Geometry-Based Index for Predicting Sink Mark in Plastic Parts

This paper describes the development of geometry-based indices that predict sink mark depth in injection molded parts. Plastic part designers need such indices to incorporate manufacturability concerns at the conceptual stage of design. These indices apply to several form features so engineers do not have to check different design rules for each geometry element. First, we propose a geometry-based sink index that can be used to predict sink mark depth as a function of process conditions such as packing pressure. Next, we explain how this relationship is identified through experiments. We also describe HyperDesign/Plastics, a Macintosh-based design aid that incorporates the sink index.

Multivariate Yield Maximization Using CAD/CAE Models: Efficient Approximations Based on Mean and Variance

A product contributes to yield if all of its performance functions fall within their upper and/or lower limits. For example, a piston connecting rod may be required to provide rigidity along several axes. The actual connecting rod deflection will vary, depending on variations in the materials and forging conditions, but the deflection must remain less than an upper limit. Designing for maximum yield for multivariate performance limits is a difficult task. Direct optimization may require excessive computing resources. We discuss two efficient methods for yield improvement: ‘performance centering’ and a method based on Taguchi’s ‘parameter design’ philosophy. Both are shown to be motivated by the Chebychev inequality. It is important to remember that these are approximate methods. An example shows that they may produce sub-optimal yield, even when the random components of the performance functions are independent and identically distributed.

Capturing and Using Designer Intent in a Design-With-Features System

The goal of “intelligent” computer-aided-design (CAD) systems is to provide greater support for the process of design, as distinguished from drafting and analysis. More supportive design systems should provide a quick and simple means of creating and modifying design configurations, automating evaluation procedures (e.g., for manufacturing), and automating interfaces to analysis procedures. In this paper we are concerned with the issues of representing in-progress designs so that such goals can be met. A feature-based representation is proposed in which features are defined as possessing not only form but also certain designer intentions regarding geometric relationships. A working experimental version of a design-with-features system using this representation for thin-walled components illustrates its use in composing a design as a configuration of feature-forms, in modifying the design geometry through automatic, intelligent incorporation and propagation of designer-initiated geometry changes, and in providing for the generation of user-defined features. In contrast to constraint-driven simultaneous equation solving methods, this system uses an intent-driven knowledge-based method to propagate and incorporate geometry modifications not only in fully-constrained designs, but also in over- and under-constrained designs. Issues of manageability, extensibility, and computationally efficiency were considered in the development of the core services of the system.

Formal Solution of N-Type Taguchi Parameter Design Problems With Stochastic Noise Factors

The Taguchi method of product design is a statistical experimental technique aimed at reducing the variance of a product performance characteristic due to uncontrollable factors. The goal of this paper is to provide a monotonicity analysis based methodology to facilitate the solution of N-type parameter design problems. The obtained design is robust, i.e., the least sensitive to variations on uncontrollable factors (noise). The performance characteristic is unbiased in the sense that its expected value equals a target or specification. The proposed loss function is based on the absolute deviation of the characteristic with respect to the target, instead of the common square error approach. Conditions, like those imposed by monotonicity analysis, on the monotonic characteristics of the performance function are proven, despite the objective function is not monotonic and contains stochastic parameters. These conditions allow the qualitative analysis of the problem to identify the activity of some constraints. Identification of active sets of constraints allows a problem reduction strategy to be employed, where the solution to the original problem is obtained by solving a set of problems with fewer degrees of freedom. Results for the case of one uncontrollable factor are independent of the probability measure on the factor. However, conclusions for the multi-parametric case must take into account the characteristics of the probability space on which the random parameters are defined.

A Manufacturability Evaluation and Improvement System

The increasing application of CAE has lead to the evolution of Concurrent Engineering — a philosophy that prescribes simultaneous consideration of the life-cycle design issues of a product. The Concurrent Engineering (CE) systems that have been developed so far have relied on knowledge bases and qualitative evaluations of a part’s manufacturability for feedback to the design engineer. This paper describes a method for developing quantitative indicators of manufacturability. Feature-based design and estimation of machining parameters are used for ascertaining a part’s manufacturing requirements. These requirements are then combined into indices which lead the designer to features that must be redesigned for improved manufacturability. This method is illustrated on a system for rotational machined parts: the Manufacturability Evaluation and Improvement System (MEIS).

In Pursuit of a Design Mathematics: Generalizing the Labeled Interval Calculus

The Labeled Interval Calculus (LIC) is a formalism for reasoning about sets of design possibilities. Examples include toleranced objects, abstract descriptions involving many possible instantiations, and varying operating conditions. It has been successful in a “mechanical design compiler”, which accepts schematics and specifications and returns catalog numbers for optimal implementations. The LIC at present operates on monotonic algebraic equations and intervals of real values, but it now appears possible to generalize it to address arbitrary types of mathematical sets and relationships. The resulting family of formalisms is expected to be useful in design by feature and other design programs.

Extensions to the Taguchi Method of Product Design

The Taguchi method of product design is an experimental approximation to minimizing the expected value of target variance for certain classes of problems. Taguchi’s method is extended to designs which involve variables each of which has a range of values all of which must be satisfied (necessity), and designs which involve variables each of which has a range of values any of which might be used (possibility). Tuning parameters, as a part of the design process, are also introduced into Taguchi’s method. The method is also extended to solve design problems with constraints, invoking the methods of constrained optimization. Finally, the Taguchi method uses a factorial method to search the design space, with a confined definition of an optimal solution. This is compared with other methods of searching the design space and their definition of an optimal solution.

Making Sense of Prototyping Technologies for Product Design

Firms that design mechanical and electro-mechanical products confront a variety of difficult issues in their prototyping activities. For a given part, how can a choice among fabrication technologies be made? Where should investments in new prototyping technology be focused? How can new and existing prototyping technologies be evaluated? Our primary goal has been to develop a systematic method of evaluating prototyping processes in order to determine the best process for a given situation. This paper reports on a field study conducted at the Kodak Apparatus Division. Our data is drawn from (1) a user survey of prototyping perceptions and needs, (2) a survey to determine the importance of various prototype part performance attributes, and (3) estimates of the fabrication time, cost, and part performance for 104 parts and four prototyping processes.

Evaluation of Group Design in Engineering

In group design simultaneous input from multiple individuals, communications between group members, and management of the group effort make the process difficult to study. This paper describes the preliminary results of a computer-based system which was developed to study group design in engineering. Three groups of engineering designers were studied using the computer based group design system (GDS). Observations were made regarding the impact of management style (methodology) on the design process, the impact of communications on the overall group activity, the sequence of transactions leading to the conceptual design decision, and the involvement of group members. The use of systematic procedures appears to encourage more equal participation, a decision less influenced by individual choice, and reduced authority hierarchy.

Optimization of Design Utility

Design optimization is often carried out with respect to a single objective, for example, weight or manufacturing cost. Weight, cost, quality and robustness are each important considerations in design evaluation. However, when the design process is completed, a design is evaluated with respect to its performance in all of these areas, and possibly others, in addition to mechanical design considerations. The process of design should be driven from the very beginning by consideration of how the design artifact will ultimately be evaluated. We present a method for design analysis whose basic premise is that design analysis should be based on a rigorously determined multiple attribute design evaluation function. The evaluation function provides insight necessary to formulate a non-linear programming problem. A clear distinction is made between design attributes and design constraints. The objective function is maximization of utility. Explicit representations of relationships between decisions which designers make and resulting design performance in each of several attributes serve as constraints. An illustrative example of automotive bumper beam design optimization is presented. By coupling the physical relationships between the design decision variable beam gauge and the attributes weight, deflection and cost, maximization of the worth of the design in terms of the best combination of these attributes is possible. The solution can be expressed in terms of elements of a design vector over which the designer has direct control.