Numerical approach to centrality of optimal transportation networks

Abstract

We study hierarchical properties of optimal transportation networks with biological background. The networks are obtained as minimizers of an energy functional which involves a metabolic cost term of a power-law form with exponent $\gamma >0$. In the range $\gamma \in (0,1)$, most relevant for biological applications, the functional is non-convex and its local minima correspond to loop-free graphs (trees). We propose a numerical scheme that performs energy descent by searching the discrete set of local minimizers, combined with a Monte-Carlo approach. We verify the performance of the scheme in the borderline case $\gamma =1$, where the functional is convex. For~a~particular example of a leaf-shaped planar graph, we evaluate the global reaching centrality (GRC) of the (local) minimizers in dependence on the value of $\gamma \in (0,1]$. We observe that the GRC, which can be understood as a measure of hierarchical organization of the graph, monotonically increases with increasing $\gamma$. To our best knowledge, this is the first quantification of the influence of the value of the metabolic exponent on the hierarchical organization of the (almost) optimal transportation network.

7 pages, 4 figures; submitted to the Proceedings of the ENUMATH 2023 conference