【正文】 not have these options, which means the certain attribute must be selected. The last three digits are the case identification number, which means the 13 digits of the case feature are the same, and the number of these three digits is used for distinguishing them. The system also rules: “ 000” is a prototype case, which is used for retrieval, 4 and other cases are “ 001” ,“ 002” ,…, which are used for reference cases to be searched by designers. If occasionally one of them needs to be changed as the prototype case, first it must be required to apply to change the one to “ 000” , and the former is changed to referential case automatically. The construction of the case index code is shown in . Case library The case library consists of lots of predefined cases. Case representation is one of the most important issues in case based reasoning. So pounding with the index code,. Hierarchical form of Case The structure similarity of the fixture is represented as the whole fixture similarity, ponents similarity and ponent similarity. So the whole fixture case library, ponents case library, ponent case library of fixture are formed correspondingly. Usually design information of the whole fixture is posed of workpiece information and workpiece procedure information, which represent the fixture satisfying the specifically designing function demand. The whole fixture case is made up of function ponents, which are described by the function 5 ponents’ names and numbers. The ponents case represents the members. (function ponent and other structure ponents， main driven parameter, the number, and their constrain relations.) The ponent case (the lowest layer of the fixture) is the structure of function ponent and other ponents. In the modern fixture design there are lots of parametric standard parts and mon non standard parts. So the ponent case library should record the specification parameter and the way in which it keeps them. 2 Strategy of Case Retrieval In the case based design of fixtures ,the most important thing is the retrieval of the similarity, which can help to obtain the most similar case, and to cut down the time of adaptation. According to the requirement of fixture design, the strategy of case retrieval bines the way of the nearest neighbor and knowledge guided. That is, first search on depth, then on breadth。 the knowledge guided strategy means to search on the knowledge rule from root to the object, which is firstly searched by the fixture type, then by the shape of the workpiece, thirdly by the locating method. For example, if the case index code includes the milling fixture of fixture type, the search is just for all milling fixtures, then for box of workpiece shape, the third for 1plane+ 2pine of locating method. If there is no match of it, then the search stops on depth, and returns to the upper layer, and retrieves all the relative cases on breadth. Retrieval algorithms: 1)According to the case index information of fixture case library, search the relevant case library。 6 2)Match the case index code with the code of each case of the case library, and calculate the value of the similarity measure。 3)Sort the order of similarity measure, the biggest value, which is the most analogical case. Similarity between two cases is based on the similarity between the two cases. features. The calculation of similarity measure depends on the type of the feature. The value of similarity can be calculated for numerical values, for example, pareWorkpiece with the weight of 50kg and 20kg. The value can also be calculated between non numerical values, for example, now the first 13 digits index code is all non numerical values. The similarity measure of a fixture is calculated as follows: where S is the similarity measure of current fixture, n is the number of the index feature, is the weight of each feature, is the similarity measure of the attribute of the i2th feature with the attribute of relative feature of the jth case in the case library. At the same time, , the value counts as follows: 7 . Where is the value of the index attribute of the ith feature, and is the value of attribute of the relative ith feature of the jth case in case library. So there are two methods to select