A Genetic Algorithm for Developing Modular Product Architectures

The architecture of a product is determined by both the elements that compose the product and the way in which they interact with each other. In this paper, we use the design structure matrix (DSM) as a tool to capture this architecture. Designing modular products can result in many benefits to both consumers and manufacturers. The development of modular products requires the identification of highly interactive groups of elements and arranging (i.e. clustering) them into modules. However, no rigorous DSM clustering technique can be found in product development literature. This paper presets a review of the basic DSM building blocks used in the identification of product modules. The DSM representation and building blocks are used to develop a new DSM clustering tool based on a genetic algorithm (GA) and the minimum description length (MDL) principle. The new tool is capable of partitioning the product architecture into an “optimal” set of modules or sub-systems. We demonstrate this new clustering method using an example of a complex product architecture for an industrial gas turbine.

Incorporating Design Complexity Measures in Architectural Assessment

A feature of good modularity is the ease of changing a module within a product. Existing modularity methods use subjective or qualitative attributes to evaluate architectures. We develop a method to relatively compare proposed product architectures according to design complexity. Our metric represents the difficulty that different module boundary interactions, represented by flows in and out of a function, would have in terms of redesign effort. We decomposed medical injector head systems and conducted interviews in two companies to find out a relative redesign effort for various interaction types, e.g. electrical and mechanical connection, signal flows, etc. We found that to change a flow by 1%, 1–4% more design effort is required, depending on the interaction type. We also found that decreasing a flow value causes, in general, less rework than increasing a flow. Our metric proved to be a valuable tool in estimating the redesign difficulty of an architecture.

Standardization and Modularization Driven by Minimizing Overall Process Effort

Faster Product development is a major goal for companies in competitive markets. Product platform architectures support planning for addressing diverse markets and fulfilling future market desires. Applying standardization or modularization on product platform components leverages current product design effort across future products, reducing overall development costs. This work introduces a method for focusing engineering effort when applying standardization or modularization on product platform components. The method calculates the total design effort from current to future generations of the platform, as obtained by standardization or modularization of components. By comparing the total design cost of different simulations, we can direct the design team to standardization or modularization opportunities. This process has been successfully applied to two different product platforms. One is External-Drum Plate-setter for the digital prepress printing market (introduced here) and the other is TOW anti-tank missile launching system for the military market.

Partitioning Knowledge for Generative Configuration Design

Designers and design researchers both agree that developing many feasible alternatives at the conceptual design stage is useful. In this paper we introduce generative configuration design (GCD) for conceptual design. We provide a partition of knowledge accessed during GCD and use the partitioned knowledge foundation to compare design tool architectures so that computational improvements can be made. We present an improved architecture for a GCD algorithm and implement it as a tool for office chair design. Subsequent examples show tradeoffs between computational load and design variety when applying constraints for behavior testing.

Work Structure Based Collaborative Engineering Design

Collaborative engineering design requires multiple people working together to achieve a common goal. Data sharing approach and workflow management approach have been developed to support collaborative design, but the disconnection of these two approaches has led to problems of efficiency and adaptability. In this paper, we propose a work structure based approach for collaborative design. Our goal is to improve process efficiency and adaptability by integrating management processes with engineering details and allowing designers to make certain managerial decisions through peer coordination. For a specific task, a work structure is a network of engineering work items connected by dynamically acquired engineering dependencies. It is used to generate multiple processes from which the one that best fits the current situation is dynamically determined through coordination among team participants. In order to capture engineering dependencies and associate engineering details, an adaptive work process model is developed that explicitly represents engineering work, work structure, and processes. Based on this model, a set of operations and algorithms are developed for intelligent agents to provide coordination support. Experiments have shown that by following this approach, engineering design processes can dynamically adapt to both requirement and resource changes, and the process efficiency can be significantly improved.

On Validating Design Decision Methodologies

In this paper an argument for validation of design decision methods is presented. In the process of justifying the need for validation, the elements of what a valid design methodology is are derived and a formal definition is presented. Under this definition, critical evaluation of two popular decision support methods, the House of Quality and Suh’s Axiomatic Design, is presented using a simple design problem and both are shown to be flawed under the proposed definition of validity. This does not imply that the methods are ineffective. In fact, under the appropriate assumptions, they are quite useful. However, in this paper, an investigation of the validity of these assumptions is conducted, including a more general definition of validity with respect to decision support methods in design.

Representing Product Architecture

Product architecture is the transformation of function to layout. Like much of conceptual design, it is a highly dynamic process whereby engineers must consider a deluge of information in terms of both function and form. One shortcoming of current engineering practice is the absence of representations or abstractions used to aid in developing, refining, and exploiting alternative layout solutions. The purpose of this paper is to present a representation for product architecture that sufficiently captures the design factors relevant to product architecture design which are not taken into account in current practices. An example is given to illustrate the technique, and results of a validation experiment are shown.

Enumerating the Component Space: First Steps Toward a Design Naming Convention for Mechanical Parts

A standard naming convention for mechanical parts is proposed in this paper. We refer to this naming convention as the component basis. The component basis is a first step at classifying an exhaustive list of human-made mechanical transmission artifacts as functional forms, geometric shapes, simple machines, and natural forms. The proposed component basis provides a framework for the development of a suite of computational conceptual design tools. This suite of design tools includes a function-based computational concept generator and a product evolution methodology.

Decision Matrix Reduction in Preliminary Design

Matrix methods such as the house of quality (HOQ) are useful as a first step in organizing the information relating product attributes to engineering design parameters. However, they frequently result in a large number of elements, making it very difficult to focus attention on elements that bring maximum benefits. This paper presents a method for reducing a large set of engineering parameters to a more efficient subset. This subset offers the greatest potential for cost control, and is where product development, experimental, and optimization efforts should be focused. An example of a new air pollution control system illustrates the method. The most efficient set of engineering parameters are identified, and are then defined as decision variables in a cost minimization model. The optimal design identified by the method significantly reduces cost.

Iteration in Engineering Design: Inherent and Unavoidable or Product of Choices Made?

Iteration in design has different meanings, ranging from simple task repetition to heuristic reasoning processes. Determining the need to iterate is important to improve the design process on cost, time, and quality, but currently there is no categorization of iterations conducive to this goal. After exploring the possible causes and attempts to address them, we propose to classify iterations as rework, design, or behavioral. This framework suggests that design teams should try to eliminate rework iterations, perform design iterations without skipping abstraction levels, and do behavioral iterations in parallel.