DOC PREVIEW
ESTIMATING THE OPTIMAL NUMBER OF ALTERNATIVES TO BE EXPLORED IN LARGE DESIGN SPACES

This preview shows page 1-2-3-4 out of 12 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 12 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 12 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 12 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 12 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 12 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

Expected UtilityThis document contains the draft version of the following paper: S.K. Gupta and C. Xu. Estimating the optimal number of alternatives to be explored in large design spaces: A step towards incorporating decision making cost in design decision models. ASME Computers in Engineering Conference, Montreal, Canada, September 2002. Readers are encouraged to get the official version from the conference proceedings or by contacting Dr. S.K. Gupta ([email protected]).ESTIMATING THE OPTIMAL NUMBER OF ALTERNATIVES TO BE EXPLORED IN LARGE DESIGN SPACES: A STEP TOWARDS INCORPORATING DECISION MAKING COST IN DESIGN DECISION MODELS Satyandra K. Gupta Department of Mechanical Engineering and Institute for Systems Research University of Maryland College Park, MD 20742 [email protected] Xu Department of Mechanical Engineering University of Maryland College Park, MD 20742 [email protected] ABSTRACT Exploration of design spaces is an important step in decision-based design. In consumer product development, precise design specifications are not known at the beginning of the design process. It is usually design team’s responsibility to find out the specifications as a part of the design process. This results in large design spaces in consumer product development. Furthermore, market window is usually limited. Thus, it is impractical to examine all possible design alternatives. As part of the design process, design teams need to determine how many alternatives to examine and how much evaluation time should be devoted to examining each alternative. This paper presents a model for estimating the optimal number of alternatives to be explored and the optimal evaluation time for each alternative by incorporating cost of decision-making in the overall design decision model. We also describe a design case study and investigate how characteristics of design task parameters influence the optimal number of alternatives and the optimal evaluation time. Our results indicate that it is difficult to intuitively identify the optimal values of the number of alternatives and the evaluation time for even very simple design tasks. We describe the practical issues that need to be addressed to make these decisions and discuss how the model proposed in this paper can be extended to handle more general cases of design tasks. 1 INTRODUCTION Development of consumer products is an activity through which a product development organization attempts to design and manufacture products to realize profit. This activity usually consists of identifying customer needs and preferences, defining product functionality, identifying design alternatives, identifying alternative ways of manufacturing/procuring components, and selecting a design alternative. Usually the goal of most consumer product development organizations is to maximize profit or a closely related financial metric such as net present value or return on investment. Product development for consumer products typically has the following characteristics: • Design spaces are quite large. At the beginning of consumer product development, engineering specifications are not available explicitly. They need to be determined as a part of the design process. Let us consider the following scenario to understand this characteristic. Let us assume that a company is interested in developing a frying pan to be used in kitchen. The company can design a utensil with a very long service life that uses very durable but expensive coating. Alternatively, they can design a pan with a short service life and therefore use less durable and less expensive coating. Market exists for both of these types of pans. Which alternative should be selected by the company ultimately depends on which alternative is likely to result in more profit for the company. Unfortunately profits cannot be estimated without some idea about the cost incurred in manufacturing the product and the manufacturing cost cannot be estimated without knowing at least some design details. Therefore, product specifications need to be selected as a part of the design process. Quite often companies need to consider alternatives not only at the design concept level but also at the specification level. This makes design spaces very large in consumer product development. • Market window is usually small. Demand for a particular type of product exists only for a short period of time. Due to changes in customer tastes, economic climate, and advent of new technologies, a particular product usually has a small market window. As an example, consider the following case. Cassette players significantly reduced 1demand for record players. CD players significantly reduced demand for cassette players. MP3 players might reduce demand for CD players. Therefore a product needs to be conceived, designed, and marketed quickly for it to have demand and result in profit. • Evaluating an alternative from design space takes significant amount of time. Evaluating a design alternative is not an instantaneous process. Experienced designer might give a design alternative a rough and quick evaluation based on his experience of similar design. But getting more confidence in evaluation, quite often involves simulation and physical prototyping. So evaluating a design alternative is a time consuming process. • Uncertainty in evaluation reduces if more time is spent on evaluation. For example, a coarsely meshed elastic FEA evaluation may take only hours to carry out. On the other hand, a very finely meshed FEA that includes non-linear effects may take days to complete. Testing of a physical prototype may take weeks to complete. Inaccuracies in the evaluation and hence the uncertainty reduces as we go from the first type of evaluation to the third type of evaluation. • Evaluating more alternatives improves the possibility of identifying design with superior performance characteristics. Usually this characteristic is associated with diminishing returns as shown in Figure 1. Initially, the performance improves significantly as more alternatives are evaluated. However, the improvement in performance is not linearly proportional to the number of alternatives being evaluated. The extent of the diminishing returns varies from one design task to another. We use the term “degree of diminishing return on performance with respect to the number of alternatives” to quantitatively characterize this extent.


ESTIMATING THE OPTIMAL NUMBER OF ALTERNATIVES TO BE EXPLORED IN LARGE DESIGN SPACES

Download ESTIMATING THE OPTIMAL NUMBER OF ALTERNATIVES TO BE EXPLORED IN LARGE DESIGN SPACES
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view ESTIMATING THE OPTIMAL NUMBER OF ALTERNATIVES TO BE EXPLORED IN LARGE DESIGN SPACES and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view ESTIMATING THE OPTIMAL NUMBER OF ALTERNATIVES TO BE EXPLORED IN LARGE DESIGN SPACES 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?