuclear weapons that day: a ground explosion at the Nevada Test Site, viewed by thousands of dancing tourists quaffing “atomic cocktails” at blast parties along the Las Vegas Strip, and a lesser-known atmospheric detonation over the remote Johnston Atoll, west of Hawaii in the Pacific. The Soviets air dropped 262 kilotons on Novaya Zemlya.The USS Randolph’s sonobuoys bobbed in the stormy Sargasso, prosecuting C-19 (Soviet submarine B-59) while a US U-2 spy plane monitored the Soviet nuclear tests at Novaya Zemlya with high-altitude air samples and high-resolution photography above the North Pole. Flying blind into northern lights at twenty-one thousand meters, the pilot’s magnetic compass spun in frantic circles, contorting itself as it attempted to point straight down at the North Pole. The U-2 penetrated three hundred miles into Soviet airspace by mistake and scrambled six Soviet MiG fighter planes intending to shoot it down. The pilot had spent time in North Korean prison camps and had no interest in being shot down over enemy territory. He escaped death and avoided capture by racing toward an alternate landing strip at a remote Alaskan radar station. As he crossed back into American territory, the MiGs peeled off their pursuit, his fuel tanks emptied, and his engines flamed out. He averted personal disaster and military escalation by gliding his U-2 to an emergency landing.At 06:00, the CIA reported that five Soviet R-12 nuclear ballistic missile sites in Cuba appeared to be fully operational. Around midday, another U-2 pilot, US Air Force Major Rudolf Anderson Jr., flew a recon spy plane at 70,500 feet over Cuba to photograph Soviet missiles near Guantánamo Bay. Two Russian surface-to-air missiles, fired from Cuba, exploded, and shrapnel punctured his cockpit canopy and pressure suit. His lungs ruptured under explosive decompression and vaporized into a pink cloud that shot through the windshield of his disintegrating Dragon Lady at half the speed of sound. The remains of his body, strapped to the remnants of his plane, crashed on Cuban soil. His mission was so secret, he did not exist except as a lost radar contact.After days of being dogged by Task Group Alpha, Captain Valentin Savitsky, commanding officer of B-59 became convinced that all-out war had broken out. He ordered his nuclear torpedo to be prepared for launch. (Captain Nicholai Shumkov on the B-139 also ordered his nuclear torpedo to be made ready and the torpedo tube to be flooded, but his “special” weapons officer passed out when he heard the command and was unable to carry it out.) Captain Savitsky lost his composure in the asphyxiating madness and screamed, “The war has already started up there, and we are down here doing somersaults. We’re going to blast them now. We’ll die, but we will sink them all. We won’t disgrace our Navy or shame the fleet.”One man, Vasili Arkhipov, who had quelled a near mutiny aboard the infamous K-19, stepped forward. (The Soviets’ first ballistic missile submarine experienced a Chernobyl-like reactor accident at sea in 1961.) Arkhipov challenged Savitsky’s orders and convinced Savitsky to release a single sonar ping to test the American resolve to kill their submarine. With explosions deafening their ears and extreme heat broiling their brains, they argued. While their ship’s crew was passing out from oxygen deprivation and heat exhaustion, they expended their remaining strength to make a decision. With the fate of their nation and all the world at stake, they decided to save the world, to prevent its immediate consensual devastation by Armageddon. They defied death and their superiors by not initiating the act that would destroy hundreds of millions of lives and threaten everyone who might live beyond that moment.Had Captain Second Rank Vasili Aleksandrovich Arkhipov not intervened, a nuclear blast would have erupted in the Sargasso Sea. Every nation would have perceived that they were under attack, the Soviet forces, the American forces, Havana, Moscow and Washington D. C. President Kennedy had announced to the world five days earlier in his speech on October 22, “It shall be the policy of this Nation to regard any nuclear missile launched from Cuba against any nation in the Western Hemisphere as an attack by the Soviet Union on the United States, requiring a full retaliatory response upon the Soviet Union.” Sixty Soviet ballistic missiles tipped with megaton nuclear warheads, a hundred tactical nukes and forty thousand Soviet troops had been placed in Cuba.The truest fact is that the greatest US hero of all time was a Russian naval officer that almost no one knows, who stood stone-steadfast at the centroid of international absurdity, beneath the sea against a myriad of apocalyptic weaponry that powerful nations arrayed against the world at the most treacherous moment of the human epoch.The Future of Life Institute honored Vasili Arkhipov with their inaugural Future of Life award in London on October 27, 2017, the first public appreciation of his contribution to society. The posthumous recognition was accepted by his daughter, Yelena Andriukova, and his grandson, Sergei Andriukova, on behalf of Arkhipov who “never said a word to his family because it was closed, secret information—he wasn’t allowed to talk about it.” Vasili Arkhipov understood on the most dangerous day in human history that nuclear warfare is simultaneously instantaneous and everlasting.“The decision not to start world war three was not taken in the Kremlin or the White House, but in the sweltering control room of a submarine.” Edward Wilson, The Guardian, October 27, 2012.For the complete story from the Russian submarine perspective, read “The Last Saturday of October.”

There are multiple factors at play that regulators need to take into account. They should do this dispassionately and based on the evidence.First, lets look at common semi-autonomous driving features and which manufacturers have them in production vehicles.There are multiple questions regulators need to ask regarding semi-autonomous features:Do they make driving statistically safer for humans?Do they reduce non-injurious collisions, e.g. fender benders?Do they introduce new and unexpected failure modes?Do they provide additional benefits such as emissions or fuel efficiency?Do they have other societal impacts which should be considered, and should regulation be applied as a result?Is inclusion of features required or merely not prohibited?Do they make driving statistically safer for humans?So far, the jury is strongly in on behalf of anti-lock braking and forward collision automatic braking. These features have strong net safety values. The last especially is so strongly supported that there are strong calls for making it mandatory on cars.The National Transportation Safety Board cites that, in recent years, almost half of all two-vehicle crashes involved a rear-end collision–claiming about 1,700 lives per year and causing 500,000 injuries. And the IIHS estimates as many as 1.9 million total crashes could be prevented or mitigated each year if all vehicles were equipped with forward-collision systems. Further, studies by the IIHS and others show that automatic emergency braking technology could reduce insurance injury claims by as much as 35 percent.Forward-Collision Warning With Braking to Become StandardandThe Insurance Institute for Highway Safety (IIHS) has seen a 7 percent reduction in crashes for vehicles with a basic forward-collision warning system, and a 14 to 15 percent reduction for those with automatic braking.Avoiding Crashes with Self-Driving Cars - Consumer ReportsABS is unequivocally safer too:The results show;1) ABS has a reduction effect of rear-end collision (1-38% reduction) and single-vehicle collision (10-33%),http://www-nrd.nhtsa.dot.gov/pdf/esv/esv21/09-0436.pdfTraction control is less equivocally a safety positive. Basically it enables you to accelerate in more circumstances under control, which can get you out of certain classes of trouble. You are less likely to slip backwards when starting on hills and more able to accelerate away from intersections when vehicles are sliding toward you from behind.Overall, the effectiveness of traction control to reduce or prevent crashes and injuries has not been well-documented. Nonetheless, due to its often being packaged together with ABS and electronic stability control (ESC), it is reasonable to suggest that driving a vehicle equipped with this trio significantly reduces fatal crash risk by up to 50%.Traction ControlOther semi-autonomous features, the ones in the news today, are still empirically unsupported by significant evidence.Tesla reported a May fatality last week in which a driver using Autopilot collided with the side of a left-turning 18-wheeler. In this case, Autopilot interpreted the side of the truck as a suspended sign, a known, documented and communicated weakness. The driver was inattentive and did not brake themselves.Michael Barnard's answer to What are the key facts in the Tesla AutoPilot-related fatal accident?Tesla included in their statement the following:This is the first known fatality in just over 130 million miles where Autopilot was activated. Among all vehicles in the US, there is a fatality every 94 million miles. Worldwide, there is a fatality approximately every 60 million miles.A Tragic LossThis is an interesting statement because it’s both accurate and relatively uninformative. There are three key elements to this:Autopilot is only intended for a limited subset of driving, mostly the most boring kind of driving with the fewest unexpected occurrence. It does, however, represent a large percentage of miles covered while driving and typically the driving which occurs at the highest speeds, and higher speeds are strongly correlated with fatalities.A sample size of one is not statistically relevant. Another fatal collision tomorrow would reverse the statement of safety that Tesla is projecting to make it appear less safe. Until there are a sufficient number of fatalities while Autopilot is in use to achieve statistical significance, this is merely an interesting data point.As I’ve said elsewhere, Tesla’s autonomous approaches start with a vehicle which has very strong survivability built in, a classic subsumption approach. It’s quite possible that there have been multiple Autopilot collisions in which the passengers were protected by the very significant crumple zones, air bags and other safety features of the car. It’s possible that lane-following cars from other manufacturers have seen fatalities as well, but they were not reported because the other manufacturer’s are not under the microscope in the way that Tesla is, for better or worse.Overall, it’s clear that most semi-autonomous features are reducing risks related to normal driving conditions. The most recent advances do not have a good empirical basis one way or the other.Do they reduce non-injurious collisions, e.g. fender benders?This one is pretty unequivocal. The data on active braking is above, and a lot of accidents are avoided by it at lower speeds for cars that have it.one of every five motor vehicle accidents take place in a parking lot, and 14% of all claims of auto damage involve collisions therein.Parking lot accidents: statistics, causes, and liability | MyParkingSign.com BlogParking lot accidents are the most common way fleet vehicles are damaged, according to research conducted by PHH Arval. In 2012, around 20 percent of fleet accidents occurred while parked or during parking. Basic defensive driving can help minimize collisions in these situations.Top 10 Things to Know About Parking Lot CollisionsMost of these accidents occur when backing into or out of parking spots. Parking assist features eliminate most of them.Further:NHTSA estimates that 22 percent (more than one-fifth) of children between ages 5 and 9 killed in traffic crashes were pedestrians. Most of these accidents occurred because drivers failed to see kids while backing up their vehicles.Parking lot accidents: statistics, causes, and liability | MyParkingSign.com BlogThe data on this is fairly unequivocal. Parking assist is a very positive feature from a regulatory perspective.Do they introduce new and unexpected failure modes?This is where it starts to get interesting.Various cars with semi-autonomous features have been videoed colliding at low speed with pedestrians which they did not recognize as such, or having slow speed collisions with immovable objects while being Summoned or the equivalent. This is imperfect technology. However, all manufacturers are also clear with all drivers that they are still ‘in control’ and must observe the car as it performs the semi-autonomous act and be able to abort or change the action. This isn’t intended to be fire-and-forget at present.There are no reports of injury from these slow speed collisions, most of which appear to have occurred while people videoed the feature in order to test it in various edge conditions. In other words, people tried to make it fail and succeeded. There has been property damage of minor types, but very little.In general, the edge conditions of failure appear to be of lower significance than the success conditions of avoidance.The Tesla fatality is an interesting counter-example, of course.Both Tesla and Mobileye, which provides key components of Tesla’s solution, make it clear that their technologies can fail to recognize trucks at 90 degrees to the road as obstacles, interpreting them as overhead signs instead. This is something that no human would likely do, but prevents frequent false positives and related braking from actual overhead signs.This is a new failure mode, and was part of what caused the fatality of the Tesla driver. To be clear, the truck driver turned across opposing traffic in broad daylight with a clear and straight road upon which he should have seen the black Tesla approach. The Tesla driver wasn’t paying attention and did not brake. Autopilot, which is an assistive technology, didn’t interpret the truck as an obstacle but as an overhead sign, something apparently assisted by the white side of the truck against a clear and bright sky. This was a collision which Autopilot did not prevent that was caused by two human errors, not a collision that Autopilot caused.This is the only fatality due to this condition that’s been reported however. A similar error occurred on a highway where a truck was stopped in the left hand lane and a Tesla collided with it, without fatalities.Statistically, this has to be compared to the distracted driving issues which Autopilot alleviates. That’s statistical work that still has be done over time. Regulators will be looking at this carefully, and running statistical models in an attempt to quantify this.It’s conceivable that a solution will have to be mandated for this known issue. And it’s extremely likely that both Mobileye and Tesla are working on improving the solution so that it makes this class of mistake less often, without increasing false positives from overhead signs which would make the feature unusable.Do they provide additional benefits such as emissions or fuel efficiency?This is an interesting side note. Anecdotal information from Tesla users makes it clear that the Tesla is more efficient when under Autopilot than under driver control. It accelerates and decelerates less and undergoes fewer directional changes which impart some additional friction. As a result, it goes further on a charge when under Autopilot. It’s presumably similar for internal combustion cars, but in those cases it is likely over-ridden by presence or absence of dual-clutch gearboxes and car efficiency mode choices.Regulators do regulate fleet efficiency and emissions. If Autopilot and other assistive driving technologies have a significant impact, they might consider them when assessing overall fleet performance. At present they aren’t considering this, but are measuring fuel efficiency and emissions under older testing protocols which would not be affected by semi-autonomous features in any event, as they mostly occur on dynamos instead of on the road.Do they have other societal impacts which should be considered, and should regulation be applied as a result?This becomes a very interesting question very rapidly, through induced demand and behaviour in intersections.The first factor related to induced demand is that semi-autonomous and autonomous cars in highway conditions improve the flow of traffic somewhat, basically by smoothing acceleration and deceleration and avoiding collisions.The second factor is that people using semi-autonomous features are finding longer commutes more manageable and productive.The combination is akin to creating a new lane on a highway. After a short period of time, the highway is just as congested as before. With the creation of apparent openness and ease of flow of traffic, more people use the road, creating congestion. That’s why building more roads doesn’t really help much. More people get to the same place in the same period of time, but each individual person takes the same time.The intersection challenge is more a future problem for fully autonomous vehicles which are navigating intersections. Passengers are expected to have transit or passenger rail like experiences in terms of acceleration and turning. They won’t want jerkiness.While autonomous cars increase highway vehicle capacity, they could decrease intersection vehicle capacity by 18% to 53%.Autonomous Cars Likely To Increase CongestionThe more autonomous features are engaged, the greater likelihood of more vehicles being on the road, usually with one person and in the future with no people at all. Just as regulators deal with fleet emissions for taxis, future autonomous fleets would likely be considered for regulation as well.Is inclusion of features required or merely not prohibited?All of this nets out to this question: should regulators ban features, prevent them from being used without key capabilities being in place, or actively require that they be present in vehicles?Active braking is definitely becoming something regulators are considering mandating.At present, there are few regulatory frameworks that ban autonomous cars simply because the technology is advancing faster than regulation. Some US states have regulations, mostly to say that they aren’t going to regulate. In the USA, the federal NHTSA is expected to bring forth guidance regarding vehicles with Level 4 autonomy, basically completely self-driving cars.Within six months, NHTSA will propose best-practice guidance to industry on establishing principles of safe operation for fully autonomous vehicles (vehicles at Level 4 on the scale established in NHTSA’s 2013 preliminary policy statement).Automated VehiclesWhat’s important to note about this is that the Tesla fatality was operating under Autopilot, which is Level 2 Autonomy per that scale. It will not be affected by these guidelines. It might, however, be affected by other guidelines, and the failure will undoubtedly be assessed for the NHTSA’s upcoming guidance.Other Sources:Semi-Autonomous Cars Compared! Tesla Model S vs. BMW 750i, Infiniti Q50S, and Mercedes-Benz S65 AMG - FeatureLike my content? Help it spread via Patreon. Get confidential consulting via OnFrontiers. Email me if you’d like me to write for you.