SAMS Subject Classification: 25, 26

The demand for agricultural development to build stable and sustainable food systems has resulted in the need for plant breeding and genetic optimisation. The significant crop improvements required for adapting to the growing global population demands development of innovative optimisation methods within plant breeding programs. In practice, however, as crop populations (cultivars) increase in size, problem complexity increases exponentially. To approach the optimisation of these problems with significant complexity, the implementation of metaheuristics is required to yield high-quality solutions with high computational efficacy. The diversity of plant breeding programs and crop scenarios calls for investigation into developing a decision support framework for selecting appropriate algorithmic designs with suggested functionality to improve optimisation performance. The framework is aimed at providing breeders with improved performance over generic algorithm designs available in the literature and typically implemented in practice.

A generic framework for algorithm design selection in the context above is proposed in this poster, and further follow-up work anticipated is described; including a computerised instantiation of the framework in the form of a practical decision support system.