Ants build transportation networks that optimize cost and efficiency at the expense of robustness. (17th October 2014)

• Record Type: Journal Article
• Authors: Cabanes, Guénaël; van Wilgenburg, Ellen; Beekman, Madeleine; Latty, Tanya
• Is Part Of: Behavioral ecology. Volume 26:Number 1(2015:Jan./Feb.)
• Page Start: 223
• Page End: 231
• Publication Date: 2014-10-17
• Subjects: ant -- construction costs -- efficiency -- graph theory -- robustness -- transportation network.; Animal behavior -- Periodicals; Behavior evolution -- Periodicals; Ecology -- Periodicals; Psychology, Comparative -- Periodicals; 591.5
• DOI: 10.1093/beheco/aru175
• Languages: English
• ISSNs: 1045-2249

Abstract

Lay Summary: Engineers face the challenge of designing transportation networks that are robust against damage, cheap, and efficient. Unfortunately, these design criteria cannot be optimized simultaneously. Trail-making ants face a similar dilemma but must build networks without the aid of computers or centralized leadership. We find that meat ant trail networks are more efficient and less costly than random networks. However, networks were less robust than random, suggesting that meat ants prioritize efficiency and cost over robustness. Abstract : Like modern human societies, many biological systems are dependent on transportation networks for the efficient distribution of resources and information. Network builders face the daunting challenge of optimizing conflicting network criteria such as robustness, efficiency, and cost, which cannot be optimized simultaneously. Here, we use graph and network theory to examine the trail networks of the polydomous meat ant Iridomyrmex purpureus . Meat ants build and maintain physical trails that connect their multiple nests to each other and to food resources. The resulting transportation network is used to distribute workers, brood, and food resources. We found that meat ants built low-cost trail networks that were relatively efficient. However, networks were less robust than comparable simulated networks, suggesting that meat ants prioritize cost and efficiency over robustness. Populous nests had higher connectivity than did less populous nests, implying they play a key role in resource distribution throughout the network. We propose that meat ant networks are an ideal model system for the development of network optimization heuristics.