620-262 Decision Making

620-261.

This subject introduces the essential features of decision-making situations encountered in operations research investigations. It develops a number of basic mathematical approaches to such situations and the techniques used to solve decision-making problems represented by these approaches. The theoretical foundations of these techniques are also considered. Students should develop the ability to construct formal mathematical models for practical decision-making situations; to solve a number of two-person games, including zero-sum and non-zero-sum games, cooperative and non-cooperative games; to use linear programming and dynamic programming techniques in the solution of a number of multi-objective optimisation problems; and to solve stochastic decision problems using techniques from probabilistic dynamic programming and Markov decision processes. This subject demonstrates the complexity of decision-making situations that may arise from business, economics, management, industry, etc., the extent and limitations of a number of operations research techniques used to solve such problems, and the important role that linear algebra, calculus and probability theory play in the development of these techniques.

A selection of topics in decision analysis will be covered, including single-stage and multi-stage decision models, in particular those using linear programs; zero-sum games; preference relations and optimisation; multi-criteria decision-making; decision trees; and probabilistic dynamic programming. Use of computer packages and internet resources will also be covered.

Up to 24 pages of written assignments due during the semester (10%); a 3-hour written examination in the examination period (90%).