How too many tradeoffs can kill a project

It’s often about focus: designing a product to meet two overlapping objectives may be nice in principle but often leads to harsh tradeoffs, substandard performance, and eventual failure in practice.

Engineering design is more than just ensuring the various hardware and software components fit together and work as intended. During both planning and implementation phases, there will inevitably be tradeoffs among key performance parameters, power, size, weight, and other factors. (They are often called “compromises” to make them sound benign.) Balancing these tradeoffs while deciding how much weight to give each is where the art of engineering is added to the experience, expertise, analysis, and hard numbers.

But some projects try to do too many things, so the tradeoff matrix is harsh. Consider the flying car. We’ve heard about these since the 1950s, including the one popularized in the classic cartoon series “The Jetsons” (1962-1963) (Figure 1).

Every five years or so, there’s a spate of publicity about how flying-car prototypes are nearing final development, and we should be seeing them on the road and in the air soon. This “they are just around the corner” cycle then dies down, only to reappear about five years later. For example, just a few years ago, companies such as Terrafugia (a PR-coverage “star”) were getting lots of favorable attention, but now they are pretty much gone (Reference 1 and Figure 2).

However, when you think about it, the flying car is an example of very conflicting goals, expectations, and mandates. Start with the obvious: an airplane wants a streamlined body roughly in the shape of a cylinder, while a car needs a rectangular-like envelope with four wheels for stability. Most small aircraft have three wheels (nose wheel plus two), so right there, we have a conflict.

The list goes on, as the user expectations for a car as well as regulatory mandates (rear-view camera, advanced braking mechanisms, to cite a few) add to the complexity, weight, and power requirements but have little or no use in the air. For flying, many of the primary requirements are counter to those of the car, and some are actually counterproductive rather than just tolerable.

(Please note that when I say flying cars, I mean just that and not air taxis. There’s been a lot of interest lately in electric-powered short-hop air taxis with dozens of start-ups, and while those may not come to fruition, they do have a singular functional focus (see References  2 through 4.)

So why the enthusiasm for the flying car? I suspect it’s because at least “on paper” it sounds like a great idea, even if you ignore issues of managing airspace and related operational issues. The issue is this: can you build a single vehicle that meets our minimal current standards for cars and also small aircraft?

I think that’s an almost unresolvable conundrum even with our advanced electronics, materials, modeling tools, and more. The reality is that trying to do both will likely kill the project’s viability despite how attractive it seems in theory.

Sometimes it does work

There are cases where combining two functions does bring that much-desired 1+1>2 synergy effect; the basic digital multimeter (DMM) for measuring voltage, current, and resistance is an obvious one (Figure 3).

There are also oscilloscopes that merge their traditional time-domain function with a frequency-domain spectrum analyzer capability. This minimizes front-panel and bench-top space and reduces the number of leads and internal complex front-end circuitry. At the same time, the ability to look at both domains simultaneously and have the two functions share data can really enhance the test and measurement process.

But that is often not the case. After all, if you were doing some furniture assembly at home, you’d probably prefer a dedicated single-function screwdriver to the multifunctional, very popular Leatherman (Figure 4). The latter is optimized for minimizing weight and size, while the former is optimized for performance at one specific task. Those are legitimate tradeoffs to consider for a given project.

An excellent example is the development and evolution of lunar lander from the Apollo moon missions. The original plan was to have the third stage of the Saturn launch vehicle be the lander and return capsule. It would go to the moon and land (actually, it would back down), and the astronauts would then use that same third stage to blast off for the return trip. That multipurpose third stage would provide life support, navigation, and communication, plus a heat shield for reentry as well.

But as the planning team went through the details and numbers, they could not get the physics or operational issues to “work.” That third stage would be too heavy, too encumbered, and too unwieldy to be a lander, liftoff, and return-trip vehicle all in one. John Houbolt, working on his own at great career risk, proposed and championed the radical idea of instead using a super-light, disposable lander optimized solely for landing from the lunar module orbiting the moon, returning them from the surface to that orbiter.

Once that decision was made – and it was a very long battle to make it so (see EE World Related Content, Items 1 and 2) – the entire frame of reference and engineering “mindset” for the lander’s design changed completely. It could now be an ultralight, flimsy, ungainly, non-aerodynamic, and disposable unit that could barely stand on its own on Earth, as it no longer had to meet so many conflicting goals (Figure 5). The entire engineering basis shifted dramatically and decisively by focusing on a much more limited set of objectives.

When someone in engineering (or, more likely, in marketing) suggests adding additional features and functions to make a product satisfy the needs of multiple missions, it’s important to be honest about the ripple impact and resultant. Ask and re-ask: Why? What does it “cost?” What conflicts will arise if we try to do more? In what ways will trying to do more burden the project with “dead weight” and severely impact the original, primary function?

Will it be a case of 1+1 = 0.5 or perhaps even zero? It’s like trying to use a smartphone loaded with apps to try to do “everything” ­– it can do some of these very well, but for many others, it is awkward and perhaps using that desktop PC for some tasks makes more sense (as so many work-at-home people are now re-discovering). Sometimes, as the cliché goes, it is true that “less is more” – in this case, trying to do less is more likely to succeed.

EE Word Related Content

1. When audacious engineering leads to major success, Part 3: Apollo Mission profile
2. When audacious engineering leads to major success, Part 4: Apollo Lunar module
3. Be wary when they say, “That change should be no big deal “

References

1. Flight Global, “Terrafugia to shut down US operations: reports”
2. IEEE Spectrum, “Air Taxis Are Safe—According to the Manufacturers”
3. IEEE Spectrum,” EVTOL Companies Are Worth Billions—Who Are the Key Players? “
4. IEEE Spectrum, “Following the Money in the Air-Taxi Craze”