Browsing by Author "Pearn, W. L."

Now showing 1 - 20 of 104

A Bayesian approach for group supplier selections based on the popular process-capability-index C-pk
(2016)
Wu, Chin-Chieh
;
Shiau, Jyh-Jen Horng
;
Pearn, W. L.
;
Hung, Hui-Nien
;
統計學研究所
Supplier selection is an important part of supply chain management. In the initial stage of production setting, the decision-maker usually faces the problem of selecting the "best\'\' one(s) from available manufacturing material suppliers. For this purpose, process capability indices (PCIs) are commonly used in the literature to rank the suppliers under selection; and among these PCIs, C-pk could be the most popular one. This problem of selecting the best supplier(s) has received considerable attention in the literature but mainly from the frequentist point of view. In this paper, we tackle the so-called group supplier selection problem via the Bayesian approach, namely selecting a group of suppliers that would include the supplier of the largest C-pk value with a high level of confidence in a Bayesian sense. Based on the observed data and available prior information, we develop a practical procedure for the group supplier selection, which is useful for practitioners operating in-plant applications.
A Close Form Solution for the Product Acceptance Determination Based on the Popular Index C-pk
(2013-07-01)
Pearn, W. L.
;
Wu, C. H.
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Product acceptance determinations are practical tools for quality control applications involving quality contract on product orders between the vendor and the buyer. It provides the vendor and the buyer rules for product acceptance to meet the preset product quality requirement. As the rapid advancement of manufacturing technology, more than one quality characteristic must be simultaneously considered to improve the product quality because of the product design. In this article, we introduce an efficient product acceptance procedure on the basis of the generalization C-pk(T) index, to deal with lot sentencing problem with very low fraction of defectives. We tabulate the required sample size n and the corresponding critical acceptance value c(0) for various alpha-risk, beta-risk, and the levels of the lot fraction of defectives that correspond to acceptance and rejecting quality levels. Practitioners can use the proposed method to make reliable decisions in product acceptance. Copyright (C) 2012 John Wiley & Sons, Ltd.
A comparison of methods for estimating loss-based capability index
(2009)
Wu, Chien-Wei
;
Shu, Ming-Hung
;
Pearn, W. L.
;
Cheng, Feng-Tsung
;
工業工程與管理學系
The process capability index Cpm, sometimes called the loss-based index, has been proposed to the manufacturing industry for measuring process reproduction capability. This index incorporates the variation of production items with respect to the target value and the specification limits preset in the factory. To estimate the loss-based index properly and accurately, certain frequentist and Bayesian perspectives have been proposed to obtain lower confidence bounds (LCBs) for providing minimum process capability. The LCBs not only provide critical information regarding process performance but are also used to determine whether an improvement was made in a capability index and by extension in reducing the fraction of non-conforming items. In this paper, under the assumption of normality, based on frequentist and Bayesian senses, several existing approaches for constructing LCBs of Cpm are presented. Depending on the statistical methods used, we then classify these existing approaches into three categories and compared them in terms of the coverage rates and the mean values of the LCBs via simulations. The relative advantages and disadvantages of these approaches are summarized with some highlights of the relevant findings.
A comparison of two methods for transforming non-normal manufacturing data
(2007-01-01)
Chung, S. H.
;
Pearn, W. L.
;
Yang, Y. S.
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Many statistical methods applied to manufacturing quality control and operations management have been under the assumption that the process characteristic investigated is normally distributed. If the process characteristic is not normally distributed, a popular approach is to transform the non-normal data into a normal one. In this paper, we consider the Box-Cox transformation, and compare the transformation power using two different parameter estimation methods, including the maximum likelihood estimator (MLE) and the method of percentiles (MOP). The performance comparison is based on the pass rate under the Shapiro-Wilk normality test. The results show that, in general, the MOP has better pass rate, while the MLE has smaller power variation for most cases investigated. For small sample size (n = 5, 10) both methods perform equally well. For large sample size, the MOP is recommended due to its simplicity and significantly higher pass rate.
A Note on "Capability Assessment for Processes with Multiple Characteristics: A Generalization of the Popular Index Cpk"
(2013-03-01)
Pearn, W. L.
;
Wu, C. H.
;
Tsai, M. C.
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
The generalized yield index C-pk(T) establishes the relationship between the manufacturing specifications and the actual process performance, which provides a lower bound on process yield for two-sided processes with multiple characteristics. The results attended are very practical for industrial application. In this article, we extended the results in cases with one-sided specification and multiple characteristics. The generalized index C-PU(T) was considered, and the asymptotic distribution of the natural estimator (C)over-cap(PU)(T) was developed. Then, we derived the lower confidence bounds as well as the critical values of index C-PU(T). We not only provided some tables but also presented an application example. Copyright (c) 2012 John Wiley & Sons, Ltd.
A note on Group Selection with multiple quality characteristics: power comparison of two methods
(2019-03-04)
Pearn, W. L.
;
Lin, Chen-ju
;
Chen, Y. H.
;
Huang, J. Y.
;
工業工程與管理學系
The Group Selection problem is an essential problem in the supplier selection process. The objective of the problem is to select a subset of suppliers containing the best among multiple candidate suppliers. Manufacturers should procure parts from the selected suppliers to produce high-quality products. Lin, C.J., W.L. Pearn, J.Y. Huang, and Y.H. Chen [2017. "Group Selection for Processes with Multiple Quality Characteristics." Communications in Statistics - Theory and Methods. doi:] considered the problem under multiple quality characteristics, and proposed the Modified Bonferroni method and the Multiple Comparisons with the Best (MCB) method to tackle the problem. The two methods, however, may select different subset containing the best depending on the magnitude of the differences among the k estimated index values. In this paper, we derive the power function for the Modified Bonferroni method, and compare the power of the two methods with extensive simulations. The results show that the MCB method is more powerful than the Modified Bonferroni method when the actual number of the best process is one. On the other hand, the Modified Bonferroni method significantly outperforms the MCB method when the actual number of the best process is greater than one. The results provide practitioners with useful reference about the properties of the two methods for supplier selection.
A note on production yield measure for multiple lines
(2016)
Pearn, W. L.
;
Tai, Y. T.
;
Wu, C. H.
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
In today\'s globally competitive environment, processes involving multiple manufacturing lines are quite common due to economies of scale considerations. Production yield measure indices C-pk and C-pu are the most popular yield-based indices and have been widely used in manufacturing industries for two-sided and one-sided specification limits, respectively. In this paper, two production yield indices for multiple lines C-pk(M) and C-pu(M) are considered and the approximate distributions of two natural estimators C-pk(M) and C-pu(M) are derived. For the convenience of industry applications, the lower confidence bounds and critical values are provided.
A powerful test for two-sided multiple independent characteristics supplier selection problem
(2015-03-04)
Wu, C. H.
;
Pearn, W. L.
;
Chuang, C. C.
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
In this article, we consider the supplier selection problem for two-sided processes with multiple independent characteristics. We review the existing division method and develop a new exact approach called the subtraction method. For practical applications, a two-phase selection procedure is established based on the subtraction method. The decision powers of two methods are compared. We show that the subtraction method we proposed is indeed more powerful than the existing division method. Several figures are presented to display the required sample sizes with various powers and various values of (C-pk1(T), C-pk2(T)). For the convenience of practitioners, the required sample sizes for the two methods with various powers and various values of (C-pk1(T), C-pk2(T))) are also tabulated.
A recursive method for the F-policy G/M/1/K queueing system with an exponential startup time
(2008-06-01)
Wang, Kuo-Hsiung
;
Kuo, Ching-Chang
;
Pearn, W. L.
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
This paper deals with the optimal control of a finite capacity G/M/1 queueing system combined the F-policy and an exponential startup time before start allowing customers in the system. The F-policy queueing problem investigates the most common issue of controlling arrival to a queueing system. We provide a recursive method, using the supplementary variable technique and treating the supplementary variable as the remaining interarrival time, to develop the steady-state probability distribution of the number of customers in the system. We illustrate a recursive method by presenting three simple examples for exponential, 3-stage Erlang, and deterministic interarrival time distributions, respectively. A cost model is developed to determine the optimal management F-policy at minimum cost. We use an efficient Maple computer program to determine the optimal operating F-policy and some system performance measures. Sensitivity analysis is also studied. (c) 2007 Elsevier Inc. All rights reserved.
A variables sampling plan based on C-pmk for product acceptance determination
(2008-01-16)
Wu, Chien-Wei
;
Pearn, W. L.
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Process capability indices are useful management tools, particularly in the manufacturing industry, which provide common quantitative measures on manufacturing capability and production quality. Most supplier certification manuals include a discussion of process capability analysis and describe the recommended procedure for computing a process capability index. Acceptance sampling plans have been one of the most practical tools used in classical quality control applications. It provides both vendors and buyers to reserve their own rights by compromising on a rule to judge a batch of products. Both sides may set their own safeguard line to protect their benefits. Two kinds of risks are balanced using a well-designed sampling plan. In this paper, we introduce a new variables sampling plan based on process capability index C-pmk to deal with product sentencing (acceptance determination). The proposed new sampling plan is developed based on the exact sampling distribution hence the decisions made are more accurate and reliable. For practical purpose, tables for the required sample sizes and the corresponding critical acceptance values for various producer's risk, the consumer's risk and the capability requirements acceptance quality level (AQL), and the lot tolerance percent defective (LTPD) are provided. A case study is also presented to illustrate how the proposed procedure can be constructed and applied to the real applications. (C) 2006 Elsevier B.V. All rights reserved.
A wind turbine evaluation model under a multi-criteria decision making environment
(2012-12-01)
Lee, Amy H. I.
;
Hung, Meng-Chan
;
Kang, He-Yau
;
Pearn, W. L.
;
工業工程與管理學系
Due to the impacts of fossil and nuclear energy on the security, economics, and environment in the world, the demand of alternative energy resources is expanding consistently and tremendously in recent years. Wind energy production, with its safe and environmental characteristics, has become the fastest growing renewable energy source in the world. The construction of new wind farms and the installation of new wind turbines are important processes in order to provide a long-term energy production. In this research, a comprehensive evaluation model, which incorporates interpretive structural modeling (ISM) and fuzzy analytic network process (FANP), is constructed to select suitable turbines when developing a wind farm. A case study is carried out in Taiwan in evaluating the expected performance of several potential types of wind turbines, and experts in a wind farm are invited to contribute their expertise in determining the importance of the factors of the wind turbine evaluation and in rating the performance of the turbines with respect to each factor. The most suitable turbines for installation can finally be generated after the calculations. The results can be references for decision makers in selecting the most appropriate wind turbines. (C) 2012 Elsevier Ltd. All rights reserved.
Accuracy analysis of the percentile method for estimating non normal manufacturing quality
(2007)
Wu, Chien-Wei
;
Pearn, W. L.
;
Chang, C. S.
;
Chen, H. C.
;
工業工程與管理學系
Vannman ( 1995) proposed a superstructure C-p(u, v) of capability indices for processes with normal distributions, which include C-p, C-pk, C-pm, and C-pmk as special cases. Pearn and Chen (1997) considered a generalization of C-p(u, v), called C-Np(u, v), to cover processes with non normal distributions. Pearn and Chen ( 1997) also proposed a sample percentile estimator for the generalization C-Np(u , v). In this article, we investigate the performance of the sample percentile estimator. We perform some simulation study, which covers the normal distribution and various non normal distributions including the uniform distribution, chi-square distribution, student's t distributions, F distribution, beta distribution, gamma distribution, Weibull distribution, lognormal distribution, triangular distribution, and Laplace distribution, with selected parameter values. Extensive simulation results, comparisons, and analysis are provided.
An analytical closed-form solution for multiple line supplier selection problem
(2019-07-04)
Pearn, W. L.
;
Tai, Y. T.
;
Tseng, S. C.
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Supplier selection is an essential issue, particularly, for semiconductor manufacturing. Nowadays, since high demands in the globally competitive semiconductor manufacturing environment, multiple-line processes are quite commonly applied. In the paper, we consider a two-stage method composing of quality verification and selection decision for multiple-line supplier selection problems. Since quality performance is an important criterion in supplier selection, we test whether the new supplier has a significantly better capability than the existing supplier based on the multiple-line yield index in the stage of quality verification. In selection decision, when the magnitude of capability outperformance is significant, switching contracts to a new entrant supplier is suggested. In addition, analytical closed-form solutions for asymptotical critical values and required sample sizes are derived and tabulated for the investigated multiple-line supplier selection problem. We demonstrate the applicability of the proposed supplier selection method by presenting a real-world example taken from a semiconductor packaging shop floor located in the Science-based Industrial Park in Hsinchu, Taiwan.
An effective decision making method for product acceptance
(2007-02-01)
Pearn, W. L.
;
Wu, Chien-Wei
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Acceptance sampling plans are practical tools for quality assurance applications involving quality contract on product orders. The sampling plans provide the vendor and buyer decision rules for product acceptance to meet the preset product quality requirement. As the rapid advancement of manufacturing technology, suppliers require their products to be of high quality with very low fraction of defectives often measured in parts per million. Unfortunately, traditional methods for calculating fraction of defectives no longer work since any sample of reasonable size probably contains no defective product items. In this paper, we introduce an effective sampling plan based on process capability index C-pk to deal with product acceptance determination for low fraction of defectives. The proposed new sampling plan is developed based on the exact sampling distribution rather than approximation. Practitioners can use the proposed method to determine the number of required inspection units, the critical acceptance value, and make reliable decisions in product acceptance. (c) 2005 Elsevier Ltd. All rights reserved.
An effective powerful test for one-sided supplier selection problem
(2011)
Pearn, W. L.
;
Hung, H. N.
;
Chuang, Y. S.
;
Su, R. H.
;
統計學研究所
In this paper, we consider the supplier selection problem, which deals with comparing two one-sided processes and selecting a better one that has a higher capability. We first review two existing approximation approaches, and an exact approach proposed which we refer to as the division method. We then develop a new exact approach called the subtraction method. We compare the two exact methods on the selection power. The results show that the proposed subtraction method is indeed more powerful than the division method. A two-phase selecting procedure is then developed based on the subtraction method for practical applications. Some computational results are tabulated for practitioners' convenience.
An effective powerful test for one-sided supplier selection problem with multiple independent characteristics
(2016)
Pearn, W. L.
;
Wu, Chia-Huang
;
Chuang, Ching-Ching
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
In this paper, we use the CT PU index to investigate the supplier selection problem for one-sided processes with multiple independent characteristics. We first review the existing approach, which we refer to as the division method, then develop a new exact approach called the subtraction method. A two-stage selection procedure is developed based on the subtraction method for practical applications. We then compared the two methods with regard to selection power. The results show that the subtraction method we propose is indeed more powerful than the existing division method. A thin-film transistor type liquid-crystal display (TFT-LCD) application example is provided to illustrate the testing procedure.
An Effective Powerful Test for Supplier Selection Evaluation with Multiple Characteristics
(2012-07-01)
Pearn, W. L.
;
Tai, Y. T.
;
Wang, C. Y.
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
"Supplier selection in high-tech manufacturing industries has become more essential and has received considerable attention. In this paper, we consider a supplier selection problem with multiple characteristics in which we compare two-sided processes with multiple characteristics and select the better one with the higher capability. We first review an approach referred to as the division method. We then develop a new approach called the subtraction method. We compare the two methods in terms of their selection power. The results show that the proposed subtraction method is indeed more powerful than the division method. Two supplier-selection procedures are developed for practical applications. For practitioners' convenience in applying our procedures, the various sample sizes required for designated selection powers are tabulated and discussed. For illustration purposes, a case is considered and solved for supplier selection evaluation."
An Effective Procedure for Calculating Weibull Production Yield with Mean Shift
(2012-05-01)
Hsu, Ya-Chen
;
Pearn, W. L.
;
Li, Yuan-Yi
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Motorola, Inc. introduced its six sigma production quality initiative to the world in the 1980s. Six sigma production quality is estimated by assuming a 1.5 sigma shift to the process mean. Bothe provided a statistical reason for considering such a shift in the normal process mean, and he also provided a much more accurate capability calculation to measure the production yield (Bothe, D. R., "Statistical Reason for the 1.5 sigma Shift," Qual. Eng., Vol. 14(3), 2002, pp. 479-487). In this paper, we consider Weibull processes, which cover a wide class of non-normal processes. We calculate the mean shift adjustments under various sample sizes n and Weibull parameters, with the power fixed at 0.5. Then, we implement the derived results to develop an effective procedure to accurately calculate the production yield. Finally, to demonstrate the applicability of the proposed approach, the proposed procedure is applied to a real manufacturing process with mean shift.
An Effective Procedure for Supplier Selection Applied to Glass Substrate Processes with Multiple Lines
(2014-11-01)
Pearn, W. L.
;
Wu, C. H.
;
Chuang, C. C.
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
In this paper, we consider a supplier selection problem, which compares two glass substrate processes with multiple lines and selects the one that has a higher capability. A test statistic obtained by a subtraction method is employed to establish a hypothesis test with two phases. Critical values of the test are obtained to determine the selection decisions. Sample sizes required for various designated selection powers and confidence levels are also investigated. Practitioners can use the proposed method to make reliable decisions. This paper presents a real-world case of a glass substrate process. The selection procedure uses data to reach a decision in supplier selection. We compare the results of the division method to the proposed subtraction method. The results show that the proposed subtraction method is indeed more powerful than the division method.
An Effective Scheduling Approach for Maximizing Polyimide Printing Weighted Throughput in Cell Assembly Factories
(2009-07-01)
Chung, S. H.
;
Tai, Y. T.
;
Pearn, W. L.
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Polyimide printing (PI) is an important process operation but also often the cause of bottlenecks in capital-intensive cell assembly factories. Therefore, the development of an effective scheduling method to maximize throughput in this PI process is essential and difficult. In the polyimide printing scheduling problem (PISP), jobs are given weights and clustered by their product types, which must be processed on identical parallel PI machines. The setup times for two consecutive jobs between different product types in the PI machines are sequence-dependent. In this paper, the PISP is formulated as a mixed integer linear programming model. The PISP is also transformed into a multiple tour maximum collection problem (MTMCP), a well-known network problem which has been investigated extensively. Based on this transformation, one can therefore solve the PISP near-optimally using the efficient algorithm.