Constructal Law and Increasing Returns: The Evolution of Machines and Catching the Next Wave

In The Evolution of Airplanes, A. Bejan, J.D. Charles, and S. Lorente propose a compelling framework which describes the evolution of the airplane based on constructal law.

Constructal law states that, for a finite-size system to persist in time, to live, it must evolve in such a way that it provides easier access to the imposed currents that flow through it.

As an example, the engine mass of an airplane is proportional to the size of its body. The engine, or a component thereof, is alive with currents that flow inside it by overcoming resistances and generating entropy. This entropy penalty is smaller when the engine is larger. So larger is better.

In the same time, the airplane must burn fuel in order to transport its engine. So the fuel penalty is proportional to the weight of the engine. And, in this case, smaller is better.

From this conflict emerges a trade off. And airplane size and engine size must be proportional: whether on the Farman F60 or on the Boeing 737.

But which of the two, the F60 or the 737 or both, is a better solution to this problem of optimization?

Taking the analogy with biological evolution one step further, we would like to suggest that the trade off is reached for a given place in the economy of nature: those systems that are better adapted to their market will flourish.

The 5-ton F60 was adapted to its market in the 1920s. The 50-ton 737 was adapted to its (newly created, or newly available) market in the 1970s.

And in the 1970s, airplanes such as Embraer’s 5-ton EMB110 took over from the F60 as they were better adapted to this segment of the economy of nature.

Can we Take this Framework One Step Further? #

We would like to suggest that, as appealing as this may be, it only addresses one part of the picture–optimization–but leaves out another important component: innovation.

Looking at the evolution of airplanes during the past half century, there has been an average of 1.2% yearly decrease in fuel consumption per passenger. As important as this tendency may be, it is clearly the result of successive small improvements, or optimization.

But how do airplanes compare to other means of transportation such as cars, or even bicycles and horses?

If we attempt a link between constructal law and W. Brian Arthur’s increasing returns, the metaphor of the evolution of technology may be interpreted as follows.

• One technological wave corresponds to one constructal genus, such as the airplane, which has evolved from the Farman F60, and diversified into the Boeing 737 and Embraer EMB110, and then into the Airbus A380, Boeing 787, and Bombardier Dash 8.
• Two technological waves correspond to two distinct constructal genera, such as the airplane and automobile, or the smartphone and feature-phone.

And if we define a new genus–a new, distinct product category–as one that does not fit with the allometric rules of an existing genus, then we can propose that developing a new, adapted genus, leads to successfully catching a next-wave, which leads to a positive feedback loop and increasing returns.

Does it Add Up? #

The figure below suggests proportionality between the weights and speeds of different technological species within one genus. For example, the weights and speeds of different Aircrafts (such as the A320, A380, B737, and B747) seem to be proportional and distributed around the same line.

It also shows that, although a similar trend may be conserved, different allometric rules seem to apply to different technological genera. For example, Aircrafts, Watercrafts, and Cars are distributed around different lines.

Of course, this is merely a first look at the matter. The sample size is small and the data must be double-checked.

But the range of possibilities is inviting. Like Bejan et al., we note that the Concorde is a misfit. We also notice, at first look, that a technological jump has occurred between classical Rotorcrafts and the High-speed types currently being contemplated by the industry (i.e. X2, X3, V22).

A similar technological jump has occurred between early automobiles and post-1970 cars, rendering them into two different (albeit close) genera. Of course, Formula 1 racing cars form a genus of their own.

The Bicycle-like genus is also an intriguing one as it contains the Segway, canting horses, and the pioneering Benz Patent-Motorwagen!

Application to Other Industries #

Following the same logic, the figure below suggests that Feature-phones, Smartphones, Mini-tablets and Tablets are four different genera. And it indicates proportionality between the processor speeds and screen sizes of the species within each genus.

The speeds and screen sizes of Feature-phones are proportional, and so are the speeds and screen sizes of Smartphones, Mini-tablets, and Tablets.

Again, the sample size is small and the data must be double-checked. But a number of facts are noteworthy when looking at the matter from this perspective.

For instance, the figure above suggests that BlackBerry was not the first to catch the Smartphone wave: in effect, the BlackBerry Bold and BlackBerry Curve series, for example, are on the red-square Feature-phone line and not on the blue-diamond Smartphone line. Put in another way, BlackBerry did not invent the Smartphone, they developed a powerful Feature-phone.

Instead, the first Smartphones seem to have been introduced by Nokia. The Nokia 7710, Nokia 9210, Nokia 9300, and Nokia 9500 are all to the left of that blue-diamond Smartphone line. But they were perhaps ahead of their time, lacking a clear market segment in the economy of nature.

With the introduction of the original iPhone, Apple may have succeeded in catching a new wave by tending to (or creating) a new market segment. Instead of a bicycle, they proposed a car—and the market was ready for cars.

Nokia, on the other hand, almost caught the next wave. But the market or technology was not ripe yet.

The Nokia 9210 Smartphone, first released in 2000.

Evolution vs. Revolution #

In this framework, a clear distinction is made between new products in pre-existing categories and entirely new product categories. The former are similar to new species within an existing genus, whereas the latter are entirely new genera.

For an aircraft manufacturer, developing the A380/B747 after having developed the A320/B737 amounts to optimization, or the evolution and diversification of existing species within one genus. Within this framework, even if Airbus or Boeing were someday to develop an aircraft that is ten times larger than the A380 and that flies at almost the speed of sound, they would still not be innovating.

The corollary is that in order to innovate, a company needs to create a completely new product category—such as space transportation, massive public transportation, or perhaps unipersonal flight systems.

An important take-away from this is that both evolution and revolution are key. The trick is to be crystal clear about what we are aiming at with each new product development because the stakes and consequences are not the same. The two cannot be managed in the same way.

And, if it holds, the framework presented in this paper can help us determine the nature of our industrial endeavors.

The Case of the Concorde #

As for the Concorde, there are two ways to look at this industrial undertaking.

• An efficiency seeking manager may see the Concorde as one of those products that shouldn’t have been made. It clearly did not obey robust allometric rules and was vowed to a probable commercial failure.
• A visionary leader, on the other hand, may see the Concorde as a next-wave that wasn’t. Like in almost all breakthroughs, the successes are rare exceptions. So even though the Concorde failed commercially, it was still worth a try.

Author: E. Dib #

References #

• A. Bejan, J. D. Charles, S. Lorente, The Evolution of Airplanes
• W. Brian Arthur, Increasing Returns and the Two Worlds of Business
• E. Dib, Increasing Returns: Innovate to Remain Relevant

Further Reading #

• C. Darwin, On the Origin of Species