Browsing by Author "Chang, Yung-Chia"

Now showing 1 - 20 of 32

A flexible web-based simulation game for production and logistics management courses
(2009-08-01)
Chang, Yung-Chia
;
Chen, Wen-Chih
;
Yang, Yung-Nien
;
Chao, Hui-Cheng
;
工業工程與管理學系
This study develops a flexible simulation game environment called SIMPLE (Simulation of Production and Logistics Environment) in order to raise teaching effectiveness and improve classroom teaching in emerging production paradigms. Instructors can tailor SIMPLE by simply setting the appropriate parameters. SIMPLE can thus be adopted at various teaching stages to teach different major business concepts, such as inventory management, capacity management, pricing determination and negotiation, and information-sharing between players. Meanwhile, SIMPLE was employed in two courses to evaluate the degree of students' acceptance of using SIMPLE in classrooms. The results also showed that SIMPLE was generally well received by students. (C) 2009 Elsevier B.V. All rights reserved.
A Novel AHP-Based Benefit Evaluation Model of Military Simulation Training Systems
(2015-01-01)
Chang, Kuei-Hu
;
Chang, Yung-Chia
;
Chung, Hsiang-Yu
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
With the constantly changing patterns of war, more technologically exquisite weapons are designed, increasing in cost and complexity. Training maneuvers with live ammunition are expensive and are prone to accidental casualties. Thus, many countries are gradually adopting simulation training systems to replace some actual exercises to reduce casualties and still maintain maximum combat readiness. However, each simulation training system has a different background with regard to time, source, function goal, and quality of environment and staff. It is also more difficult to assess the benefit of simulation training systems. Moreover, traditional benefit assessments of military simulation training systems have merely considered their efficiency, not safety, causing biased conclusions. To solve these issues, this paper integrates the analytic hierarchy process (AHP), important-performance analysis (IPA), and the 2-tuple fuzzy linguistic representation model to determine the benefits of simulation training systems. To verify the proposed approach, a numerical example of the evaluation of a training simulator system's benefit is adopted. Compared with the traditional AHP method, the proposed method does not lose any valuable information that is provided by experts and also considers training safety. Further, these data are presented in 2-dimensional graphs for managers to further guide the decision making process.
A Novel Credit Risk Assessment Model Using a Granular Computing Technique
(2014-11-01)
Chang, Yung-Chia
;
Chang, Kuei-Hu
;
Hsiao, Chien-Wen
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Information about financial risk almost always contains a problem of class imbalance. Class-imbalanced data refers to the asymmetric categories of data, and it is divided into a major class and a minor class. If we guide all information into the training sample to model of this situation, it may happen that the accuracy rate of the major class is high, but the accuracy rate of the minor class is too low. Many risk assessment models have been developed in many studies, but most of them only use sampling methods to deal with the class-imbalanced data; this may cause the distortion of information. In order to effectively solve the problem of class imbalance in credit risk assessment, this paper proposed a novel credit risk assessment model using a granular computing technique to construct a risk assessment model to provide a better insight into the essence of data and effectively solve class imbalance problems. On the other hand, in order to improve the lack of granular computing and enhance the efficiency of the credit risk assessment model, this paper adds a new index, "% of minor class (PM)," to avoid a situation in which minor class data spread to the major class granular. Finally, this paper also compares the results of the area under the receiver operating characteristic curve (AUC) and G-means methods for dealing with class-imbalanced data. The results demonstrate that the proposed granular computing credit assessment model would have better results than other sampling models.
A Novel Method for Designing Information Technology Services
(2016-01-01)
Kao, Yi-Chih
;
Chang, Yung-Chia
;
Peng, Sheng-Lung
;
Chang, Ruay-Shiung
;
交大名義發表
Information technology (IT) services are diverse and complex. Numerous service design methods have been developed for designing and developing products. However, owing to the limited availability of useful tools in IT service design, these methods are ineffective. This study proposes an innovative three-dimensional method for designing IT services. The proposed method considers user requirements, an organization's business requirements, service providers' management requirements as well as the decision-making criteria of management representatives to ensure a smooth implementation of a designed IT service. Using this method, a prototype system to improve campus wireless local area network services was developed and tested at a university in Taiwan. The prototype system reduced the need for repetitive authentication and the time required to solve service problems and address user complaints. In addition, the service design team observed an approximately 70% reduction in project cycle time. The proposed method provided a systematic means to organize the design and implementation of IT services.
A NOVEL NETWORK SIMULATION SOFTWARE FOR SUPPLY CHAIN MANAGEMENT COURSES
(2018-01-01)
Chang, Yung-Chia
;
Chang, Kuei-Hu
;
Lu, Mei-Te
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Supply chain management courses require more content that focuses on real practice and applications to increase students' interest in learning. A simulation software is an effective method that simulates a virtual environment to allow students to apply and comprehend the knowledge they have learned from lectures. Recently, Chang, Chen, Yang, and Chao (2009) developed a simulation of a production and logistics environment (referred to as SIMPLE hereafter) software as a teaching aid which has been adapted in different courses teaching productions and logistics. However, in SIMPLE, only a linear supply chain model can be constructed, which means that only one affiliate in each echelon. Therefore, SIMPLE cannot be applied when teaching material related to supply chain network design. In order to encourage students to explore the knowledge of supply chain design without actually managing a real supply chain, this study constructed simulation software to allow students design their own supply networks to meet customers' demand and to compare the performance resulted from different supply chain networks. The functions of this software include adjustable facility selection for different members of supply chains, each of which can be associated with different inventory replenishment policy and transportation arrangement. The network constructed is displayed graphically and can also be saved for reuse. This software can set varied scenarios for lecturing or can be used as a mean for students to compete with each other by applying their knowledge learned in lectures. After using this software in a graduate-level Supply Chain Management course in a national university in Taiwan, it is found that students were attracted to this game/toy-based learning and were willing to explore related material on their own.
An ant colony optimization heuristic for an integrated production and distribution scheduling problem
(2014-04-03)
Chang, Yung-Chia
;
Li, Vincent C.
;
Chiang, Chia-Ju
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Make-to-order or direct-order business models that require close interaction between production and distribution activities have been adopted by many enterprises in order to be competitive in demanding markets. This article considers an integrated production and distribution scheduling problem in which jobs are first processed by one of the unrelated parallel machines and then distributed to corresponding customers by capacitated vehicles without intermediate inventory. The objective is to find a joint production and distribution schedule so that the weighted sum of total weighted job delivery time and the total distribution cost is minimized. This article presents a mathematical model for describing the problem and designs an algorithm using ant colony optimization. Computational experiments illustrate that the algorithm developed is capable of generating near-optimal solutions. The computational results also demonstrate the value of integrating production and distribution in the model for the studied problem.
An enhanced model for SDBR in a random reentrant flow shop environment
(2014-03-19)
Chang, Yung-Chia
;
Huang, Wen-Tso
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
This study proposed an enhanced simplified drum-buffer-rope (SDBR) model to be applied in a reentrant flow shop (RFS) in which job processing times are generated from a discrete uniform distribution and machine breakdowns are subject to an exponential distribution. In this enhanced SDBR model, the due-date assignment method, order release rule and dispatching rule were improved. The due dates and release dates of orders were determined by considering the total planned load of the capacity-constrained resource (CCR) in a random RFS. The deviation rate of buffer status is used as a dispatching rule to eliminate the influence of machine breakdowns. Simulations based on a real case company are used to evaluate the effective of the proposed model. The experimental results showed that our approach yields better performance than the other methods in terms of six due-date-related indexes when the product mix is with a large proportion of multi-reentrant orders and when the utilisation of CCR increases from 60 to 90%.
AN INNOVATIVE APPROACH INTEGRATING 2-TUPLE AND LOWGA OPERATORS IN PROCESS FAILURE MODE AND EFFECTS ANALYSIS
(2012-01-01)
Chang, Kuei-Hu
;
Chang, Yung-Chia
;
Wen, Ta-Chun
;
Cheng, Ching-Hsue
;
工業工程與管理學系
Process failure mode and effects analysis (FMEA) is used in the high-tech industry to assure process robustness. Traditionally, process FMEA uses the risk priority number (RPN) to evaluate the risk of failure. However, the RPN method has several main shortcomings. In order to improve the method of RPN evaluation, this paper proposes an innovative approach, integrating 2-tuple and the linguistic ordered weighted geometric averaging (LOWGA) operator in process FMEA. The main purpose of this study is to effectively resolve the main shortcomings of RPN evaluation to assure product and process robustness. In numerical verification, process FMEA of the color super-twisted nematic (CSTN) product is presented to further illustrate the proposed approach. After comparing the result that was obtained from the proposed method with both the traditional RPN method and the LOWGA method, it was found that the proposed approach provided a more accurate and reasonable ranking of the risk of failure. Moreover, the proposed method does not lose the useful information provided by the experts.
Application of eXtreme gradient boosting trees in the construction of credit risk assessment models for financial institutions
(2018-12-01)
Chang, Yung-Chia
;
Chang, Kuei-Hu
;
Wu, Guan-Jhih
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
The majority of the studies on credit risk assessment models for financial institutions during recent years focus on the improvement of imbalanced data or on the enhancement of classification accuracy with multistage modeling. Whilst multistage modeling and data pre-processing can boost accuracy somewhat, the heterogeneous nature of data may affects the classification accuracy of classifiers. This paper intends to use the classifier, eXtreme gradient boosting tree (XGBoost), to construct a credit risk assessment model for financial institutions. Cluster-based under-sampling is deployed to process imbalanced data. Finally, the area under the receiver operative curve and the accuracy of classifications are the assessment indicators, in the comparison with other frequently used single-stage classifiers such as logistic regression, self-organizing algorithms and support vector machine. The results indicate that the XGBoost classifier used by this paper achieve better results than the other three and can serve as a superior tool for the development of credit risk models for financial institutions. (C) 2018 Elsevier B.V. All rights reserved.
Applied column generation-based approach to solve supply chain scheduling problems
(2013-07-01)
Chang, Yung-Chia
;
Chang, Kuei-Hu
;
Chang, Teng-Kai
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
More and more enterprises have chosen to adopt a made-to-order business model in order to satisfy diverse and rapidly changing customer demand. In such a business model, enterprises are devoted to reducing inventory levels in order to upgrade the competitiveness of the products. However, reductions in inventory levels and short lead times force the operation between production and distribution to cooperate closely, thus increasing the practicability of integrating the production and distribution stages. The complexity of supply chain scheduling problems (integrated production and distribution scheduling) is known to be NP-hard. To address the issues above, an efficient algorithm to solve the supply chain scheduling problem is needed. This paper studies a supply chain scheduling problem in which the production stage is modelled by an identical parallel machine scheduling problem and the distribution stage is modelled by a capacitated vehicle routing problem. Given a set of customer orders (jobs), the problem is to find a supply chain schedule such that the weighted summation of total job weighted completion time and total job delivering cost are minimised. The studied problem was first formulated as an integer programme and then solved by using column generation techniques in conjunction with a branch-and-bound approach to optimality. The results of the computational experiments indicate that the proposed approach can solve the test problems to optimality. Moreover, the average gap between the optimal solutions and the lower bounds is no more than 1.32% for these test problems.
Applied Variable Neighborhood Search-Based Approach to Solve Two-Stage Supply Chain Scheduling Problems
(2016-05)
Chang, Yung-Chia
;
Chang, Kuei-Hu
;
Kang, Tien-Chi
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
When a company uses a make-to-order or direct-order business model, it needs to keep a low finished goods inventory level to be competitive. However, a low inventory level will cause the company to have little buffer to face the unforeseeable demand surge, which strengthens the close relationship between the stage of production and distribution. This paper probes an integrated production and distribution scheduling problem in which jobs are first produced by a set of unrelated parallel machines and then directly distributed to the corresponding customers by vehicles with limited capacity, with no inventory kept at the production stage. The objective is to find a joint production and distribution schedule such that the total cost, considering both customer service level and total distribution cost, is minimized. This paper integrated the earliest available machine first (EAMF) and variable neighborhood search (VNS) to solve the two-stage scheduling problem and used EAMF, which considers the load of the unrelated parallel machine to generate a set of production schedules to be systematically optimized by VNS, along with optimized routing, and applied a shaking mechanism of VNS to escape from the local optimal solution. The results of the computational experiments showed that the proposed method is better than ant colony optimization (ACO) heuristic methods in the computation time and solution quality.
Applying the concept of exponential approach to enhance the assessment capability of FMEA
(2014-12-01)
Chang, Kuei-Hu
;
Chang, Yung-Chia
;
Lai, Pei-Ting
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Failure modes and effects analysis (FMEA) has been used to identify the critical risk events and predict a system failure to avoid or reduce the potential failure modes and their effect on operations. The risk priority number (RPN) is the classical method to evaluate the risk of failure in conventional FMEA. RPN, which ranges from 1 to 1000, is a mathematical product of three parameters-severity (S), occurrence (O), and detection (D)-to rank and assess the risk of potential failure modes. However, there are some shortcomings of the conventional RPN method, such as: the RPN elements have many duplicate numbers; violate the assumption of measurement scales; and have not considered the weight of S, O, and D. In order to improve the aforementioned shortcomings of the conventional RPN calculation problem, this paper presents an easy yet effective method to enhance the risk evaluation capability of FMEA. The new method is named exponential risk priority number (ERPN), which uses a simple addition function to the exponential form of S, O, and D to substitute the conventional RPN method, which is a mathematical product of three parameters. Two practical cases are used to demonstrate that the ERPN method can not only resolve some problems of the conventional RPN method but also is able to provide a more accurate and reasonable risk assessment in FMEA.
Cost-Benefit Analysis of RFID Application in Apparel Retailing for SME: A Case from Taiwan
(2010-06-01)
Wen, Ta-Chun
;
Chang, Yung-Chia
;
Chang, Kuei-Hu
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Decision Support System for Inventory Management by TOC Demand-Pull Approach
(2010)
Hung, Kun-Tung
;
Liou, Edwin D.
;
Tsai, Hsin-Feng
;
Shi, Chi-Sheng
;
Chang, Yung-Chia
Theory of Constraint (TOC) suggests the application of demand-pull approach combined with buffer management to effectively manage inventory. The initial target buffer size, the timing to adjust the buffer and the amount of size to be adjusted are major managerial decisions that determine whether the demand-pull approach can be successfully applied in practice. However, there exist few studies that discuss the determination of parameters of buffer management in depth. Therefore, we constructed a decision support system (DSS) to assist users in finding applicable parameters when using demand-pull replenishment combined with buffer management. This DSS provides two modules: the search-of-parameter and the specific-parameter modules. The search-of-parameter module allows users to set the desired range of each strategic parameter to be simulated. Based on the range assigned, the system automatically generates various combinations of the strategic parameters and provides one simulation run for each combination. The specificparameter module only simulates one specific combination of parameters assigned and provides a more detail graphical display for the simulated results. The DSS also provides various charts for users to analyze the performance of a specific combination of parameters. To demonstrate the application of this DSS, we use a set of historical data for an integrated circuits (IC) product provided by a wafer foundry in Taiwan. From the results in the simulation, feasible actions to the inventory control will be suggested.
Determining the optimal re-sampling strategy for a classification model with imbalanced data using design of experiments and response surface methodologies
(2011-04-01)
Tong, Lee-Ing
;
Chang, Yung-Chia
;
Lin, Shan-Hui
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Imbalanced data are common in many machine learning applications. In an imbalanced data set, the number of instances in at least one class is significantly higher or lower than that in other classes. Consequently, when classification models with imbalanced data are developed, most classifiers are subjected to an unequal number of instances in each class, thus failing to construct an effective model. Balancing sample sizes for various classes using a re-sampling strategy is a conventional means of enhancing the effectiveness of a classification model for imbalanced data. Despite numerous attempts to determine the appropriate re-sampling proportion in each class by using a trial-and-error method in order to construct a classification model with imbalanced data (Barandela, Vadovinos, Sanchez, & Ferri, 2004; He, Han, & Wang, 2005; Japkowicz, 2000; McCarthy, Zabar, & Weiss, 2005), the optimal strategy for each class may be infeasible when using such a method. Therefore, this work proposes a novel analytical procedure to determine the optimal re-sampling strategy based on design of experiments (DOE) and response surface methodologies (RSM). The proposed procedure, S-RSM, can be utilized by any classifier. Also, C4.5 algorithm is adopted for illustration. The classification results are evaluated by using the area under the receiver operating characteristic curve (AUC) as a performance measure. Among the several desirable features of the AUC index include independence of the decision threshold and invariance to a priori class probabilities. Furthermore, five real world data sets demonstrate that the higher AUC score of the classification model based on the training data obtained from the S-RSM is than that obtained using oversampling approach or undersampling approach. (C) 2010 Elsevier Ltd. All rights reserved.
Enhancement of Inventory Management for the Wafer Manufacturing Industry by Combining Market Demand Forecast and Demand-Pull Replenishment
(2015-07-01)
Chang, Yung-Chia
;
Chang, Kuei-Hu
;
Sun, Wei-Chan
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Theory of constraints (TOC) uses a demand-pull replenishment strategy with buffer management (DPBM) to effectively manage inventory. This strategy has been demonstrated to perform better than traditional inventory strategies, such as (s, S), (s, Q), (R, S), (R, s, S) policies. However, the replenishment lead time is long, and demand changes rapidly in the semiconductor industry. These characteristics cause problems in managing semiconductor inventory and complicate the inventory replenishment strategy. In order to effectively solve the issue of inventory management, this paper proposes a novel approach to improve the classical DPBM by using market demand forecast information and demand-pull replenishment to effectively manage inventory in the semiconductor industry. In numerical verification, this paper used real data provided by a wafer foundry in Taiwan to demonstrate the feasibility and effectiveness of the proposed approach. Simulated scenarios that account for different patterns of demand and different degrees of accuracy of the demand forecasts were generated to test the performance of the proposed approach. After comparing the result that was obtained from the proposed approach with the one from the classical DPBM, it was found that the proposed approach can reduce the average inventory and does not adverse the level of service as long as the demand forecasts are not too far from the real demand.
Enhancing FMEA assessment by integrating grey relational analysis and the decision making trial and evaluation laboratory approach
(2013-07-01)
Chang, Kuei-Hu
;
Chang, Yung-Chia
;
Tsai, I-Tien
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Failure modes and effects analysis (FMEA) is used widely to improve product quality and system reliability, employing a risk priority number (RPN) to assess the influence of failures. The RPN is a product of three indicators-severity (S), occurrence (0), and detection (D)-on a numerical scale from 1 to 10. However, the traditional RPN method has been criticized for its four chief shortcomings: its (1) high duplication rate; (2) assumption of equal importance of S, 0, and D; (3) not following the ordered weighted rule; and (4) failure to consider the direct and indirect relationships between failure modes (FMs) and causes of failure (CFs). To resolve these drawbacks, we propose a novel approach, integrating grey relational analysis (GRA) and the decision-making trial and evaluation laboratory (DEMATEL) method, to rank the risk of failure, wherein the GRA is used to modify RPN values to lower duplications and the ordered weighted rule is followed; then, the DEMATEL method is applied to examine the direct and indirect relationships between FMs and CFs, giving higher priority when a single CF causes FMs to occur multiple times. Finally, an actual case of the TFT-LCD cell process is presented to verify the effectiveness of our method compared with other methods in providing decision-makers more reasonable reference information. (C) 2013 Elsevier Ltd. All rights reserved.
Establishing decision tree-based short-term default credit risk assessment models
(2016)
Chang, Yung-Chia
;
Chang, Kuei-Hu
;
Chu, Heng-Hsuan
;
Tong, Lee-Ing
;
工業工程與管理學系
Traditional credit risk assessment models do not consider the time factor; they only think of whether a customer will default, but not the when to default. The result cannot provide a manager to make the profit-maximum decision. Actually, even if a customer defaults, the financial institution still can gain profit in some conditions. Nowadays, most research applied the Cox proportional hazards model into their credit scoring models, predicting the time when a customer is most likely to default, to solve the credit risk assessment problem. However, in order to fully utilize the fully dynamic capability of the Cox proportional hazards model, time-varying macroeconomic variables are required which involve more advanced data collection. Since short-term default cases are the ones that bring a great loss for a financial institution, instead of predicting when a loan will default, a loan manager is more interested in identifying those applications which may default within a short period of time when approving loan applications. This paper proposes a decision tree-based short-term default credit risk assessment model to assess the credit risk. The goal is to use the decision tree to filter the short-term default to produce a highly accurate model that could distinguish default lending. This paper integrates bootstrap aggregating (Bagging) with a synthetic minority over-sampling technique (SMOTE) into the credit risk model to improve the decision tree stability and its performance on unbalanced data. Finally, a real case of small and medium enterprise loan data that has been drawn from a local financial institution located in Taiwan is presented to further illustrate the proposed approach. After comparing the result that was obtained from the proposed approach with the logistic regression and Cox proportional hazards models, it was found that the classifying recall rate and precision rate of the proposed model was obviously superior to the logistic regression and Cox proportional hazards models.
EZ-Net BYOD Service Management in Campus Wireless Networks
(2017-07-01)
Kao, Yi-Chih
;
Chang, Yung-Chia
;
Chang, Ruay-Shiung
;
工業工程與管理學系
;
資訊技術服務中心
We consider management issues associated with bring your own device (BYOD) services on a university campus. More than 70% of the users at the studied university access wireless networks at the campus through BYOD services. Thus, we propose the EZ-Net BYOD service management system, which implements a fail-through authentication function to an existing wireless local area network controller via multiple externally established authentication servers. The proposed system requires no client-side configuration and is not limited to any particular wireless network equipment. The proposed system was deployed and evaluated on campus. Results show that the EZ-Net system saves an average of 60 min of authentication time per month for each user. In addition, it offers a low-cost and efficient alternative to commercial BYOD service management software.
Improving the Efficiency of TOC Demand-Pull and Buffer Management by Incorporating Demand Information Using EWMA in Semiconductor Manufacturing
(2014-01-01)
Chang, Yung-Chia
;
Liang, Shu-Wen
;
工業工程與管理學系
;
Department of Industrial Engineering and Management
Products in the semiconductor manufacturing industry usually have characteristics of having short life-spans, volatility in demands and long lead times. Applying the demand-pull approach and buffer management (DPBM) suggested by the Theory of Constraints (TOC) to manage the inventory of such products does not efficiently respond to the unpredicted demand change thus could potentially result in either excessive inventories or shortages. This research proposed a new buffer adjustment mechanism to improve DPBM approach by applying the methodology of Exponentially Weighted Moving Average (EWMA) to integrate the true demands and the rolling forecasts of market in order to make the stock replenishing decision. Real product demand data provided by a wafer foundry in Taiwan is analyzed to demonstrate the effectiveness of the proposed approach. The study results show that the proposed idea can indeed keep a lower inventory level and provide a higher service standard in contrast to the traditional DPBM method.