Finding Analogies to Braess’s Paradox in Biological Systems
Referenced article: Investigating Los Angeles’ urban roadway network from a biologically-formed perspective
Over the past few weeks, I have been particularly intrigued by Braess’s paradox, especially because I have grown up dealing with the frustrations of Los Angeles freeway traffic. The academic article, “Investigating Los Angeles’ urban roadway network from a biologically-formed perspective,” conducted by biology and engineering researchers Sophia Deen, Tatiana Kuzmenko, Hossein Asghari, and Demian A. Willette at Loyola Marymount University reveals a fascinating opportunity to connect Braess’s Paradox to academic research and unique biological processes.
In their experiment, researchers sought to use biologically-created networks to research the efficiency of existing infrastructure systems. The slime mold Physarum polycephalum was their organism of choice to research. The slime is essentially a one enlarged cell. It is not an animal, plant, or fungus. Nonetheless, it has exhibited very intelligent “decision-making” properties when placed in novel environments because it can grow through an entire environment, and then re-contract its body to follow the quickest and most efficient path. The researchers created and used proportional topographic maps of an area of downtown Los Angeles, which is a region of the city at the epicenter of four major freeway intersections: the 5, the 10, the 110, and the 101. They placed slime food source locations – oat flakes – to proportionally align with major intersections of highways and major roadways in downtown LA. The map models were sanitized to prevent any other fungal growth or contamination. 17 trials of a 72 hour slime exploration period were conducted. The results showed that the slime’s decision-making of the Los Angeles roadway model found shorter paths than the existing Los Angeles roads. To be precise, there were networks that were 65.6 km (plus or minus 13.8 kilometers due to standard error deviation) of proportional length shorter than the lengths of LA’s current road network.
A key difference in physarum’s methods of decision-making and the decision-making of human civil engineers is the existence of preferential attachment in the construction of roadway networks. Preferential attachment states that nodes will connect to other nodes that have more existing connections and resources. In the existing freeways, new shopping centers or residential areas will be preferentially connected to a node with its own high population center. However, this will ultimately diminish the efficiency of the travel time. Preferential attachment is related to Braess’s Paradox, because when new roads are built with the sole intent of connecting an isolated residential area, for example, directly to a major intersection, more time is spent commuting as a result.
The findings from this experiment can tell us how to make transport networks more efficient. While the existing roadways that aim to connect multiple nodes to large, central intersections are unlikely to be completely taken down in the near future, the findings show that, as impatient many drivers get while navigating freeways, adding more routes between major roadways can raise travel time and increase congestion. After looking at the visual representations of real world highways and the slime mold’s exploration, shown below, the illustration of the slime mold’s network emulates that of high-speed trains or subways. There are fewer edges between the nodes. Los Angeles has already begun implementing energy-efficient metro rail trains, and I believe this is a step in the right direction.
