Browsing by Author "Su, Chao-Ton"

Now showing 1 - 8 of 8

A case study on the application of Fuzzy QFD in TRIZ for service quality improvement
(2008-10-01)
Su, Chao-Ton
;
Lin, Chin-Sen
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
The improvement of service quality so as to enhance customer satisfaction has been widely mentioned over the past few decades. However, a creative and systematic way of achieving higher customer satisfaction in terms of service quality is rarely discussed. Recently, TRIZ, a Russian acronym which means "Theory of Inventive Problem Solving," has been proven to be a well-structured and innovative way to solve problems in both technical and non-technical areas. In this study, a systematic model based on the TRIZ methodology is proposed to generate creative solutions for service quality improvement. This is done by examining first the determinants of service quality based on a comprehensive qualitative study in the electronic commerce sector. Then the correlation between the imprecise requirements from customers and the determinants of service quality is analyzed with Fuzzy Quality Function Deployment (QFD) in order to identify the critical determinants relating to customer satisfaction. After which, the corresponding TRIZ engineering parameters can be effectively applied in the TRIZ contradiction matrix to identify the inventive principles. A case study is illustrated to demonstrate the effectiveness of our approach in an e-commerce company, and its results are presented to show the applicability of the TRIZ methodology in the e-service sector.
Application of hoshin kanri for productivity improvement in a semiconductor manufacturing company
(2007-01-01)
Yang, Tsung-Ming
;
Su, Chao-Ton
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Purpose - The planning approach used is issue-based planning or formal strategic planning, leading to local optimization and duplicated redundancy. An approach is presented in the rapid growth of the business scale which emphasizes both customization and effectiveness, not only operation efficiency. The paper addresses these issues. Design/methodology/approach - A two-phase approach is applied to obtain a smooth implementation. A lean operation concept is executed in the first phase to reduce the engineer's workload, and the business process re-engineering (BPR) technique is used in phase two to transform the working model. Findings - This paper presents an extension model of hoshin kanri to translate strategies into achievable policies and actions to fulfil the objectives of the enterprise. This model not only comprises a complete plan and executive structure which is simple and effective, but also conducts complicated communication processes at different levels to achieve the common goals as envisioned by the executive management. Practical implications - A case study is introduced to show the complete policy deployment process and realize the actual "vital few". This study has been successfully implemented in the case company, and the labor productivity was improved by 6 percent over a two-year period. Originality/value - The paper presents a successful application of hoshin kanri for a semiconductor manufacturing company within a rapid growth environment.
Data mining for the diagnosis of type II diabetes from three-dimensional body surface anthropometrical scanning data
(2006-03-01)
Su, Chao-Ton
;
Yang, Chien-Hsin
;
Hsu, Kuang-Hung
;
Chiu, Wen-Ko
;
工業工程與管理學系
Diabetes mellitus has become a general chronic disease as a result of changes in customary diets. Impaired fasting glucose (IFG) and fasting plasma glucose (FPG) levels are two of the indices which physicians use to diagnose diabetes mellitus. Although this is a fairly accurate approach, the tests are expensive and time consuming. This study attempts to construct a prediction model for Type II diabetes using anthropometrical body surface scanning data. Four data mining approaches, including backpropagation neural network, decision tree, logistic regression, and rough set, were used to select the relevant features from the data to predict diabetes. Accuracy of classification was evaluated for these approaches. The result showed that volume of trunk, left thigh circumference, right thigh circumference, waist circumference, volume of right leg, and subjects' age were associated with the condition of diabetes. The accuracy of the classification of decision tree and rough set was found to be superior to that of logistic regression and backpropagation neural network. Several rules were then extracted based on the anthropometrical data using decision tree. The result of implementing this method is not only useful for the physician as a tool for diagnosing diabetes, but it is sophisticated enough to be used in the practice of preventive medicine. (C) 2006 Elsevier Ltd. All rights reserved.
Feature selection for the SVM: An application to hypertension diagnosis
(2008-01-01)
Su, Chao-Ton
;
Yang, Chien-Hsin
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
A support vector machine (SVM) is a novel classifier based on the statistical learning theory. To increase the performance of classification, the approach of SVM with kernel is usually used in classification tasks. In this study, we first attempted to investigate the performance of SVM with kernel. Several kernel functions, polynomial, RBF, summation, and multiplication were employed in the SVM and the feature selection approach developed [Hermes, L., & Buhmann, J. M. (2000). Feature selection for support vector machines. In Proceedings of the international conference on pattern recognition (ICPR'00) (Vol. 2, pp. 716-719)] was utilized to determine the important features. Then, a hypertension diagnosis case was implemented and 13 anthropometrical factors related to hypertension were selected. Implementation results show that the performance of combined kernel approach is better than the single kernel approach. Compared with backpropagation neural network method, SVM based method was found to have a better performance based on two epidemiological indices such as sensitivity and specificity. (c) 2006 Elsevier Ltd. All rights reserved.
Hoshin Kanri planning process in human resource management: recruitment in a high-tech firm
(2015-02-01)
Su, Chao-Ton
;
Yang, Tsung-Ming
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
The Hoshin Kanri (HK) planning process is widely used in Japanese and Western companies for strategy planning and performance measurement in the manufacturing sector. Human resource management (HRM) plays an important role in business operations; HRM practices have a significant positive effect on implementing total quality management and on employee and customer satisfaction. However, HRM has always been considered reactive, uncoordinated, and unprofessional, delivering poor, slow, and ineffective services. This study presents the envision, identify, diagnose, prioritise, execute, and review (EIDPER) model, an extension of HK. EIDPER translates strategies into achievable actions to fulfil organisational objectives. We introduce a case study to illustrate applying EIDPER to the recruitment process of the HRM department in a high-tech manufacturing firm. The approach was successfully applied in the HRM department of the case company and effectively reduced hiring costs by US$360,000 in 5 years and improved recruitment cycle time by 40%. Managerial implications are also discussed at the end of the study.
MTSbag: A Method to Solve Class Imbalance Problems
(2018-01-01)
Hsiao, Yu-Hsiang
;
Su, Chao-Ton
;
Fu, Pin-Cheng
;
Chen, Mu-Chen
;
運輸與物流管理系註：原交通所+運管所
Class imbalance is a common problem in classification problems. The Mahalanobis-Taguchi System (MTS) has been shown to be robust in addressing class imbalance problems owing to its inherent properties of classification model construction. The bagging learning approach often has been applied as a superior strategy to reduce the learning bias of classification algorithms. In this study, we propose MTSbag, which integrates the MTS and the bagging-based ensemble learning approaches to enhance the ability of conventional MTS in handling imbalanced data. We perform numerical experiments involving multiple datasets with various class imbalance levels to demonstrate the effectiveness of MTSbag, especially for datasets with high imbalance levels.
Optimization of the Cu wire bonding process for IC assembly using Taguchi methods
(2011-01-01)
Su, Chao-Ton
;
Yeh, Cheng-Jung
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
The yield of IC assembly manufacturing is dependent on wire bonding. Recently, the semiconductor industry demands smaller IC designs and higher performance requirements. As such, bonding wires must be stronger, finer, and more solid. The cost of gold is continuously appreciating, and this has become a key issue in IC assembly and design. Copper wire bonding is an alternative solution to this problem. It is expected to be superior over Au wires in terms of cost, quality, and fine-pitch bonding pad design. To obtain the best wire bonding quality, we employed Taguchi methods in optimizing the Cu wire bonding process. With Cu wire bonding technology, the production yield increased from 98.5% to 99.3% and brought approximately USD 0.7 million in savings. (C) 2010 Elsevier Ltd. All rights reserved.
Performance evaluation of a multi-echelon production, transportation and distribution system: A matrix analytical approach
(2007-01-16)
Wang, Fong-Fan
;
Su, Chao-Ton
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
This study proposes an originative method to evaluate complex supply chains. A tentative multi-echelon production, transportation and distribution system with stochastic factors built-in is employed as a test bed for the proposed method. The supply subsystem formulated in this study is a two-stage production facility with constant probability of feedback and stochastic breakdowns. The transportation subsystem is a service facility with one server. The distribution subsystem under study is a single central warehouse with M retailers. All the participants of the supply chain use base-stock policies and single-server settings. We investigated both the make-to-order (MTO) and make-to-stock (NITS) policies for different base-stock levels, as adopted at different sites. Applying quasi-birth-and-death (QBD) processes as decomposed building blocks and then using the existing matrix analytical computing approach for the performance evaluation of a tandem queue constitutes the main procedure of this study. We also discuss the possibilities of extending the current model to account for other inventory control policies as well as for multi-server case. Numerical study shows our proposed analytical model is robust for practical use. (c) 2005 Elsevier B.V. All rights reserved.