OPERATIONAL RESEARCH &
OPTIMISATION
Dr. Fang He
Email: [email protected]
Office: N7.118
University of Westminster
6DATA005W Operational research & optimisation
Network optimisation problems Outline
• We examine the characteristics of network models;
• Formulate some examples of these models;
• Study approach to their solution.
6DATA005W Operational research & optimisation
General Minimum Cost Flow Problem
6DATA005W Operational research & optimisation
General Minimum Cost Flow Problem
6DATA005W Operational research & optimisation
General Minimum Cost Flow Problem
6DATA005W Operational research & optimisation
General Minimum Cost Flow Problem- Important special
structure
6DATA005W Operational research & optimisation
General Minimum Cost Flow Problem
Flow
balance
Flow
capacity
6DATA005W Operational research & optimisation
• Minimum cost flow, a unified approach:
• Transportation problem; (no transhipment nodes, no capacity limits)
• Assignment problem; (a special type of transportation problem)
• Recap
• New:
• Maximum flow problem;
• Shortest-path problem;
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
Maximum flow-a prototype example
6DATA005W Operational research & optimisation
Scenario
6DATA005W Operational research & optimisation
Maximum flow problem
• Exercise: formulate the problem;
6DATA005W Operational research & optimisation
Note: The objective function; the values of bi; Check if it presents the special structure of the network model;
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
• Exercise: formulate the problem
6DATA005W Operational research & optimisation
Note: the direction of the arcs; the values of bi;
6DATA005W Operational research & optimisation
Network models
• Minimum cost flow, a unified approach:
• Transportation problem; (no transhipment nodes, no capacity limits)
• Assignment problem; (a special type of transportation problem)
• Maximum flow problem;
• Shortest-path problem;
• Minimum spanning tree:
6DATA005W Operational research & optimisation
• Slides 21,23,24,37 are from: Robert Sedgewick and Kevin Wayne Copyright © 2006
http://www.Princeton.EDU/~cos226
6DATA005W Operational research & optimisation
Minimum spanning tree from Wikipedia
• A telecommunications company trying to lay cable in a new neighbourhood. If
it is constrained to bury the cable only along certain paths (e.g. roads), then
there would be a graph containing the points (e.g. houses) connected by
those paths. Some of the paths might be more expensive, because they are
longer, or require the cable to be buried deeper; these paths would be
represented by edges with larger weights.
• A spanning tree for that graph would be a subset of those paths that has no
cycles but still connects every house; there might be several spanning trees
possible. A minimum spanning tree would be one with the lowest total cost,
representing the least expensive path for laying the cable.
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
• Exercise: how many cycle can you find with number of arcs
of 3, 4, 5,6?
6DATA005W Operational research & optimisation
• Another slightly better way to do it:
• Add the sub-tour (cycle) elimination constraints iteratively.
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
Minimum Spanning Tree Problem
• More efficient method maybe:
• Heuristics method!
6DATA005W Operational research & optimisation
Heuristic
• The word “heuristic” means “to find” or “to discover” by trial and
error.
• A heuristic is a rule of thumb to guide decision making or to find
approximate solutions to optimisation problems.
• Heuristic Search:
• Heuristic search refers to the techniques with the aim of
finding “good” solutions for a very hard optimisation and
decision problems within a reasonable amount of
computational time.
• But, it inability to produce guaranteed optimal solutions. And
it is difficult to know, in many cases, how far away from the
optimum a heuristic solution might be…
6DATA005W Operational research & optimisation
Heuristics
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
(greedy) Prim’s algorithm
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
6DATA005W Operational research & optimisation
• Q&A
