Study of Snow Disaster Loss Evaluation Model Based on Projection Pursuit

It is a complex problem of multi-attribute and multi-factor to evaluate snow disaster loss. In order to resolve the non-uniformity problem of evaluation results of disaster loss indices, and to raise the evaluation result precision of snow disaster loss, this paper proposes a snow disaster loss evaluation model based on projection pursuit (PP). Accordingly, the paper converts high dimensional indices data to projection value of single dimension by PP which can find optimal projection direction using accelerating genetic algorithm (AGA). A scheme of PP modeling is also presented to reduce the computational amount, and a new function of projection indices is given. It is suggested that both the function of projection indices and the parameters of PP model can be optimized by using a real coding based accelerating genetic algorithm. Furthermore, the case study shows that the model has the advantage of simplicity and objectiveness of the evaluation result.