Multi-objective intelligent coordinating optimization blending system based on qualitative and quantitative synthetic model

A multi-objective intelligent coordinating optimization strategy based on qualitative and quantitative synthetic model for Pb-Zn sintering blending process was proposed to obtain optimal mixture ratio. The mechanism and neural network quantitative models for predicting compositions and rule models for expert reasoning were constructed based on statistical data and empirical knowledge. An expert reasoning method based on these models were proposed to solve blending optimization problem, including multi-objective optimization for the first blending process and area optimization for the second blending process, and to determine optimal mixture ratio which will meet the requirement of intelligent coordination. The results show that the qualified rates of agglomerate Pb, Zn and S compositions are increased by 7.1%, 6.5% and 6.9%, respectively, and the fluctuation of sintering permeability is reduced by 7.0%, which effectively stabilizes the agglomerate compositions and the permeability.

Analogy-based software effort estimation using multi-objective feature selection

The feature selection in analogy-based software effort estimation （ASEE） is formulized as a multi-objective optimization problem. One objective is designed to maximize the effort estimation accuracy and the other objective is designed to minimize the number of selected features. Based on these two potential conflict objectives, a novel wrapper- based feature selection method, multi-objective feature selection for analogy-based software effort estimation （MASE）, is proposed. In the empirical studies, 77 projects in Desharnais and 62 projects in Maxwell from the real world are selected as the evaluation objects and the proposed method MASE is compared with some baseline methods. Final results show that the proposed method can achieve better performance by selecting fewer features when considering MMRE （mean magnitude of relative error）, MdMRE （median magnitude of relative error）, PRED （ 0. 25 ）, and SA （ standardized accuracy） performance metrics.

Application of Multi－objective Optimization Method to Extracton Separtion Process

A multi-objective decision procedure was proposed to select the optimal precase parameters of extractionseporating zirconium and hafnium from sulphuric acid solution with Alamine-336 as an extractant in this pa-per.Adaptive Random Search(ARS) algorithm was adopted for optimizing calculation. The results obtainedshow that the multi-objective decision method is much better than sangle-objecive decision method.

COST AND SERVICE-LEVEL-BASED MODEL FOR A SERU PRODUCTION SYSTEM FORMATION PROBLEM WITH UNCERTAIN DEMAND

With increasing demand diversification and short product lifecycles, industries now encounter challenges of demand uncertainty. The Japanese seru production system has received increased attention owing to its high efficiency and flexibility. In this paper, the problem of seru production system formation under uncertain demand is researched. A multi-objective optimization model for a seru production system formation problem is developed to minimize the cost and maximize the service level of the system. The purpose of this paper is to formulate a robust production system that can respond efficiently to the stochastic demand. Sample average approximation （SAA） is used to approximate the expected objective of the stochastic programming. The non-dominated sorting genetic algorithm II （NSGA-II） is improved to solve the multi-objective optimization model. Numerical experiments are conducted to test the tradeoffbetween cost and service level, and how the performance of the seru production system varies with the number of product types, mean and deviation of product volume, and skill-level-based cost.