Percolation controled in large-scale network

The approach is considered to study of the large-scale (complex) networks, connected with model percolation upon their rib information, transport etc. traffic. They are considered appearing herewith clusters of object and statistical phenomenon's, connected with them. Alongside with threshold stochastic percolation is researched point to maximum clusterization, in which number formed clusters is maximum. It Is entered notion percolation controlled of the zone of the service, which is realized in two phases: stochastic base on the first with comparatively small concentration object, not providing stochastic percolation but on the second in inter-cluster intervals are entered additional objects to achieve the most short percolation way through stochastic base under relatively low gross amount object and minimization of the general expenses. The results are received by way of statistical modeling of the greater networks on square matrix. The got results are used for analysis greater information, computing and social net-works, as well as for calculus of the safe ways in disadvantage casual ambience.