Abstract: A system and method for routing the maximum number of trunks on a network with resource constraints is described. The invention provides a unique modification of a multi-commodity mathematical model that maximizes flow in a capacity constrained network. The problem of routing the maximum number of trunks through a communications network can be described as a multi-commodity problem by mapping each trunk as a commodity. However, in the multi-commodity model, the resources or capacity used by a unit of any commodity is the same. This condition is too restrictive for contemporary multi-rate broadband networks because a high bandwidth traffic trunk consumes more bandwidth than a trunk with low bandwidth requirement. The invention provides a method to modify the multi-commodity model so that the capacity utilized by different trunks does not have to be identical.

Abstract: A system and method optimize a fiber optic network arrangement in terms of load balancing and equipment placement at network sites. The present invention accomplishes this optimization by transforming a network of rings into a directed graph of nodes and arcs. The two parameters, which are load balancing and equipment placement, are set up as state variables in a mathematical algorithm for optimization. Following the transformation, descriptive linear equations are derived from the directed graph. The linear equations are then optimized for load distribution and equipment placement at network sites. The present invention may be applied to equipment placement and load distribution in new networks, as well as new load accommodation in existing networks by equating a state variable, corresponding to equipment placement, to a constant value.

Abstract: A method for rapidly restoring disrupted communications services between nodes of a network following a loss of service due to a circuit outage such as may occur, for example, as a result of an optical fiber cable break. The method is operable, however, irrespective of the transmission technology. Treating the telecommunications traffic as a commodity, a multicommodity model is developed using a measure of the commodities desired to be restored between network nodes and the capacity available for restoral in spare and impacted edges of the network. Action variables are established by use of representational "dummy nodes pairs" interconnected by "action edges." A multicommodity algorithm is applied to the model and to the decision variables to determine reroutes and actions to be taken within nodes of the network to effect the restoral reroutes.