我们发表在ACTA MATERIALIA上的反应加工多目标优化的论文（Yuxi Jia, Sheng Sun, Lili Liu, Yue Mu and Lijia An, Acta Mater., 2004, 52: 4153-4159），被K. Mitra在论文Genetic algorithms in polymeric material production, design, processing and other applications: a review, International Materials Reviews, 2008, 53: 275-297里详细评论如下：“Jia et al. solved a weight based multiobjective optimisation problem for producing silicone rubber based on given requirements. They derived the explicit expressions between composition and mechanical properties of silicone rubber from the physics of polymer elasticity and deduced the implicit expressions among material composition, reaction conditions and reaction efficiency from chemical thermodynamics and kinetics. The multi-objective problem was to find out optimal values for material composition and reaction conditions (average number of structure units in an initial molecule, an average number of cross-linkable structure units in an initial molecule, reaction medium temperature) by achieving simultaneous minimisation of the difference between the model predicted shear modulus and reaction time and their corresponding desired values”。直译为：“贾玉玺等人以硅橡胶制品的反应加工制备为例，按照给定的要求，求解了基于权重的多目标优化问题。他们根据聚合物弹性的物理学原理，获得了材料体系组分和力学性质之间的显式函数关系式；而且，他们还根据化学热力学和动力学理论，建立了材料体系组分、反应加工条件和反应效率之间的隐式表达式。他们求解的多目标优化问题是要同时得到材料体系组分和反应加工条件的最优解（在初始分子里结构单元的平均数量、在初始分子里可交联的结构单元的平均数量、反应加工介质温度）。他们采用了基于优化模型而预测的剪切模量、反应加工时间与相应的期望值之间差异的同时最小化方法，得到了上述多决策变量的最优解”。