The IFBMDA: A Model for Mechanical Design Automation

The IFBMDA is an Information-Flow-Based model for Mechanical Design Automation. This paper first analyzes the mechanical design process from the views of design methodology and cognitive model. Then, two essential assumptions about mechanical design behavior are provided. Based on the analysis and fundamental assumptions, this paper thoroughly describes five submodels which constitute the automation model IFBMDA. They are Information Flow model, Knowledge Processing model, Non-monotonic Expansion Search model, Iterative Constraint Generation and Solution model and Design Process Stage model. Then, this paper also evaluates the model in both practical and theoretical aspects and shows that it is well-developed in both aspects. Finally, the perspective of further mechanical design automation research is outlined.

A General Method for Computing the Distance Between Two Moving Objects Using Optimization Techniques

Presented in this paper is a general method used to find the distance between two moving objects. This distance is defined as the length of the shortest path from one object to the other. The objects are assumed to be composed of arbitrary quadratic surface segments. The distance problem is formulated as a quadratic programming problem with linear and/or quadratic constraints, which is solved by efficient and robust quadratic programming techniques. Attention is focused on implementation in order to achieve computational efficiency for real-time applications. Computing tests show that the computational speed of this method is of linear order in terms of the total number of bounding surfaces of the two objects. It is also shown that, with a minor modification, this method can be used to calculate the interference between objects. A corresponding general software code has been implemented, and will be used for kinematics and dynamics modelling and simulation of space manipulators including situations with transient topologies, contact of environment, and capture/release of payloads.

A Blackboard Architecture to Support Case-Based Reasoning in Mechanical Design

One of the most significant issues in applying case-based reasoning (CBR) to mechanical design is to integrate previously unrelated design plans towards the solution of a new design problem. The total design solution (the design plan structure) can be composed of both retrieved and dynamically generated design plans. The retrieved design plans must be mapped to fit the new design context, and the entire design plan structure must be evaluated. An architecture utilizing opportunistic problem solving in a blackboard environment is used to map and evaluate the design plan structure effectively and successfuly. The architecture has several assets when integrated into a CBR environment. First, the maximum amount of information related to the design is generated before any of the mapping problems are addressed. Second, mapping is preformed as just another action toward the evaluation of the design plan. Lastly, the architecture supports the inclusion of memory elements from the knowledge base in the design plan structure. The architecture is implemented using the GBB system. The architecture is part of a newly developed CBR System called DEJAVU. The paper describes DEJAVU and the architecture. An example is also included to illustrate the use of DEJAVU to solve engineering design problems.

Optimization in a Design System for Complex Products

This paper describes a system for parametric design and optimization of complex products. In the system, the use of knowledge-based and mathematical programming methods is combined. The motivation is that while knowledge-based methods are well suited for modeling products, they are insufficient when dealing with design problems that can be given an optimization formulation. This weakness was approached by including the information necessary for stating an optimization problem in the product models. A system optimization method can then be applied. The system also performs sensitivity analysis and has an interactive optimization module. The use of the system is illustrated by an example; the design and optimization of a two-speed gearbox.

Discrete Optimal Design Formulations: Application to Gear Train Design

The use of discrete variables in optimal design models offers the opportunity to deal rigorously with an expanded variety of design situations, as opposed to using only continuous variables. However, complexity and solution difficulty increase dramatically and model formulation becomes very important. A particular problem arising from the design of a gear train employing four spur gear pairs is introduced and formulated in several different ways. An interesting aspect of the problem is its exhibition of three different types of discreteness. The problem could serve as a test for a variety of optimization or artificial intellegence techniques. The best known solution is included in this article, while its derivation is given in a sequel article.

Design of Machine Tool Spindle for Minimum Workpiece Cylindricity Error

The design of machine tool spindle for minimum cylindricity error is presented. The shape of the spindle portion between the bearings, usually cylindrical, is reconfigured to increase the spindle rigidity. Two methods of reconfiguration are presented. In the first method, the spindle is reshaped as a smooth body of revolution. In the second method, the spindle is reshaped as a stepped shaft. Effects of bearings rigidity and fatigue stresses are also considered.

Design for Assembly Evaluation of Orientation Difficulty Features

In support of the effort to bring downstream issues to the attention of the designer as parts take shape, an analysis system is being built to extract certain features relevant to the assembly process, such as the dimension, shape, and symmetry of an object. These features can be applied to a model during the downstream process to evaluate handling and assemblability. In this paper, we will focus on the acquisition phase of the assembly process and employ a Design for Assembly (DFA) evaluation to quantify factors in this process. The capabilities of a non-homogeneous, non-manifold boundary representation geometric modeling system are used with an Index of Difficulty (ID) that represents the dexterity and time required to assemble a product. A series of algorithms based on the high-level abstractions of loop and link are developed to extract features that are difficult to orient, which is one of the DFA criteria. Examples for testing the robustness of the algorithms are given. Problems related to nearly symmetric outlines are also discussed.

The Development of Virtual Concurrent Engineering and its Application to Stamped Products

It is well-known that so-called Concurrent Engineering is a desirable alternative to the largely sequential methods which tend to dominate most product development methods. However, the proper implementation of a concurrent engineering method is still relatively rare. In order to facilitate the development of a reliable concurrent engineering product development method, we start with a careful definition of concurrent engineering and, after an extensive study of all of product development, we propose three criteria which ideal concurrent engineering must satisfy. However, for labor, time, and overall cost considerations, ideal concurrent engineering is infeasible. Instead, we propose a computer-based environment which, by being constructed in accordance with the three criteria, attempts to approach ideal concurrent engineering. The result is the Virtual Concurrent Engineering method and computer implementation environment. This product development method and computer-based implementation system provide the detailed, structured information and data needed to optimally balance the product with respect to the main product development parameters (e.g., manufacturing costs, assembly, reliability). This important information includes re-design suggestions to improve the existing design. The designer can directly apply these re-design suggestions for design optimization, or he can use the results as input into a more complex design optimization or design parameterization function of his own. To demonstrate Virtual Concurrent Engineering, we use it to refine earlier work done by the authors in the Design for Producibility of stamped products. We discuss, in some detail, the results of applying Design for Producibility to complex stampings, including process plans and product producibility computations.

Catalog Design: Design Using Available Assets

Catalog design is a procedure in which a system is assembled by selecting standard components from catalogs of available components. Selection in design involves making a choice among a number of alternatives taking into account several attributes. The information available to a designer to do so during the early stages of project initiation may be uncertain. The uncertainty in information may be imprecise or stochastic. Under these circumstances, a designer has to balance limited resources against the quality of solution obtained or decisions made by accounting for uncertainty in information available. This complex task becomes formidable when dealing with coupled selection problems, that is problems that should be solved simultaneously. Coupled selection problems share a number of coupling attributes among them. In an earlier paper we have shown how selection problems, both coupled and uncoupled can be reformulated as a single compromise Decision Support Problem (DSP) using a deterministic model. In this paper, we show how the traditional compromise DSP can be extended to represent a nondeterministic case. We use fuzzy set theory to model imprecision and Bayesian statistics to model stochastic information. Formulations that can be solved with the same solution scheme are presented to handle both fuzzy and stochastic information in the standard framework of a compromise DSP. The approaches are illustrated by an example involving the coupled selection of a heat exchanger concept and a cooling fluid for a specific application. The emphasis in this paper is placed on explaining the methods.

On Global Feasible Search for Global Design Optimization With Application to Generalized Polynomial Models

A powerful idea for deterministic global optimization is the use of global feasible search, namely, algorithms that guarantee finding feasible solutions of nonconvex problems or prove that none exists. In this article, a set of conditions for global feasible search algorithms is established. The utility of these conditions is demonstrated on two algorithms that solve special problem classes globally. Also, a new model transformation is shown to convert a generalized polynomial problem into one of the special classes above. A flywheel design example illustrates the approach. A sequel article provides further computational details and design examples.