Lai K. Chan

A New Measure of Process Capability: Cpm

A new measure of the process capability (Cpm) is proposed that takes into account the proximity to the target value as well as the process variation when assessing process performance. The sampling distribution for an estimate of Cpm (Ĉpm) and some of i..

Consumer satisfaction index: new practice and findings

This paper reports the development of the Hong Kong consumer satisfaction index (HKCSI) – a new type of consumer‐oriented economic performance indicator representing the quality of products (commodities and services) sold in Hong Kong as evaluated by Hong Kong consumers – as well as the findings of HKCSI from 1998 to 2000, each year with more than 10,000 successful telephone interviews on about 60 products. According to Hong Kongs special economic structure, the HKCSI is constructed from a consumption point of view concentrating on products and the products’ consumers, rather than from the popular production point of view focusing on firms and the firms’ customers. Key features of HKCSI include: the direct introduction of consumer characteristics (such as age, education, and income) in model construction; the wide coverage of services, especially free services; and the adoption of a product weighting system based on consumer price index (CPI), not on gross domestic product (GDP). In this paper, a theoretical framework of consumer satisfaction is first presented after investigating the relevant literature, and then the large‐scale consumer survey scheme adopted to collect the data and the structural equation modeling technique employed to estimate the indexes are discussed. New considerations in the HKCSI in such areas as model structure, indicator and questionnaire design, and product classification are elaborated upon. Next, the estimated results are analyzed focusing on the reliability and validity of the model, on the relationships among consumer satisfaction and its antecedents and consequences, and especially on the effects of consumer characteristics on consumer satisfaction and the implications of such effects for marketing practice. The results are generally consistent across different products, broadly acceptable and in agreement with previous findings, and are also relevant to Hong Kongs special situation. The paper concludes with a summary and some remarks on problems in the present study and future research directions.

A mulvariate measure of process Capability

AbstractA measure of process capability for the multivariate normal case is proposed. This measure takes into account both proximity to the target and the variation observed in the process. The result is analogous to the univariate measure of process capability referred to as Cpm. Some statistical properties associated with the measure are examined. Multivariate specification limits and their creation are also discussed.

PRIORITIZING THE TECHNICAL MEASURES IN QUALITY FUNCTION DEPLOYMENT

Quality function deployment (QFD) is a quality management system designed to create high levels of customer satisfaction. This system, originated in Japan in the 1970, consists of four interlinked phases which describe the connections between the custom..

On the asymptotic distributions of some process capability indices.

This paper shows that under general conditions, the distributions of the estimates of the process capability indices Cp, Cpk and Cpm, are asymptotically normal.

Combined Robust Design

Robust design, advocated by Taguchi, has been recognized as a useful approach for building quality into a product in its design stage. It consists of (1) using parameter design to determine the optimal setting of control factors which makes the product ..

Robust Estimation of Location Using Optimally Chosen Sample Quantiles

Robust estimation of the location parameter α based on selected order statistics is considered. The distribution function is only known to belong to a subset of a set D of distributions consisting of Tukeys lambda family of symmetric distributions with the inverse distribution function of the form α + β/γ) [λγ – (1 – λ)γ], 0 ≤ λ ≤ 1, and the normal, Cauchy and double exponential distributions. The scale parameter β can be unknown. In the set D, distributions with tail length varying from short to extremely long, when the shape parameter γ varies, are included. The asymptotically best linear estimate (ABLE) based on k (k = 2(1)5) optimally chosen symmetrical sample quantiles is considered. It can be used as a robust estimate and is shown to compete favorably with optimally trimmed and Winsorized means, in the sense of giving a higher guaranteed relative asymptotic efficiency (GRAE) for subsets of D. Tables are provided so that the robust k-ABLE giving the highest GRAE for any subset of D can easily be obt...

Robust Procedures for Estimating the Scale Parameter and Predicting Future Order Statistics of the Weibull Distribution

Two robust estimators of the scale parameter of the Weibull distribution are proposed: 1) Maximin Estimator (based on a mixture of Weibull distributions and the procedure of constructing the Best Linear Invariant Estimator), and 2) Adaptive Estimator (based on an adaptive procedure which chooses between the Maximin Estimator and the Best Linear Invariant Estimator from individual Weibull distributions). They are robust compared with some well-known estimators. Using similar methods, two robust predictors of future order statistics in a sample are proposed: 1) Maximin Predictor, and 2) Adaptive Predictor. The ideas behind the construction of the estimators and predictors can be applied to other distributions.

Linear estimation of the location and scale parameters based on selected order statistics

This paper gives a review of the best linear estimates of the location and/or scale parameters based on a few order statistics selected from a complete or censored sample. Small sample and large sample cases are considered and compared. Some examples of the practical applications of the estimates are outlined.

Robust Estimators of the 3-Parameter Weibull Distribution

Robust estimators of the location, scale, and shape parameters of the Weibull distribution are proposed. The estimators are easy to calculate and have few of the disadvantages associated with the maximum likelihood estimators. Their rms-errors are considerably smaller than those of the estimators available in the literature.