Stability of optimal traffic plans in the irrigation problem

From MaRDI portal
Publication:2078348



Abstract: We prove the stability of optimal traffic plans in branched transport. In particular, we show that any limit of optimal traffic plans is optimal as well. This is the Lagrangian counterpart of the recent Eulerian version proved in [CDM19a].


In the irrigation problem one is interested in minimizing the cost of connecting two nonnegative and finite measures \(\mu^{-}\) and \(\mu^{+}\) on \(\mathbb{R}^d\), where the transport is along a \(1\)-dimensional network and the cost of moving a collection of particles of total mass \(m\) along a stretch of length \(\ell\) is proportional to \(\ell \times m^{\alpha}\), for a parameter \(\alpha\in(0,1)\). A central question in the theory concerns the \emph{stability} property, that in the \emph{Lagrangian} framework considered here can be stated as follows: if \(\{\mu^{-}_n\}_{n\in\mathbb{N}}\) and \(\{\mu^{+}_n\}_{n\in\mathbb{N}}\) converge to \(\mu^{-}\) and \(\mu^{+}\) and if \(\{\mathbf{P}_n\}_{n\in\mathbb{N}}\) is a sequence of optimal traffic plans for the marginals \((\mu^{-}_n,\mu^{+}_n)\), converging to a traffic plan \(\mathbf{P}\), is it true that \(\mathbf{P}\) is optimal for \((\mu^{-},\mu^{+})\)? In the paper the authors answer affirmatively to this question for every \(\alpha\in(0,1)\), under suitable and necessary technical assumptions. The proof is by contradiction, and takes advantage of the stability result proved in [\textit{M. Colombo} et al., Commun. Pure Appl. Math. 74, No. 4, 833--864 (2021; Zbl 1468.90031)] where the corresponding problem in the \emph{Eularian} framework has been investigated. It is important to notice that the main result of the present paper does not follow straightforwardly from the above mentioned work, since in principle some curves of the limit plan \(\mathbf{P}\) may overlap with opposite orientations, producing cancellations at the level of vector measures. This phenomenon is excluded by the authors with a careful analysis which constitutes the core of the proof presented in this manuscript.











This page was built for publication: Stability of optimal traffic plans in the irrigation problem

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2078348)