Should We Design Processes Like Airplanes?

As the ongoing discussion about standards for BPM continues, an interesting thread with Mr. Khan of Ultimus fame emerged. In response to his Don’t Forget the BPM Ecosystem article, I explained that standards always play a critical role in the development of mature industries, and drew further analogy to the airline industry that Mr. Khan felt compeled to challenge in a subsequent article. Unfortunately, his appreciation for the standardization that took place in both the commercial and general aviation markets seems to be fairly limited, and certainly not matched by facts that can be easily observed by any pilot today, myself included.

First, let’s go through Mr. Khan four points:

“i. There is no ’standard’ that I know for the design of the cockpit of an airplane or the dashboard/controls of a car. The cockpit of a Boeing 777 is totally different than that of a Cessna. The pilot’s ‘user interfaces’ are all are different!”

To the untrained eye, it might seem that all airplane cockpits are full of gauges, knobs, and screens that bare very little resemblance to each other when going from one aircraft to another. In reality though, modern instrument panels are the result of a relentless standardization process that makes the cockpit of a single engine Cessna 172 SP NAV III eerily similar to the one of a Boeing 787 (bar the number of circuit breakers). Over the past 3 years, over 95% of new General Aviation airplanes sold in the United States were equipped with computerized instrument panels (referred to as glass cockpits) built using the same technologies as those found in commercial jets, from the Airborne Heading-Attitude Reference System (AHARS) used for positioning and motion control, to the user interface replacing legacy instruments such as attitude indicator, heading indicator, turn indicator, altimeter, airspeed indicator, and vertical speed indicator.

To be sure, such standardization of instrument panels did not occur overnight. It started in the early days of aviation, but accelerated rapidly during World War II in order to facilitate the training of new pilots, allowing them to transition from one aircraft to another more easily. As a result of such a process, most traditional instrument panels (affectionately referred to as “six packs,” for they are built around the same six gauges) look pretty much the same, and any pilot trained on one can very easily transition to another. The same is true today for glass cockpits, which are increasingly using the same UI metaphors for representing standard information such as airpseed (vertical tape on the left), altitude (vertical tape on the right), or attitude indicator (inverted V bar decorated with various visual cues depicting the aircraft’s angle of bank, or the flight director).

And what is true for the cockpit of an aircraft is even more so for the dashboard of a car. The very vast majority of cars sold today display a circular speed indicator on the left, and a circular RPM indicator on the right. Only a handful of exotic cars (from Lamborghini in particular) or electric hybrids (like Toyota’s Prius) have broken away from such standardized analog displays, much like digital wristwatches did in the early 80’s. But if history is any indication, it is likely that analog dash components are here to stay. And the only reason why the RPM indicator might disappear in the future is because it makes no sense on a fully electric car.

“ii. There is no standard that I know for the engine of an airplane or a car. Yes they sometimes look similar, but they are different from each other. You cannot take the engine of one car and drop it into another car. There is no ‘portability’ of the engine. For airplanes of different sizes (a Cessna versus a Boeing 777) even the technologies used are radically different, even though they have many similarities (wings, rudder, landing gear, tail, etc.). The ‘engines’ are all different!”

This isn’t true either. Most mechanics will tell you that with the proper frame modifications, you can take the engine of any car and put it into any other. People do that not only for cars and motorcycles, but for planes as well. Modern aviation knows only three types of engines: piston engine (also called reciprocating engines), turboprop engines (turbines connected to a propeller), and jet engines (using Newton’s law of motion to provide forward momentum through action/reaction, also known as thrust). The former two are largely interchangeable (as long as you keep weights and balances in check), while the later two use the same grade of fuel (Jet-A fuel). In fact, so much standardization took place in the industry that you can mount a car engine on a plane, which is what Diamond did for the DA42 Twin Star, using a pair of Diesel engines originally designed to power entry-level cars manufactured by Mercedes-Benz.

“iii. Mr. Ghalimi mentioned the FAA in his post. I am sure the FAA uses very different standards to determine the air worthiness of a Boeing 777 versus a Cessna. So the quality and inspection standards are also different!”

As a matter of fact, they really do not. The certification procedures are pretty much the same. If you read the Federal Aviation Regulations (FAR), you will see that such certification procedures are not really based on how big the aircraft is, but how it is intended to be used (Part 91 for General Operating, Part 141 for Pilot Schools, Part 135 for Air Carriers & Operators). Even though certifying the Boeing 787 will be a longer and more expensive process than certifying the new Cessna Mustang Very Light Jet, the two certification processes will pretty much follow the same rules. And because the Mustang can be operated by a single pilot, its certification process must have been very similar to the one of the little Cessna 400 I am flying : both are certified for IFR operations, for flight into known icing conditions, and for high-altitude flying (above 18,000 feet).

“iv. You cannot take a pilot of small plane and put him in the cockpit of a Boeing 777 without a lot of training. The training is different!”

Once again, this is not true. All Airline Transport Pilot (ATP) start their training on single engine piston airplanes under Visual Flight Rules (VFR), then transition to Instrument Flight Rules (IF), multi-engine piston, turbine (turboprop), and finally move on to commercial jets. Most professional pilots will tell you that the transition to commercial jets is actually easier than the transition from VFR to IFR. In fact, an IFR-rated pilot (like myself) can get a type rating for Boeing 737 in just a week (Cf. FTI). And if you were to ask any instructor working at Flight Training International (the folks who provide such training), she would tell you that students do not struggle with the aircraft’s instruments or systems (the hydraulics of large jets are pretty complex), but rather with poor IFR proficiency. Staying current as an IFR-rated pilot is where the challenge lies, and this is precisely why so much standardization has been applied in this area, from the notation used for instrument approach plates, to the specification of ground-based equipment used to support such approaches around the world (VOR, DME, ILS, etc.).

Most of the training a pilot receives has very little to do with the aircraft she flies, but rather focuses on the missions to be flown. For example, the pilot of a Cessna 400 will receive the same high-altitude training as the pilot of a Boeing 747 or the pilot of a U2 plane, using the exact same decompression chamber. I followed such a training back in May 2008, at Beale Airforce Base. On my way there, I took a Boeing 747 Captain with me on a Cirrus SR22, and our training session took place immediately after the one given to a couple of rookie U2 pilot dressed in the same kind of suits used by the first astronauts to have been placed into orbit back in the 60’s. This standardized training taught me about the early signs of oxygen deprivation (tingling in the joints, loss of color perception, overall stupidity), all the way to its completely debilitating effects, to the point where I lost useful consciousness (Cf. article). Pretty cool stuff… But my point is: the training of a pilot is pretty much standardized, thanks to the standardization of aircraft and the overall airspace. And that’s a good thing!

After making his uninformed analogies, Mr. Khan goes on writing that standards should only be developed for individual components and interfaces, rather than complete systems. Again, the analogy to the airline industry shows the exact opposite. While very few standards have been developed for connecting heterogeneous sets of instruments to each other, massive standardization has been applied to systems themselves, be they at the aircraft level (as we’ve seen with instrument panels), or at the level of the entire airspace.

Today, the pilot of a tiny Cessna 172 can fly through Los Angeles International’s Class B airspace, alongside a Boeing 747 coming from London and an Airbus A380 on its way to Singapore. When I take a customer or a partner for a Bay tour on Cirrus SR22 (barely larger than a Cessna 172, but a bit faster), we fly directly over SFO, less than 1,000 feet away from heavy jets. And we all follow the exact same rules, fly the exact same approaches (unless the weather is so bad that you really do not want to be flying in a small plane), receive positioning information from the exact same GPS satellites, and use the exact same phraseology, on the exact same radio frequencies. And the same is true around the world, thanks to the adoption of a single language (English) for supporting communications between aircraft and the air traffic control system. The entire system has been standardized, and that is precisely what makes it all work so well.

Granted, such standardization might slow the pace of innovation, as Mr. Khan believes, but it makes the entire flying business a lot safer, and a lot more reliable for passengers. It’s also the only way to make it profitable (although barely nowadays because of the high fuel costs). Such standardization happened in the automotive industry, the PC industry, and even the database industry. While opponents to standardization of the BPM industry are quick to point out that business processes have too much variety in them to bear any meaningful amount of standardization, I would retort that we’re not trying to standardize processes. Instead, we’re trying to standardize the discipline of managing processes (Business Process Management), and the technologies used to do so (Business Process Management System).

Now, anyone should be free to develop and use a non-standard BPMS, much like anyone is free to build and fly an uncertified aircraft. From the FAA viewpoint, such a vehicle is classified under the ‘Experimental’ category, and does not need to comply to most of the requirements that other aircraft do. Unfortunately, experimental airplanes cannot be used for commercial purposes, and have the highest fatality rate of any airplane category. Would you fly such an airplane, Mr. Khan?