In the regional coverage problem, it is necessary to coordinate and control the behavior of individuals and the distribution of swarm robots. This paper combines the principle of activator inhibition and the principle of stimulus response of bee colony, and proposes an algorithm (AISRA) combining activator inhibition and stimulation response. AISRA achieves close collaboration between individuals through the principle of activator inhibition; through individuals interaction, the distribution of robots in unknown regions is adjusted to give full play to the role of each robot. Simulation results and discussion show that AISRA can realize the cooperation between individuals and adjust the position distribution of robots, thereby improving regional coverage, reducing the number of repeated coverage, and improving the regional coverage efficiency of swarm robots.

The rising of online networks have provided a new ecosystem of individual interactions. Of particular interest are social contagions on duplex networks. Previous studies assume that each network is homogeneous, however, empirical analysis shows that nearly the online networks are heterogeneous. Motivated by this, we studied the threshold model on duplex networks with different online-offline coupling. Compared with uncorrelated coupling, the maximum positive coupling weakens systematic robustness for the sparse network, while enhances the robustness for the dense network. The maximum negative coupling, however, plays an opposite role. These effects are strengthened when the online network is heterogeneous.

For the bipartite consensus problems of first-order and second-order multi-agent systems, we design a consensus protocol based on the edge dynamic to make the system achieve edge dynamic bipartite consensus. With the help of line graph and equivalent transformation, the sufficient conditions for achieving edge dynamic bipartite consensus of first-order and second-order multi-agent systems under directed graph without time-delay are obtained. Furthermore, non-uniform communication time-delay is added to the protocol. Lyapunov asymptotic stability theorem is used to obtain the sufficient conditions that first-order and second-order multi-agent systems under directed graph with time-delay that reach edge dynamic bipartite consensus. Finally, using Matlab to simulate the example, the rationality of the conclusion is verified.

The static structures of developer collaboration networks in OSS communities have been widely discussed in the literature. However, there has been limited research on the evolutionary process of these networks. In this paper, we construct the developer collaboration network of the Cloud Foundry community by code-collaboration relationships and analyze its structure and evolution. The results show that the collaboration network evolution is characterized by the structural variations of its giant component, namely “loosely connected” state, “chain-module” structure, and multi-modular small-world network with a “core-periphery” structure. Meanwhile, the evolution of the giant component is intrinsically related to the sub-projects. Besides, the collaborative relationship between developers shows the combined features of preferential-attachment, homophily, and heterophily. The overall results may be not only helpful to deepen our understandings of the collaborative pattern in OSS communities, but also enrich the studies of open innovation in the large-scale crowd.

The “Internet +” mobile phone recycling mode makes it possible to recycle obsolete mobiles in a convenient, efficient, normative and safe way. In order to reveal the influence of different factors on the operating performance of the recycling mode, the functional structure and influencing factor system of the recycling mode were analyzed based on four links, including supply, collection, recycling and sales. Accordingly, the theoretical model of the interaction among the factors was established by using the method of system dynamics. Finally, the effects of each link and its factors were simulated by changing the government capital investment. The results show that supply and collection links play significant roles in operating performance and are sensitive to capital input, while recycling and sales links have a delayed response. Consumers' perceived risk, perceived benefits and platform's service quality are dominant factors for the improvement of recycling performance. The poor economic benefits of recycler and the inactivity of second-hand platform are the main restraining factors.

Using spillovers index approach and spillover network, this paper explores the characteristics and sources of volatility spillovers effect across 11 major stock markets. The results show that the US stock market is the center of the international volatility spillover network. The three markets of UK, France and Germany have obvious mutual spillovers, but the spillovers to other markets are weak. The total volatility spillovers index is volatile, rising significantly during the period of financial turmoil. The economic fundamentals and market contagious are both important factors.

Aiming at the fourth party logistics routing optimization problem under uncertain environment, by considering that people have different behavioral characteristics in different situations, by considering the transportation time and transportation cost of transportation tasks, the fourth party logistics routing optimization model is established based on the ratio utility theory and the value function of prospect theory. According to the characteristics that the fourth party logistics choose routes and the third party logistics simultaneously, an enumeration algorithm, an ant colony system and an improved ant colony algorithm areusedto solve the model. The effectiveness of the model and the algorithm is verified bynumerical examples.

In order to resolve the dynamic characteristics of pig price fluctuations and transmission from January 2000 to September 2018, the paperestablishescomplex network and analyzes model characteristics of the network,weighted aggregation coefficient of core modal nodes of hog′s price fluctuation, transmission models of K-core and transition time characteristics. The results show that the sum of the weighted degree distributions of the six modes is as high as 79.26%, which accounts for the dominant position of conduction. Modal transmission is the main transmission pathways of price volatility which is a one-way closed loop composed of six main modes. The price fluctuation of hog′s has strong gradual, persistent and periodic characteristics, and the above modal transitions have strong temporality. The modal transitions with a downward trend mainly occur in March, April and May, while the modal transitions with an upward trend mainly occur in July, August, September, October and November.