%0 Journal Article
	%A Jagannath Aryal and  Don Kulasiri and  Dishi Liu
	%D 2008
	%J International Journal of Computer and Information Engineering
	%B World Academy of Science, Engineering and Technology
	%I Open Science Index 14, 2008
	%T Optimization Approaches for a Complex Dairy Farm Simulation Model
	%U https://publications.waset.org/pdf/935
	%V 14
	%X This paper describes the optimization of a complex
dairy farm simulation model using two quite different methods of
optimization, the Genetic algorithm (GA) and the Lipschitz
Branch-and-Bound (LBB) algorithm. These techniques have been
used to improve an agricultural system model developed by Dexcel
Limited, New Zealand, which describes a detailed representation of
pastoral dairying scenarios and contains an 8-dimensional parameter
space. The model incorporates the sub-models of pasture growth and
animal metabolism, which are themselves complex in many cases.
Each evaluation of the objective function, a composite 'Farm
Performance Index (FPI)', requires simulation of at least a one-year
period of farm operation with a daily time-step, and is therefore
computationally expensive. The problem of visualization of the
objective function (response surface) in high-dimensional spaces is
also considered in the context of the farm optimization problem.
Adaptations of the sammon mapping and parallel coordinates
visualization are described which help visualize some important
properties of the model-s output topography. From this study, it is
found that GA requires fewer function evaluations in optimization
than the LBB algorithm.
	%P 405 - 411