Defining system boundaries

Boundaries are arbitrary. Choosing the best ones is a big deal.

Boundaries are arbitrary. Choosing the best ones is a big deal.

One of the most critical parts of systems modeling is defining the boundaries between systems. Different boundaries will lead to different system models, so choosing the “best” boundaries for a modeling goal is really important. Here’s how I do it.

Continue reading


Art Markman reinvents the wheel on innovation

Art Markman wrote a piece for the Harvard Business Review on innovation (11 Jan 2012) in which he advises “Don’t Think Different, Think About Different Things.”  Unfortunately, his approach is obscure and shrouded in weird zen stuff that’s just not necessary.  I won’t be describing here how anyone can be innovative at anything, but I will describe one method, which you can use to increase the odds of being innovative, is well described by Markman’s (rather tired) example of the Dyson vacuum.

Markman writes “When you need to solve a problem in a new way, you have two options. One is pure research and development. The other requires finding knowledge (which we already know) that offers a novel solution.”  He goes on to advise us that we should seek meanings in proverbs to learn how to think differently.  He’s really talking about lateral thinking here, which is also nothing new.  Lateral thinking is hard, because there’s no real method to it; one just practises it until one gets good at it – on the assumption that one can apply lateral thinking to any domain equally well.  The assumption is ridiculous, but that’s another story.

The story of how James Dyson re-designed the vacuum cleaner, as recounted by Markman, is an excellent example of a design method that is quite well known.  We don’t need all the zen-i-ness that Markman suggests.  Here’s how it works.

First: identify a shortcoming.  In the case of the vacuum, it’s that they lose suction.  This should be apparent from user feedback about existent vacuums, product tests, etc.  However, this isn’t quite as easy as it seems, because the only way you can notice is something is bad, is if you have something good to compare it to.  After all, fish don’t know they’re wet.

Second: question the premises.  By looking at the alternative vacuum designs that all suffer the problem, one looks for the common features.  The bag is the common feature.  The assumption is that a bag is necessary.  One then simply asks Why is the bag necessary? In some ways, this step can be done without step 1; but knowing what shortcomings exist is a product can be a big help.

Third: look for functional alternatives to the assumption.  The bag serves certain functions in the vacuum cleaner.  Any technology that can provide the same function is a candidate replacement for the bag.  One can then evaluate those technologies to look for ones that are “innovative.”

That’s all there is to it.  Dyson didn’t look very far to find something.  Indeed, as with so many innovative designs, the answer lay in the designer’s own experience. It might be difficult to imagine that Dyson saw a similarity between domestic vacuums and sawmills, but it becomes simple if you think about function instead of structure.  Vacuums suck dust up using air; the vacuum must therefore internally separate the dust from the air.  That’s exactly what the function of the vacuum bag is – to trap dirt and let air through.  In the sawmill, an industrial cyclone is used to separate sawdust from air.  Notice that the functions served by the bag on the one hand and the cyclone on the other are exactly the same.  It’s much easier to see the connection when you think about function.

And that’s exactly what one needs to do to find an innovative solution to this problem: look for ways of providing the same function as the bag provides.

Indeed, this case is also an example of another technique of developing potentially innovative designs: change the state of matter used to provide a function.  The bag uses mass (the bag itself) to separate dust from air.  One could use liquid to do this – that’s the principle of those wet vacuums for carpets.  One could also use gas – which is how the Dyson vacuum works.  Finally (since matter and energy are related thanks to Einstein), one can use energy – a static electric charge (like in a Swiffer) that attracts and holds dust.

So in the end, all the mumbo-jumbo that Markman espouses might make a consultant rich, but it’s not necessary to design creatively.  Simple methods like this one can work too.

And if after reading this, you go out and invent something amazing, I won’t expect payment.  Just think kindly of me.