Stochastic Search and Fuzzy Modelling for Real World Complex Systems

The modelling of real-world complex systems is an area of ongoing interest for the research community. Real-world systems present a variety of challenges not least of which is the problem of uncertainty inherent in their operation. In this research the problem of inventory management was chosen, the goal was to discover whether Interval Type-2 Fuzzy Logic was an appropriate choice to represent the intrinsic uncertainty present in a large Supply Chain operation. Stochastic search algorithms were used with a series of Interval Type-2 Fuzzy Logic models to identify suitable inventory plans for a set of problems culminating in a large-scale real-world problem supplied by collaborators on a Technology Strategy Board research project (ref: H0254E). In addition to illustrating the suitability of Interval Type-2 Fuzzy Logic for such a problem, this research also considers how the advantages offered by Type-2 Fuzzy Logic can be exploited to produce a solution that is preferable to an equivalent Type-1 Fuzzy Logic system.