Quality Characteristic Of Interest Research Articles

Control Charts for Radial Error

In applications such as the drilling of holes or placement of components at a targeted location, the quality characteristic of interest is the radial distance of the center from its desired location. Control charts for radial error are established for the case where the center displacement errors in two orthogonal directions are normally and independently distributed with the same variance about the true center location. It is shown that a second chart based on the ratio of individual observed values of radial error can be used to distinguish between a radial shift in the means of the error distributions and an increase in their variability.

Analysis of Hard-Facing Appearance of Specific Powdered Superalloys for PTA-Coating Processes

Artificial neural network (ANN) modeling and multiple linear regression (MLR) analysis have been used to develop a powder hard-facing process using high-energy plasma-transferred (HEPT) heating. HEPT heating can produce coatings with minimal surface roughness. An optimal procedure was developed involving the least number of process parameters but producing the most desirable performance characteristic. The quality characteristic of interest is the surface roughness after HEPT processing, utilizing the “the-smaller-the-better” criterion. Process performance was evaluated with respect to the signal-to-noise ratios, which were obtainable through experiments. The experimental results conclude that ANN models demonstrate a greater accuracy of predicting the surface appearance than the MLR models in terms of prediction error and the coefficient of determination. The results also reveal the most significant process control parameters. The predicted value of powder hard-facing roughness, through the implementation of optimal settings, produces a satisfactory result. The confirmation experiment showed that the ANN method achieved the expected optimal design goals for the HEPT powder hard-facing, thereby justifying the reliability and feasibility of the approach.

The Effect of Correlation on Chain Sampling Plans

AbstractThe Dodge chain sampling plan (ChSP‐1) and its extensions are very useful in situations where testing is either destructive or costly. Its underlying assumption is that all units to be inspected are from the same process and the quality characteristic of interest follows an identical independent distribution. Although this assumption makes the model relatively simple and easy to implement, it may not hold for today's manufacturing processes with high production volume, in which correlation exists between products within the same process. In this paper, we propose a Markov chain model for chain sampling plans to model the dependency (correlation) between testing units. To achieve this, we assume that product units within each sample follow a Markov chain model and assume that they are independent when they are from different lots. The resulting OC curves and AOQ curves show that the discriminating power of chain sampling plans improves when there is a negative correlation between product units and deteriorates when the correlation is positive. Copyright © 2004 John Wiley & Sons, Ltd.

Joint ―X and R Charts with Two‐stage Samplings

AbstractWhen joint ―X and R charts are in use, samples of fixed size are regularly taken from the process, and their means and ranges are plotted on the ―X and R charts, respectively. In this article, joint ―X and R charts have been used for monitoring continuous production processes. The sampling is performed, in two stages. During the first stage, one item of the sample is inspected and, depending on the result, the sampling is interrupted if the process is found to be in control; otherwise, it goes on to the second stage, where the remaining sample items are inspected. The two‐stage sampling procedure speeds up the detection of process disturbances. The proposed joint ―X and R charts are easier to administer and are more efficient than the joint ―X and R charts with variable sample size where the quality characteristic of interest can be evaluated either by attribute or variable. Copyright © 2004 John Wiley & Sons, Ltd.

Comparing Multi‐response Design Methods with Mixed Responses

AbstractThe multi‐response optimization process is commonly applied to engineering design problems involving more than one quality characteristic of interest. Define quality characteristics as output responses that fall into three general categories: larger‐the‐better (LTB), smaller‐the‐better (STB), and nominal‐the‐best (NTB). In the multi‐response optimization process, the objective is to define a set of input factors that, when combined, provide the optimal overall response. This paper investigates the relationship between response types when they are mixed (i.e. LTB, and/or STB, and/or NTB) and the choice of approach. The paper demonstrates that the mix of response types impacts the choice of final input parameters and response levels. In addition, the choice of an additive versus a multiplicative approach may adversely impact the response relationship. That is, it may produce two optimal combinations of factor settings or drive responses away from their target values and toward their extreme values. Copyright © 2004 John Wiley & Sons, Ltd.

Índices de capacidade de relações funcionais lineares e não-lineares

São apresentados métodos para determinação dos índices de capacidade tais como Cp e Cpk aplicados no "design for six sigma" (DFSS) quando a característica de qualidade é não-observável e dada por Y = f(X1, X2, X3,....Xk), onde Xi são variáveis aleatórias com média e variância respectivamente iguais a miXi e sigma2Xi; f é uma função conhecida. A determinação é baseada na série de Taylor e o procedimento será ilustrado com exemplos práticos.

Training for Shainin’s approach to experimental design using a catapult

In the quest for continuous quality improvement of both products and processes, experimental design (ED) based on Shainin’s variables search method is still playing a major role in the manufacturing community. This article presents a simple methodology in a step‐by‐step manner for Shainin’s variables search method, in terms of uncovering the key process variables or factors which influence a response or quality characteristic of interest. In order to illustrate the potential of this powerful methodology as a problem solving tool, a simple catapult experiment was carried out. The authors believe that this simple experiment can be used in the classroom to train engineers and scientists on variables search method in order to identify the most critical variables which influence the process or product performance.

Using Neural Networks to Detect and Classify Out-of-control Signals in Autocorrelated Processes

This paper presents an artificial neural network model for detecting and classifying three types of non-random disturbances referred to as level shift, additive outlier and innovational outlier which are common in autocorrelated processes. To the best of our knowledge, this is the first time that a neural network has been considered for simultaneous detection and classification of such non-random disturbances. An AR (1) model is considered to characterize the quality characteristic of interest in a continuous process where autocorrelated observations are generated over time. The performance of the proposed procedure is evaluated through the use of a numerical example. Preliminary results indicate that the procedure can be used effectively to detect and classify unusual shocks in autocorrelated processes. Copyright © 2003 John Wiley & Sons, Ltd.

The Optimum Target Value under Single and Two-Stage Screenings

In this paper we consider the problem of determining the optimum target value of the quality characteristic of interest and the screening limits for a correlated variable under single and two-stage screenings. In single-stage screening, inspection is performed directly on the quality characteristic of interest or on a variable that is correlated with the characteristic. In two-stage screening, the correlated variable is inspected first to decide if an item should be accepted, rejected, or whether additional observations should be taken. If additional observations are required, the quality characteristic is then directly observed in order to classify the undecided items. Models are constructed that involve selling and discounted prices as well as production, inspection, and penalty costs for both single and two-stage screenings. Methods for finding the optimum process mean and the screening limits are presented when the quality characteristic and the correlated variable are assumed to be jointly normally distributed. A numerical example is presented.

A Smooth Nonparametric Approach to Multivariate Process Capability

The determination of the capability of a stable process with a multivariate quality characteristic using standard methods usually requires the assumption that the quality characteristic of interest follows a multivariate normal distribution. Unfortunately, multivariate normality is a difficult assumption to assess. Further, departures from this assumption can result in erroneous conclusions. In this article, I propose assessing the capability of a process using a nonparametric estimator. This estimator is based on a kernel estimate of an integral of a multivariate density. Bandwidth selection for this method is based on a smoothed bootstrap estimate of the mean squared error of the estimator. I also address the issue of constructing approximate confidence intervals. An example is presented that applies the proposed method to bivariate nonnormal process data. The performance of the resulting estimator is then compared to the sample proportion and a normal parametric estimate in a simulation study.

OPTIMUM PROCESS MEAN AND SCREENING LIMITS FOR A PRODUCTION PROCESS WITH THREE-CLASS SCREENING

The optimum process mean and screening limits are provided for a production process where the accepted item is sold in one of two alternative markets and the rejected item is reworked. All items are subject to screening where the screening variable may be the major quality characteristic of interest (performance variable) or the surrogate variable, which is highly correlated with the performance variable. Each item is classified into three quality grades A, B, and C: grade A items are sold to primary market, grade B items are sold to secondary market, and grade C items are reworked by the same manufacturing process. Assuming that performance and surrogate variables are jointly normally distributed, two profit models are constructed which include selling price, production, inspection, rework and penalty costs. Methods of finding the optimum process mean and the screening limits are presented and an illustrative example is given.

Characterization of a solder paste printing process and its optimization

The objectives of this research are to model the screening process with planned experiments and to identify the optimal setting of the process parameters so as to minimize the printing defects. The percentage volume matching (PVM) and defects per unit (DPU) are the two quality characteristics of interest. A fractional factorial design was employed to study simultaneously the effects of eight process factors on the PVM and DPU and their possible interactions. Subsequent analysis shows that a low level of the stencil cleaning interval and low temperature results in the minimum DPU while maintaining a PVM very close to 100 per cent. Empirical relationships between these two quality characteristics and the important factors were formulated using regression analysis and close matches were found during subsequent validation experiments.

Optimum mean value and screening limits for production processes with multi-class screening

This paper presents the problem of jointly determining the optimum process mean and screening limits for each market in situations where there are several markets with different price/cost structures. All items are inspected prior to shipment, and the inspection is performed on a variable which is highly correlated with the quality characteristic of interest. It is assumed that that quality characteristic is normally distributed with an unknown mean and a known variance. A profit model is constructed which involves selling price, production cost, penalty cost and inspection cost. Methods for finding the optimum process mean and screening limits for each market are presented.

A smooth nonparametric approach to process capability

The determination of the capability of a stable process using the standard process capability indices requires that the quality characteristic of interest be normally distributed. Departures from normality can result in erroneous conclusions when using these indices. In this paper we propose assessing the capability of a process using a nonparametric estimator. This estimator is based on kernel estimation of the distribution function. Bandwidth selection for this method can be based either on a normal reference distribution or on a nonparametric estimate. An example is presented that applies this proposed method to non-normal process data. The performance of the resulting estimator is then compared with the sample proportion and a normal-based estimate in a simulation study. © 1998 John Wiley & Sons, Ltd.

A smooth nonparametric approach to process capability

The determination of the capability of a stable process using the standard process capability indices requires that the quality characteristic of interest be normally distributed. Departures from normality can result in erroneous conclusions when using these indices. In this paper we propose assessing the capability of a process using a nonparametric estimator. This estimator is based on kernel estimation of the distribution function. Bandwidth selection for this method can be based either on a normal reference distribution or on a nonparametric estimate. An example is presented that applies this proposed method to non-normal process data. The performance of the resulting estimator is then compared with the sample proportion and a normal-based estimate in a simulation study. © 1998 John Wiley & Sons, Ltd.

The Cpk index and its confidence intervals

Introduces a practical method of constructing confidence intervals for the Cpk index. Constructs these confidence intervals based on the asymptotic normality of estimators for the sub‐indices of the Cpk index. As a result, the underlying distribution of the quality characteristic of interest need not be normal nor be known. Explains these sub‐indices and the Cp index and presents numerical examples and results of simulation studies of the confidence intervals.

DETERMINATION OF THE OPTIMAL PROCESS MEAN WHEN INSPECTION IS BASED ON A CORRELATED VARIABLE

Consider a production process where items are produced continuously. A lower limit is specified for the quality characteristic of interest, and an item is defective if its value of the quality characteristic is smaller than the specification limit. The process mean (target value) may be set higher to reduce the costs incurred by producing defective items. Using a higher process mean, however, results in a higher production cost when production cost is an increasing function of the quality characteristic. An inspection procedure is used to screen outgoing items from the defective ones. In this paper, we consider a situation where inspection is based on a surrogate variable instead of the quality characteristic of interest. A model is developed to jointly determine the most economical process mean and the inspection specification limit.

STRATEGIES FOR ROBUST MULTIRESPONSE QUALITY ENGINEERING

Virtually every manufactured product has more than one characteristic by which its overall quality is determined. Since those quality characteristics may be correlated, some changes in the levels of the controllable product design and process design variables may improve one of the quality characteristics while adversely affecting one or more of the other quality characteristics. In this paper we define a quadratic loss function for use with multiple quality characteristics. We discuss several strategies that could be employed for robust quality engineering of products and processes when there is more than one quality characteristic of interest.

Optimal control of a production process subject to AOQL constraint

A production process which exhibits a decreasing pattern in the mean quality during the course of production is considered. A lower limit is specified for the quality characteristic of interest, and an item is classified as defective if its quality characteristic is below this lower limit. The major concern of the manufacturer is the average outgoing quality (AOQ). Hence, the process has to be adjusted after a time to avoid producing a large proportion of defective items due to deteriorating quality. To meet consumer specifications, an average outgoing quality limit (AOQL) is specified. A process-control scheme is developed in which decisions as to the upper and lower limit of the process mean are made based upon the choice of an AOQL. We discuss the decision problem of selecting the starting level of the process mean, and the level at which the process mean should be adjusted back to the starting level, so that the AOQ is not larger than AOQL. We consider a cost model which includes a fixed cost for adjustment, and a production cost which is assumed to be a function of the quality level. Due to the complexity of the model, a search procedure is used to find the optimal solution. In addition, an approximate solution which requires only simple calculations is developed, and is shown to be very effective in finding near-optimal solutions.