Reflection on my research at Chalmers

My research as a post-doc in the Department of Product and Production Development at Chalmers looks into variation in product design and manufacturing and how it affects important product development goals like sustainability. So, what does all of that mean?

Sustainability is a hot word these days (in fact, the growth in the use of the word is itself unsustainable: http://xkcd.com/1007/), and everybody seems to be trying to get into research and businesses that can tout sustainability. It sounds good, it's "green", it's good for the planet, and, in some cases, it sells. Today, governments care about sustainability, so they fund research projects, customers care about sustainability, so they buy things that they think are good for the planet, and hippies care about sustainability, so they hug trees and don't shower. But, what is sustainability? Most people usually think about the environment and how our actions are causing global warming or destroying the ozone layer or killing the polar bears. Some academic circles have adopted an idea called "triple sustainability", which involves three types of sustainability: People, planet, and profits. In other words, there is social sustainability, ecological sustainability, and economic sustainability. The truth is, developing a business around the first two types without caring about economics means that your business will fail before it can have a lasting impact on social and environmental goals. Social is the hardest one to define and measure, since it deals with things like community programs, health and safety of workers and customers, and human rights. This one is kind of on the back burner for my research right now, so I'm mostly looking at economic and ecological sustainability. This means I care about how design decisions affect manufacturing costs and revenues (economic) and the impact on the environment through resource consumption, emissions, and waste (ecological). One of my big questions is to find out what tradeoff exists between these two objectives. In other words, how many dollars/euros/kronor does it cost to reduce one pound of carbon dioxide emissions or some other environmental measure?

Now, back to this idea about variation in product design and manufacturing. In case you didn't know, no two items are ever exactly alike. The internet says that an iPhone 5 is 4.87 inches long, but if you measure the length of one iPhone 5, it may be 4.8701 inches, while a second iPhone 5 may be 4.8697 inches long (note: I have no idea what the actual tolerances are for iPhone parts, but I'm just using these numbers for the sake of argument). Chances are, you're either thinking: "Who cares?", or "That's ridiculous that a $900 device isn't built with higher precision!". I agree with both of you, and the question is really about where you draw the line. You could increase the precision and make it fit between 4.869999 and 4.870001 inches, or you could reduce the precision and make it fit between 4.8 and 4.9 inches long. The problem occurs when the difference is noticeable. It is especially important when you have a bunch of parts that have to fit together - if you noticed that the front cover of your phone is too wide and sticks out in relation to the back part, you probably won't be very happy with the appearance and feel of your phone. Even more problematic is when it affects the functionality of your phone; if the wiring inside isn't precise enough and something becomes disconnected, then you have some serious problems, angry customers, and phones being thrown away and going into landfills. Why don't we just make things with the highest precision possible? Because the more precise it is designed to be, the more expensive it is to make. So, designers have to balance how much variation they can tolerate with the costs.

So what exactly am I doing about this? My research group at Chalmers is involved in analyzing how to design products while taking this type of variation into account. We have developed software that can analyze how variation in a part (like the length of a wire) affects variation in an assembly (like the way two wires fit together for soldering) and how all of that variation might affect functionality (like does the phone/robot/car work?). My part in this is to bring in sustainability outcomes and look into how design decisions affect things like economic, ecological, and social objectives. I use models and simulations to predict how a decision like the tolerance of a dimension (the amount of variation the designer says is tolerable in the manufactured product, for instance, "within 1 millimeter of the designed length") or the material chosen will affect things like economic, ecological, and social objectives.

Finally, the big question: Why do we really care? Well, for one, I'd like to know how the three types of sustainability trade off when it comes to certain design decisions. I can choose to allow more variation in a part to reduce the cost by a dollar (economic sustainability), but that might raise the social or ecological costs. When a designer makes these decisions, they probably want to optimize for some objective, which is usually economic, but if they add in some amount of ecological or social criteria, the optimal design will change. Obviously, from a society standpoint, we want companies to value environmental and social goals, so how do we do that? There are two ways that I'm currently looking into. The first is letting the customers choose, which requires that they have complete information about the sustainability repercussions of every product on the market. One challenge there is in governments or industries mandating that this information is standardized, accurate, and available (imagine every item in the store having some "greenness" score right next to the price tag), but that probably can't happen until we prove that it's important. For one study, I ran a survey to find out how people value a product's price, environmental friendliness, and appearance of quality for a hypothetical smart phone, which shows us how people trade these things off when they buy something (or how they say that they trade them off). With all of the craze for eco-friendly or "green" products, along with the outrage over the worker health and safety scandals with things like child sweatshops and fires in textile factories, the market could give producers financial incentives to improve all three sustainability goals. The second way is to raise taxes for environmental and social impacts (like a gasoline tax), so another study I'm working on is to show how different types of legislation would force social and environmental impacts to become economic motivation. This could result in some meaningful recommendations for how governments can promote triple sustainability and how designers can optimize their products.

Is this the same kind of thing I did at Michigan? Sort of. At Michigan, I also looked at design optimization and tradeoffs using computer simulations, but that work dealt specifically with car and military vehicle safety rather than sustainability. There, I also looked at how customer demand and government intervention can influence the way companies design their vehicles. So, why did I move to Sweden to do this? First: "Why not?" Second, I wanted to apply my modeling and optimization and tradeoff analysis techniques to more general product design, which is an opportunity that my department at Chalmers could afford me. My current position also is entirely funded by the Swedish government, which means that I'm free to work on any research topic or product that I'd like to (i.e., I'm not restricted to working on any specific product or any specific research question). Some day, I'd like to influence the design of products either by designing them myself or helping policy makers develop laws, regulations, and guidelines that will positively impact how companies design things. Hopefully this can result in more environmentally and socially sustainable behavior by people, effectively saving the world from all of its problems.