0-introduction

Network Flow

We will extend the network problem-solving model to encompass a flowing through the network, with different costs attached to different routes. These extensions allow us to tackle a surprisingly broad variety of problems with a long list of application.

These problems and application can be handled within a few natural models that can be inter-related using reduction.

Network-flow models encompass huge range of problems that fall into general categories known as distribution problems, matching problems, and cut problems.

Distribution problems deal with moving objects from one place to another within a network and typify the challenges involved in managing a large and complex operation.

Merchandise distribution

Communication

Traffic Flow

Matching problems, the network is all possible ways to connect a pair of vertices, and our goal is to choose connection without including any vertex twice.

Job Placement

Minimum-distance point matching

Cut problems we are required to cut network into two or more pieces and are fundamentally related to graph connectivity problems.

Network Reliability

Cutting supply lines

We will study to important particular problems within the network-flow model the maxflow and mincost-flow problems.

In basic properties of flow network, where we interpret a network’s edge weights as capacities and consider properties of flows, which are a second set of edge weights that satisfy certain natural constraints. Next, we consider maxflow problem, which is to compute a flow that is best in specific technical sense.

Maxflow algorithms and implementation prepare for us to discuss even more important and general model mincost-flow problem, where there is cost ( another set of edge weights) and define flow costs, so we do a maxflow problem that is of minimal cost.

We do a classical generic solution to mincost-flow problem known as cycle-canceling algorithm; then we give particular implementation of cycle canceling algorithm known as network simplex.