A D-optimal design approach to robust design under constraints: a new Design for Six Sigma tool

Kovach, Jamison V.; Cho, Byung Rae

doi:10.1504/ijssca.2006.011567

Cited by 18 publications

(10 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…D-Optimality is the most common criterion for computer-generated optimal designs in conjoint analysis (Cuervo et al, 2017). In addition, the D-optimal design can be used when there are constraints such as not all possible combinations of levels being feasible (Kovach and Cho, 2006). For example in our study, the retail channel attribute, luxury channel and the price level similar to non-luxury fashion brand's retail price are incompatible in the marketplace and are therefore not included as a possible combination.…”

Section: Data Collection and Instrument Designmentioning

confidence: 99%

Exploring young consumer's decision‐making for luxury co-branding combinations

Rothenberg

Moore

2020

IJRDM

View full text Add to dashboard Cite

PurposeThe purpose of this paper is to identify the realistic trade-offs young consumers make when evaluating luxury co-branding combinations based on signalling theory.Design/methodology/approachConjoint analysis was employed to evaluate the relative impact of four major attributes (i.e. brand combinations, retail channels, uniqueness and price) on consumer desirability for luxury co-branding combinations. The data were analysed using desirability indices.FindingsBrand combinations, uniqueness and price significantly impact consumer desirability of luxury co-branding combinations. The luxury brand and sportswear combination results in the highest desirability when price is more similar to the sportswear constituent and participants perceive that the collaboration as exclusive.Practical implicationsThe results suggest that luxury brands need to consider the partnering brand's retail format primarily for co-branding strategy. Luxury brand collaborations with sportswear and premium priced streetwear brands are more likely to result in higher desirability among consumers compared to collaborations with fast fashion and mass-market brands. Additionally, uniqueness may not be effective as a point of differentiation in cases where luxury brands cannot guarantee a single yearly collaboration.Originality/valueThe decision to use existing brands for the fictitious combinations developed more sensible scenarios for respondents. In addition, rather than discrete questions, attribute-based combinations provide a more realistic depiction of consumers' decision making on luxury co-branding. Finally, the results provide marketing practitioners with practical directions for future development of fashion luxury co-branding strategy.

show abstract

Section: Data Collection and Instrument Designmentioning

confidence: 99%

Exploring young consumer's decision‐making for luxury co-branding combinations

Rothenberg

Moore

2020

IJRDM

View full text Add to dashboard Cite

show abstract

“…Using RSM, Huele and Engel 44 introduce a method of tolerance design for parameters, allowing non-normally distributed noise factors. Finally, the consideration of multiple quality characteristics in the context of RD has been studied in Govindaluri and Cho 45,46 and Kovach and Cho [47][48][49][50] .…”

Section: Literature Studymentioning

confidence: 99%

“…Some researchers recommend that the first two statistical moments (i.e. mean and standard deviation) be traded off (see Fan 35 , Kovach and Cho 47,48 , and Shin and Cho 62 ).…”

Section: Optimization Models and Solutionsmentioning

confidence: 99%

Development of a censored robust design model for time‐oriented quality characteristics

Shaibu

Cho

Kovach

2008

Quality & Reliability Eng

Self Cite

View full text Add to dashboard Cite

Robust design (RD) techniques, which are based on the concept of building quality into products or processes, are increasingly popular in industry primarily because of their practicality. Traditional RD principles have often been applied to situations in which the quality characteristics of interest are time-insensitive. However, when time-oriented quality characteristics are studied, censored data often occur. As a result, current RD models reported in the literature may not be effective in finding solutions based on such data. To address such practical needs, this paper develops a censored RD model. We also propose an estimation method that is closely related to the expectation-maximization algorithm and compare it with the method of maximum likelihood estimation via a numerical example. Model validation is conducted, and comparative studies are discussed for model verification.

show abstract

“…Therefore, the crucial notion behind environmental quality improvement should not be any different from just any other quality improvement attempt which is to find optimized solutions before massive operations receive the "all systems go" signal. In standard Six Sigma toolset, this is achieved during product and process pre-design work resorting to design of experiments (DOE) methodologies (Kovach and Cho, 2006;Goh, 2002). There is a lot of literature published recently on the use, misuse and applications of DOE (Arvidsson and Gremyr, 2008;Ilzarbe et al, 2008).…”

Section: Introductionmentioning

confidence: 99%

Eco‐design in total environmental quality management

Besseris

2012

The TQM Journal

View full text Add to dashboard Cite

Purpose -The purpose of this paper is to propose a methodology that may aid in assessing ecological quality multi-trait screening through the use of simple and robust tools while exerting minimal effort in conducting trials and interpreting results. Design/methodology/approach -Response data for two popular site-monitoring environmental indicators, chemical oxygen demand (COD) and biochemical oxygen demand (BOD), are arranged by implementing an 8-run saturated orthogonal array proposed by Taguchi. Unreplicated data consolidation is performed through the Super-Ranking translation. This permits converting the two eco-traits to a single master response which becomes much easier to manipulate statistically. Distribution-free multi-factor contrasting provides the data reduction engine to filter-out non-active eco-design variables for a waste treatment unit in a large dairy-products plant. Findings -Environmental quality improvement is achieved by accumulating structured eco-data sets through an unreplicated-saturated L 8 (2 7 ) Taguchi design. Concurrent minimization of the two selected eco-responses, COD and BOD, promotes in a statistically significant fashion the quantity of incoming wastes, set at its minimum load, as the sole active eco-factor. Practical implications -Brief but robust experimentation is exploited in gaining information about the phenomenological behavior of environmental quality indicators, namely COD and BOD, in facilities that manage wastewater treatment. Design for environment is enforced through standard DOE planning schemes. Collected multi-metric eco-quality data are translated non-parametrically in an easy-to-comprehend manner that requires no assist from software aids while bypassing more statistical intensive techniques which may demand involvement of more experienced personnel. The methodology is accessible to any level of statistical competence seamlessly intertwined to optimization demands for rapid inference needs. Originality/value -The method mixes up three distinctive "design-and-analysis" elements in order to provide optimal solution in a design-for-environment project. The sampling capabilities of Taguchi's orthogonal arrays in concert with Super-Ranking transformation fuse multi-eco-characteristics to a single easy-to-handle master unitless eco-response. Order statistics tables recently published in terms of true probabilities have been adopted for supplying the proper cutoff points to be utilized for gauging against observed rank sums for an 8-run orthogonal array screening. Quality managers and environmental engineers who contribute routinely to continuous eco-improvement projects in their Total Environmental Quality Management (TEQM) program may find this approach attractive and viable en route to a typical industrial pollution prevention control deployment.

show abstract

A D-optimal design approach to robust design under constraints: a new Design for Six Sigma tool

Cited by 18 publications

References 34 publications

Exploring young consumer's decision‐making for luxury co-branding combinations

Exploring young consumer's decision‐making for luxury co-branding combinations

Development of a censored robust design model for time‐oriented quality characteristics

Eco‐design in total environmental quality management

Contact Info

Product

Resources

About