Search

The highly endangered giant panda (Ailuropoda melanoleuca) is the world's most widely recognised conservation icon. Population dynamics models can get on track to a healthy population for giant pandas, and assess its development over time. This paper proposes a new way to study population dynamics of giant pandas in captivity by means of membrane computing, a bio-inspired computational paradigm based on processing multisets of objects within a cell structure, following a series of evolution rules. This framework is used to model the intrinsically stochastic real-life evolution processes of giant pandas, based on pedigree data. This is the first attempt to collect and analyse so complete source, and to investigate population dynamics of giant pandas based on them. Pedigree data consisting of the number of giant panda individuals per age are provided by Chengdu Research Base of Giant Panda Breeding of China. A special simulator has been developed to assist in the design and formal verification of the mathematical model presented. Particularly, the model feasibility, effectiveness, soundness and robustness have been validated by the simulator, which also enables decision-making based on the simulation results of conducted virtual experiments. [ABSTRACT FROM AUTHOR]

Even though the giant panda's extinction status is downgraded from endangered to vulnerable, but the animals still face an uphill battle for survival. Moreover, the number of giant pandas is also rare. With the increase of captive individuals, wild release has become an inevitable task that human face. It is important to evaluate at least how much giant pandas are remained for being released to the wild. Several methods have been presented in the literature to assess minimum viable population (MVP) of giant pandas. However, each of these methods is designed only based on population sizes. In this paper, a membrane computing model based on behavioral biology is proposed to evaluate the MVP of the species. An application of the proposed data processing to binary tree whose nodes are composed of population sizes is also proposed. An binary tree is generated by integrating the following three techniques: (a) a random technique to generate the vast majority of populations; (b) a modified binary search method to obtain an order population size sequence; (c) a modified inorder traversal to generate an order binary tree and a parent tree. This is the first attempt to apply membrane computing model to assess the MVP of giant pandas. The effectiveness of the presented general PDP system is verified by means of the numbers or amounts of subpopulations. Besides, we also provide the population distribution of the obtained MVP. [ABSTRACT FROM AUTHOR]