Skip to Main Content
Robust network design for multispecies conservationAuthor(s): Ronan Le Bras; Bistra Dilkina; Yexiang Xue; Carla P. Gomes; Kevin S. McKelvey; Michael K. Schwartz; Claire A. Montgomery
Source: Proceedings of the Twenty-Seventh AAAI Conference on Artificial Intelligence. 27: 1305-1312.
Publication Series: Paper (invited, offered, keynote)
Station: Rocky Mountain Research Station
PDF: Download Publication (1.26 MB)
DescriptionOur work is motivated by an important network design application in computational sustainability concerning wildlife conservation. In the face of human development and climate change, it is important that conservation plans for protecting landscape connectivity exhibit certain level of robustness. While previous work has focused on conservation strategies that result in a connected network of habitat reserves, the robustness of the proposed solutions has not been taken into account. In order to address this important aspect, we formalize the problem as a node-weighted bi-criteria network design problem with connectivity requirements on the number of disjoint paths between pairs of nodes. While in most previous work on survivable network design the objective is to minimize the cost of the selected network, our goal is to optimize the quality of the selected paths within a specified budget, while meeting the connectivity requirements. We characterize the complexity of the problem under different restrictions. We provide a mixed-integer programming encoding that allows for finding solutions with optimality guarantees, as well as a hybrid local search method with better scaling behavior but no guarantees. We evaluate the typical-case performance of our approaches using a synthetic benchmark, and apply them to a large-scale real-world network design problem concerning the conservation of wolverine and lynx populations in the U.S. Rocky Mountains (Montana).
- You may send email to firstname.lastname@example.org to request a hard copy of this publication.
- (Please specify exactly which publication you are requesting and your mailing address.)
- We recommend that you also print this page and attach it to the printout of the article, to retain the full citation information.
- This article was written and prepared by U.S. Government employees on official time, and is therefore in the public domain.
CitationLe Bras, Ronan; Dilkina, Bistra; Xue, Yexiang; Gomes, Carla P.; McKelvey, Kevin S.; Schwartz, Michael K.; Montgomery, Claire A. 2013. Robust network design for multispecies conservation. Proceedings of the Twenty-Seventh AAAI Conference on Artificial Intelligence. 27: 1305-1312.
Keywordswildlife conservation, robust network design
- The Steiner Multigraph Problem: Wildlife corridor design for multiple species
- Large landscape conservation-synthetic and real-world datasets
- Wildlife corridors based on the spatial modeling of the human pressure: A Portuguese case study
XML: View XML