In order to obtain low-cost, high-performance products, in today's society of adoption new materials replacing traditional materials, the importance of materials selection has become increasingly prominent. Therefore, the establishment of engineering materials comprehensive evaluation system to realize materials selection scientific and reasonable has become a hot research field at home and abroad.This dissertation researched the establishment of the evaluation index system, quantitative analysis of evaluation indices, determination of weights for evaluation indices and the decision of evaluation methods, respectively. Comprehensive evaluation system for engineering materials was realized by programming through the development platform of Delphi 7 according to the researching mentioned above.On the base of introduction selection principles of evaluation indices and advanced idea of materials selection, evaluation index system of engineering materials was determined, belonging to tree structure, and established, adopting goal decomposition method, according to its character. Engineering materials evaluation index system of static and dynamic combination was constructed.On the base of introduction the method of quantification qualitative evaluation index, and normalization evaluation indices for engineering materials, referring to the value of a certain candidate evaluation index for engineering materials might identical or very close with that of others, and data information obtained from evaluation indices of engineering materials might overlap, the conception of distinction degree and non-correlation degree of evaluation indices were defined. The quantitative analysis methods of evaluation indices for engineering materials were proposed as the criterion of index quantitative selection in this paper based on distinction degree and accumulative non-correlation degree. Eight kinds of candidate evaluation indices of six kinds of low temperature storage container materials, stainless steel, and thirteen kinds of candidate evaluation indices of five kinds of stainless steel for hot moulds, were quantitative analyzed according to the criterion mentioned above, respectively. The results verify the reasonableness of quantitative analysis methods mentioned above.According to analytic hierarchy process (AHP) method, the subjective weights model of evaluation indices for engineering materials was constructed. The subjective weights of eight evaluation indices for low temperature materials were confirmed by using this model. Referring to the difference among the evaluation indices weights obtained from entropy method does not match with actual situation, the objective weights model of evaluation indices for engineering materials was proposed and constructed based on improved entropy method. The objective weights of nine evaluation indices for joint implant materials were determined by using this model. The intelligent weights model of evaluation indices for engineering materials was constructed based on back propagation network (BPN), and the intelligent weights of eight evaluation indices for coating materials were confirmed by using BPN. On the base of analysis the advantages and disadvantages of three models confirming weights above, the conception of weights confidence was presented, and the Combination Determining Weights Method (CDWM) which combines subjective, objective and intelligent weights of evaluation indices for engineering materials was constructed based on confidence, and the combination weights of eight evaluation indices for low temperature materials were confirmed by using CDWM. The result shows that the weights obtained from CDWM are more scientific and accurate.For the shortcomings of Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) and Principal Component Analysis (PCA), the improved methods were carried out respectively in this dissertation. Electric insulating materials, and low temperature materials were ranked by three improved methods mentioned above, respectively. The results show that the rank effects of candidate materials obtained by improved methods are more reasonable than that of traditional methods. At the same time, gas turbine blades, bearing materials and cutting tool materials were evaluated by Weighted Average, Improved Grey Relational Analysis (IGRA)and Fuzzy Assessment, respectively. On the base of analyzing the advantage and disadvantage of five evaluation methods above, aimed for the deficiency of single evaluation method, Combination Evaluation Model (CEM) was established by combining five kinds of multi-objective method mentioned above based on the Theory of Serial Number Summation and Mode Theory. The low temperature materials were evaluated using CEM with combing weights obtained from CDWM of evaluation indices. The result shows that the conclusion obtained from CEM is more reasonable and reliable because the CEM eliminates the deficiency of single evaluation method.Regard Delphi 7 as development platform, adopting OOP (Object Oriented Programming) method, using MDI (multi documents interface) form design style, a comprehensive evaluation system for engineering materials was developed with the establishment of the evaluation indices system, quantitative analysis of evaluation indices, determination the weights of evaluation indices and the decision of the evaluation methods.