Purpose The purpose of this paper is to carry out an assembly tolerance analysis by means of a combined Jacobian model and skin model shape. The former is based on small displacements modeling of points using 6 × 6 transformation matrices of open kinematic chains in robotics. The latter easily models toleranced features with all kinds of geometric deviations. Design/methodology/approach This paper presents the procedure of performing tolerance analysis by means of the Jacobian model and skin model shape for assemblies. The point cloud-based discrete representative is able to model the actual toleranced surfaces instead of the ideal or associated ones in an assembly, which brings the simulation tools closer to reality. Findings The proposed method has the advantage of skin model shape which is suitable for geometric tolerances management along the product life cycle and contact analysis of kinematic small variations, as well as, with the Jacobian, enabling transformation of locally expressed parts deviations to globally expressed functional requirements. The result of the case study shows the accuracy of the method. Research limitations/implications The proposed approach has not been developed fully; other functional features such as the pyramid are still ongoing challenges. Practical implications It is an effective method for supporting design, manufacturing and inspection by providing a quantitative analysis of the effects of multi-tolerances on the final functional key characteristics and for predicting the quality level. Originality/value The paper is original in taking advantages of both Jacobian model and skin model shape to consider all geometric tolerances in assembly.

Tolerance specification involves selecting tolerance types for functional or assembly features to control the variation of features. General methods tend to formulate a frame to specify all the features of part, while the specification methods or reasoning rules for specific feature (point, line, plane, cylinder, etc.) are less studied. This paper focuses on the tolerance-type selection of the plane feature. The theory of axiomatic design is introduced to select the tolerance type for the plane feature, and the problem is interpreted as a redundant decoupled design. To achieve the functional requirements, design parameters and constraints of physics domain are determined. The mapping rules, which are between design parameters and functional requirements, are generated based on the independent axiom. Considering the large number of solutions of the design, the constraints such as cost and inspection methods are introduced to reduce the number of solutions. The minimum information axiom is introduced for the optimum mapping rules and the tolerance types are selected by the optimum mapping rules for the plane feature. Finally, the specification process is concluded and demonstrated by means of an example.

In the mechanical product design, tolerance specification is a crucial process to select tolerance types to control the geometrical variation of the features on a workpiece. Tolerance specification is a kind of decision-making process which needs to consider many influential factors of the manufacturing cost and quality requirement of a workpiece. In this context, those factors are classified into two categories, that of static factors from the rules and standards, and that of dynamic factors based on the manufacturing site's information. This paper presents a novel method for computer-aided tolerance specification to evaluate both types of factors while the traditional methods only take the static factors into considerations. Those factors are assessed by the developed MADM (Multiple Attributes Decision-Making) algorithm, assisted with the rule-based algorithm and Axiomatic Design algorithm. A case study is undertaken, and its result shows that the effect of dynamic factors on the output of tolerance specification. The results adhere to the ISO standard.

A datum selection strategy based on statistical learning is proposed. The datum selection is an important part of tolerance specification which is the base of geometric tolerance selection and tolerance principle selection. The problem of datum selection is to deduce the datum reference frame (DRF) based on geometrical, contact, and positioning characteristics. Currently, heuristic rules are used for DRF selection, leading to suboptimal choice of DRF in many cases. The proposed strategy formulates normalized vectors computed from the geometric, contact, and positioning characteristics of surfaces. The surfaces of different parts can be compared by their normalized vectors. Then the statistical learning method is used for building a classifier which can discriminate datum feature vectors based on training samples. Finally, a case study is given to verify the strategy and the different algorithms are compared and discussed.

No abstract