Backward Mapping Methodology for Design Synthesis

Design efficiency and robustness at early stage of parametric engineering design play a critical role in reducing cycle time and improving product quality in the overall product development process. Usually, the “forward mapping” approach, is used to find designs, where the desirable performances are satisfied through large iterations of analysis and evaluation from design space to performance space. However, these approaches are time-consuming and involve blind search if the engineering system simulation models and/or initial conditions are not appropriately selected. On the other hand, common “reverse engineering” methods use domain-specific assumptions and are not effective in generic scenarios where the presumed conditions are violated. In this paper, a Backward Mapping Methodology for Design Synthesis (BMDS) is presented that can help conduct design synthesis rapidly and robustly at early stage of parametric engineering design. BMDS is a surrogate model-based approach that combines the strengths of metamodeling and statistics. It can help designers explicitly identify the robust design solutions that satisfy the designer-specified performance requirements through a “backward mapping” from the performance space directly to the design space. Preliminary case studies show its effectiveness and potential to be used as a generic early stage parametric design synthesis methodology in the future.

Supporting Product Introduction Processes Through Product Structures

This paper reports research on approaches for accessing and controlling product data within the Product Introduction Process (PIP) through the specification, prototyping, and integration of a set of product structures, product viewpoints and their relationships. A significant achievement has been the validation of a collection of computer-based models of product structures that enable the capture of work breakdown, product specification, product functionality, and the physical parts that comprise the product. Furthermore, a number of relationships between these product structures and the business processes that use and create them have been identified. The work presented in this paper demonstrates the importance of representing relationships between product introduction processes and products through the adoption of product structures. The potential for exploiting product structures to improve understanding of relationships between products and processes is illustrated. A case study captures flows of related functional and physical product structures within PIPs across a two-tier supply chain, and discusses the issues involved in supporting such related product and process information.

A Performance-Based Representation for Engineering Design

A performance-based representation is presented, which uses the Performance Orientation Chart (POC) to aid the designer throughout an interactive design process. Assuming that all performance attributes can be expressed as functions of the design parameters, three types of graphical matrix are shown in the POC: 1) The design form depicts the performance attributes varying with the correspondent design parameters; 2) The performance dependency addresses the trade-off information among the multiple specifications based on Pareto optimal solutions; 3) The parameter constraint space defines the feasible region of the design, parameters within the, active specification limits. Guided by these graphical matrices, the designer can interactively develop the design solution to satisfy multiple specifications. The methodology was applied to a practical design problem to explicate how the POC can help the designer acquire a satisfying design solution with extensive confidence. Finally, the discussion, indicates that the performance-based representation is significantly compatible with other current engineering design methodologies.

Identifying Product Portfolio Architecture Modularity Using Function and Variety Heuristics

This paper proposes a method for identifying product portfolio architecture alternatives based upon customer needs and product function. Customer needs and uses are interpreted according to the variation in performance target values across the market and within the set of individual customer uses. Product uses are represented by function structures consisting of the functions necessary for achieving the use. These individual product use function structures are combined into a monolithic function structure to represent the entire product portfolio. This monolith is then partitioned according to function and product variety heuristics into function clusters that anticipate product modules. This candidate modularity can then be used to deliver product variety across the product portfolio given functional constraints. A portfolio of xerographic products is used as the working example.

Genetic Algorithms in Fuzzy Engineering Design

Problems in the field of engineering design represent an important class of real world problems that typically require a fuzzy and imprecise representation. This article presents and discusses a new approach to model this type of problem, by incorporating linguistic descriptions together with a variety of user-defined trade-off strategies. An interactive computer application is introduced, using stochastic optimization to solve the design task by producing a specially desired output under the given environmental conditions which are partly caused by the personal preferences of the engineer and by the expectations of the customer. It utilizes a randomized evolutionary technique, made suitable for the class of problems at hand, to generate and to optimize design solutions that are later identified by a clustering algorithm. Moreover test problems that were solved by the application are considered. In all cases the good solutions were obtained by evaluating only an extremely small fraction of all possible designs.

Introducing Knowledge-Based Engineering Into an Interconnected Product Development Process

Recently, a large automobile company implemented a Knowledge-based Engineering (KBE) application to help design an engine component. While the KBE developers aimed to facilitate a single engineer’s ability to design this component using only the KBE application, it can be shown that in fact this component’s design is tightly coupled to that of several others. Can KBE handle situations like this? How common are they? To address these and other questions, Design Structure Matrix (DSM) models were made of this component at three levels: system interactions, assembly of the component, and individual parts. The size, row names, and internal entries of these matrices were compared to matrices constructed from several conventional written design guides and a flowchart of the KBE application. In each case, the DSM contained more rows or more matrix entries per row, especially at the system interaction level. Since the DSMs were constructed by interviewing experienced engineers, one implication is that while low-aggregation information may be documented, system level information at this company mostly resides in people’s heads. An informal measure of “knowledge content” based on the number of matrix entries per row was shown to be consistent with similar measurements made on DSMs obtained by several other researchers. These results indicate some of the scope and complexity challenges that KBE faces.

Introducing a Use Perspective in Product Design Theory and Methodology

In this paper it is argued that methods are needed for the design of a larger variety of product aspects than is feasible with mechanical engineering design methodology of today. Design methods found within the European schools of design are inadequate for the design of products other than machine systems of transforming character. The reason for this is that the underlying theories only describe the nature of ‘operand-transforming’ technical systems, and that the description of the process and function systems are too narrowly defined to be useful for the design of ‘non-transforming’ products, or for products where the human is involved as an active user. The paper takes as the standpoint that the functional language, in accordance with established foundation in engineering design theory, is a successful means to treat usability aspects of human-product systems. An extended process modeling view based on product life-phase thinking including a ‘use-process’ is presented, focusing the attention towards the use, and not merely the workings, of the product. Also, extended definitions of a number of concepts are proposed, and function-classes of the human-product system, leading to a more generally applicable use of functions as a modeling tool when describing products, is introduced. The proposed functional language is illustrated in a product case example.

Toward a Reasoning Framework of Design As Synthesis

Most of the previous research efforts for design process modeling had such assumptions as “design as problem solving,” “design as decision making,” and “design by analysis,” and did not explicitly address “design as synthesis.” These views lack notion and understanding about synthesis. Compared with analysis, synthesis is less understood and clarified. This paper discusses our fundamental view on synthesis and approach toward a reasoning framework of design as synthesis. To do so, we observe the designer’s activity and formalize knowledge operations in design processes. From the observation, we propose a hypothetical reasoning framework of design based on multiple model-based reasoning. We discuss the implementation strategy for the framework.

Constrained Shape Scaling of Trimmed NURBS Surfaces

A method to scale and deform a trimmed NURBS surface while holding the shape and size of specific features (trimming curves) unchanged is presented. The new surface is formed by scaling the given surface according to the scaling requirement first; and then attaching the (original) features to the scaled NURBS surface at appropriate locations. The attaching process requires several geometric operations and constrained free-form surface deformation. The resulting surface has the same features as the original surface and same boundary curves as the scaled surface while reflecting the shape and curvature distribution of the scaled surface. This is achieved by minimizing a shape-preserving objective function which covers all the factors in the deformation process such as bending, stretching and spring effects. The resulting surface maintains a NURBS representation and, hence, is compatible with most of the current data-exchange standards. Test results on several car parts with trimming curves are included. The, quality of the resulting surfaces is examined using the highlight line model.

Educational Interactive Multimedia Software: Can it Effectively Teach Engineering Concepts?

This paper provides a rigorous statistical analysis of the educational effectiveness of a curriculum module pertaining to direct current (DC) motors and motorized systems. Educational interactive multimedia software played an important role in the module, which also included lectures, a design project and other more traditional pedagogics. Pre- and post-tests were administered to measure learning during the module and individual test questions targeted specific areas of the curriculum to determine the contribution of the various resources to the entire learning environment. The data indicate that the module was indeed educationally successful and that the software contributed to the curriculum’s success. The module was particularly effective for promoting learning by female engineering students in the course and mitigated some discrepancies in prior knowledge. Students were also grouped into four preference groups based on analysis of students’ preference questionnaire responses. The data indicate that each preference group responded differently to the various teaching techniques employed in the curriculum.