Is This the Future of Modeling Complex Systems?


This experiment gives a whole new meaning to “Bio-Computing.”

Talented and dedicated engineers spent countless hours designing Japan’s rail system to be one of the world’s most efficient. Could have just asked a slime mold.

When presented with oat flakes arranged in the pattern of Japanese cities around Tokyo, brainless, single-celled slime molds construct networks of nutrient-channeling tubes that are strikingly similar to the layout of the Japanese rail system, researchers from Japan and England report Jan. 22 in Science. A new model based on the simple rules of the slime mold’s behavior may lead to the design of more efficient, adaptable networks, the team contends. Read More at Wired. Read the Abstract in Science.

What I like about this story is the use of a technology and non-intuitive way. We spend a lot of time here at CSTEP creating models of complex dynamic systems, often employing agent based models which have so many interacting parts that there are often “surprising and unexpected results.”

In this case, a team of researchers set up and executed what could be described as a “analog computing program”. Obviously, if one wanted to change the parameters after the initial run, it is a bit more costly than simply changing a parameter in a piece of digital code. On the other hand, there may be some advantages to creating predictive and descriptive non-linear models using biological agents. I am particularly interested in what the programmers at CSTEP think.


Post a Comment

Your email is never shared. Required fields are marked *