Kwang-Jae Kim

Fuzzy multicriteria models for quality function deployment

Abstract An integrated formulation and solution approach to Quality Function Deployment (QFD) is presented. Various models are developed by defining the major model components (namely, system parameters, objectives, and constraints) in a crisp or fuzzy way using multiattribute value theory combined with fuzzy regression and fuzzy optimization theory. The proposed approach would allow a design team to reconcile tradeoffs among the various performance characteristics representing customer satisfaction as well as the inherent fuzziness in the system. In addition, the modeling approach presented makes it possible to assess separately the effects of possibility and flexibility inherent or permitted in the design process on the overall design. Knowledge of the impact of the possibility and flexibility on customer satisfaction can also serve as a guideline for acquiring additional information to reduce fuzziness in the system parameters as well as determine how much flexibility is warranted or possible to improve a design. The proposed modeling approach would be applicable to a wide spectrum of design problems where multiple design criteria and functional design relationships are interacting and/or conflicting in an uncertain, qualitative, and fuzzy way.

Determination of an optimal set of design requirements using house of quality

Abstract Quality Function Deployment (QFD) has been used to translate customer needs and wants into technical design requirements in order to increase customer satisfaction. QFD utilizes the house of quality (HOQ), which is a matrix providing a conceptual map for the design process, as a construct for understanding Customer Requirements (CRs) and establishing priorities of Design Requirements (DRs) to satisfy them. Some methodological issues occurring in the conventional HOQ are discussed, and then a new integrative decision model for selecting an optimal set of DRs is presented using a modified HOQ model. The modified HOQ prioritization procedure employs a multi-attribute decision method for assigning relationship ratings between CRs and DRs instead of a conventional relationship rating scale, such as 1–3–9. The proposed decision model has been applied to an indoor air quality improvement problem as an illustrative example.

Simultaneous optimization of mechanical properties of steel by maximizing exponential desirability functions

Summary. A modelling approach to optimize a multiresponse system is presented. The approach aims to identify the setting of the input variables to maximize the degree of overall satisfaction with respect to all the responses. An exponential desirability functional form is suggested to simplify the desirability function assessment process. The approach proposed does not require any assumptions regarding the form or degree of the estimated response models and is robust to the potential dependences between response variables. It also takes into consideration the difference in the predictive ability as well as relative priority among the response variables. Properties of the approach are revealed via two real examples -one classical example taken from the literature and another that the authors have encountered in the steel industry.

Evaluation of product usability: development and validation of usability dimensions and design elements based on empirical models

Abstract Usability defined in this study consists of the following two groups of dimensions: objective performance and subjective image/impression, which are considered equally important in designing and evaluating consumer electronic products. This study assumes that the degree of each usability dimension can be estimated by the design elements of the products. A total of 48 detailed usability dimensions were identified and defined in order to explain the usability concept applicable to the consumer electronic products. The user interface of the consumer electronic products was decomposed into specific design elements (defined as human interface elements: HIEs). A total of 88 HIEs were measured for 36 products by using a measurement checklist developed in this study. In addition, each usability dimension was evaluated by using the modified free modulus method. Multiple linear regression techniques were used to model the relationship between the usability and the design elements. As a result, 33 regression models were developed. The models are expected to help the designers not only identify important design variables but also predict the level of usability of a specific consumer electronic product. The approach used in this study is expected to provide an innovative and systematic framework for enhancing the usability of the consumer electronic products as well as other consumer products with minor modifications. Relevance to industry This study presents a systematic approach to enhancing the objective and subjective usability of consumer electronic products. It can be used in the design and evaluation stage of the development process in order to help the designers and developers identify important design elements, diagnose usability problems, and predict the level of usability of consumer electronic products. In addition, the approach developed in this study is also applicable to other consumer products (such as appliances, automobiles, communication devices, etc.) with minor modifications.

Dual Response Surface Optimization: A Fuzzy Modeling Approach

In modern quality engineering, dual response surface methodology is a powerful tool. In this paper, we introduce a fuzzy modeling approach to optimize the dual response system. We demonstrate our approach in two examples and show the advantages of our m..

Fuzzy versus statistical linear regression

Abstract Statistical linear regression and fuzzy linear regression have been developed from different perspectives, and thus there exist several conceptual and methodological differences between the two approaches. The characteristics of both methods, in terms of basic assumptions, parameter estimation, and application are described and contrasted. Their descriptive and predictive capabilities are also compared via a simulation experiment to identify the conditions under which one outperforms the other. It turns out that statistical linear regression is superior to fuzzy linear regression in terms of predictive capability, whereas their comparative descriptive performance depends on various factors associated with the data set (size, quality) and proper specificity of the model (aptness of the model, heteroscedasticity, autocorrelation, nonrandomness of error terms). Specifically, fuzzy linear regression performance becomes relatively better, vis-a-vis statistical linear regression, as the size of the data set diminishes and the aptness of the regression model deteriorates. Fuzzy linear regression may thus be used as a viable alternative to statistical linear regression in estimating regression parameters when the data set is insufficient to support statistical regression analysis and/or the aptness of the regression model is poor (e.g., due to vague relationship among variables and poor model specification).

An interactive desirability function method to multiresponse optimization

Multiresponse optimization problems often involve incommensurate and conflicting responses. To obtain a satisfactory compromise in such a case, a decision maker (DM)s preference information on the tradeoffs among the responses should be incorporated into the problem. This paper proposes an interactive method based on the desirability function approach to facilitate the preference articulation process. The proposed method allows the DM to adjust any of the preference parameters, namely, the shape, bound, and target of a desirability function in a single, integrated framework. The proposed method would be highly effective in generating a compromise solution that is faithful to the DMs preference structure.

A New Loss Function-Based Method for Multiresponse Optimization

A new loss function-based method for multiresponse optimization is presented. The proposed method introduces predicted future responses in a loss function, which accommodates robustness and quality of predictions as well as bias in a single framework. Properties of the proposed method are illustrated with two examples. We show that the proposed method gives more reasonable results than the existing methods when both robustness and quality of predictions are important issues.

QFD optimizer: a novice friendly quality function deployment decision support system for optimizing product designs

Abstract A novice-friendly decision support system prototype for quality function deployment (QFD) called QFD Optimizer is developed based upon an integrated mathematical programming formulation and solution approach. QFD Optimizer not only helps a design team build a house of quality chart, but also supports them in understanding and analyzing the system interrelationships, as well as obtaining optimal target engineering characteristic values. QFD Optimizer was tested experimentally and in a real design setting on students and practitioners to ascertain its potential viability and effectiveness. The results suggest that it has the potential to help users find improved feasible designs yielding higher customer satisfaction (i.e., improving quality of design) more rapidly (i.e., reduce the design cycle time), compared with the current manual, ad hoc approach. QFD Optimizer can be used by novice as well as expert users, and leads to a better understanding of complex interrelationships between customer needs and the engineering characteristics and among the engineering characteristics. Hence, it can and has been used as an effective quality improvement training tool, and shows promise for application in practice.

On assessing the H value in fuzzy linear regression

Abstract There are certain circumstances under which the application of statistical regression is not appropriate or even feasible because it makes rigid assumptions about the statistical properties of the model. Fuzzy regression, a nonparametric method, can be quite useful in estimating the relationships among variables where the available data are very limited and imprecise, and variables are interacting in an uncertain, qualitative, and fuzzy way. Thus, it may have considerable practical applications in many management and engineering problems. In this paper, the relationship among the H value, membership function shape, and spreads of fuzzy parameters in fuzzy linear regression is determined, and the sensitivity of the spread with respect to the H value and membership function shape is examined. The spread of a fuzzy parameter increases as a higher value of H and/or a decreasingly concave or increasingly convex membership function is employed. By utilizing the relationship among the H value, membership function, and spreads of the fuzzy parameters, a systematic approach to assessing a proper H parameter value is also developed. The approach developed and illustrated enables a decision makers beliefs regarding the shape and range of the possibility distribution of the model to be reflected more systematically, and consequently should yield more reliable and realistic results from fuzzy regression. The resulging regressing equations could, for example, also be used as constraints in a fuzzy mathematical optimization model, such as in quality function deployment.