17
CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Maximum Flow Neil Tang Neil Tang 3/30/2010 3/30/2010

CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

Embed Size (px)

Citation preview

Page 1: CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

CS223 Advanced Data Structures and Algorithms 1

Maximum Flow Maximum Flow

Neil TangNeil Tang3/30/20103/30/2010

Page 2: CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

CS223 Advanced Data Structures and Algorithms 2

Class OverviewClass Overview

The maximum flow problem Applications A greedy algorithm which does not work The Ford-Fulkerson algorithm Implementation and time complexity Another approach: linear programming An Application: maximum matching in a bipartite graph

Page 3: CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

CS223 Advanced Data Structures and Algorithms 3

The Maximum Flow ProblemThe Maximum Flow Problem

The weight of a link (a.k.a link capacity) indicates the maximum amount of flow allowed to pass through this link.

The maximum flow problem: Given a weighted directed graph G, a source node s and a sink node t, find the maximum amount of flow that can pass from s to t and a corresponding feasible link flow allocation.

Flow feasibility: Both the flow conservation constraint and the capacity constraint must be satisfied.

Page 4: CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

CS223 Advanced Data Structures and Algorithms 4

The Maximum Flow ProblemThe Maximum Flow Problem

Page 5: CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

CS223 Advanced Data Structures and Algorithms 5

ApplicationsApplications

Computer networks: Data traffic routing for throughput maximization.

Transportation networks: Road construction and traffic management.

Graph theory: Matching, assignment problems.

Page 6: CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

CS223 Advanced Data Structures and Algorithms 6

Flow Graph and Residual GraphFlow Graph and Residual Graph

Page 7: CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

CS223 Advanced Data Structures and Algorithms 7

Basic IdeaBasic Idea

Keep finding s-t augmenting paths until no such paths can be found in the residual graph.

Update the flow and residual graph according to the augmenting path in each step.

Page 8: CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

CS223 Advanced Data Structures and Algorithms 8

A Greedy Algorithm which Does Not WorkA Greedy Algorithm which Does Not WorkFind an augmenting path s-a-d-t with flow value 3 and update the flow and residual graphs as follows:

Page 9: CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

CS223 Advanced Data Structures and Algorithms 9

The Ford-Fulkerson AlgorithmThe Ford-Fulkerson AlgorithmFind an augmenting path s-a-d-t with flow value 3 and update the flow and residual graphs as follows:

Page 10: CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

CS223 Advanced Data Structures and Algorithms 10

The Ford-Fulkerson AlgorithmThe Ford-Fulkerson AlgorithmFind an augmenting path s-b-d-a-c-t with flow value 2 and update the flow and residual graphs as follows:

Page 11: CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

CS223 Advanced Data Structures and Algorithms 11

The Implementation and Time ComplexityThe Implementation and Time Complexity

If all the link capacities are integers, then the time complexity of the Ford-Fulkerson algorithm is bounded by O(f|E|), where f is the max flow.

A bad example for random path selection.

Page 12: CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

CS223 Advanced Data Structures and Algorithms 12

The Implementation and Time ComplexityThe Implementation and Time Complexity

In each step, find an augmenting path which allows largest the increase in flow using a modified Dijkstra’s algorithm. It has been proved that it terminates after O(|E|logCapmax) steps, so the time complexity is O(|E|2log|V|logCapmax).

The Edmonds-Karp algorithm: In each step, find an augmenting path with minimum number of edges using BFS. It has been proved that it terminates after O(|E||V|) steps. So the time complexity is O(|E|2|V|).

Page 13: CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

13

Another Approach: Linear Programming Another Approach: Linear Programming

LP in the standard form

CS223 Advanced Data Structures and Algorithms

An LP can be solved by existing algorithms in polynomial time.

Page 14: CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

14

Maximum Flow Problem - LPMaximum Flow Problem - LP

CS223 Advanced Data Structures and Algorithms

Page 15: CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

15

Shortest Path Problem - ILPShortest Path Problem - ILP

CS223 Advanced Data Structures and Algorithms

Page 16: CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

16

Maximum Matching in A Bipartite GraphMaximum Matching in A Bipartite Graph

CS223 Advanced Data Structures and Algorithms

A matching (a.k.a. independent edge set): a set of edges without common

vertices.

The maximum matching problem: find the matching with the maximum number of edges.

Page 17: CS223 Advanced Data Structures and Algorithms 1 Maximum Flow Neil Tang 3/30/2010

17

Maximum Matching in A Bipartite GraphMaximum Matching in A Bipartite Graph

CS223 Advanced Data Structures and Algorithms

A max-flow based algorithm