Multi-objective decision making based on particle filter
Xiaoyu Zhang, Shiqiang Hu · 2010
Particle filter, which is proposed for implementing recursive Bayesian filter to calculate posterior probability density function, is applied to solve the incommensurability of multi-objective decision making problem here. This method, which is based on the principle of particle filter, can convert the values of all alternatives under every criterion into probability, once the expectable values of all criterions are given. Then the incommensurability among different criterions in multi-objective decision making can be eliminated. At last together with the weight of every criterion, weighted sum of every alternative can be made. Hence a sorting of all alternatives can be make out. The effectiveness of this method is shown by two examples.