Ching-Shih Tsou
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Ching-Shih Tsou has a total of 16 co-authors. Here are the top co-authors by number of shared publications:
Chin-Hsiung Hsu
Department of Financial Engineering and Actuarial Mathematics, Soochow University, Taipei, Taiwan
3 shared publicationsKun-Jen Chung
Chung Yuan Christian University, College of Business , Taiwan , R.O.C
3 shared publicationsLongsheng Cheng
Nanjing University of Science and Technology
2 shared publicationsC.-H. Kao
Shih Hsin University , Taipei 11604, Taiwan, R.O.C
1 shared publication- Chin Ta Wu
College of Business, National Taipei University of Business, Taipei, Taiwan
1 shared publication Ching-Chiang Yeh
Department of Business Administration, National Taipei College of Business, Taiwan
1 shared publication
Latest Publications
- Open Access
- 28 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 93012-93018
Data Augmentation With CycleGAN to Build a Classifier for Novel Defects From the Dicing Stage of Semiconductor Package Assembly
Industry 4.0, a concept first proposed by Germany, has resulted in an increasing number of companies adopting a mass customization strategy. This strategy is widely used across various industries, enabling the production of small batches of diversified products to meet the diverse needs of customers. It encompasses the business process of providing customized goods that best fulfill individual customer requirements, thereby necessitating small-scale production of multiple products. Therefore, the product life cycle of mass customization is much shorter than other production strategies. When product line changes are frequent and customized products have high yield rates, accurately detecting potential defects from a limited number of images is a daunting challenge. If the defect identification classification model needs to maintain a certain level of identification accuracy and the model needs to be deployed quickly, it is impossible to wait until a large number of defect images are collected before deploying an accurate model for new defects. Obtaining a high-precision defect identification classification model is crucial. In this study, we employed the style transfer method of CycleGAN, which takes advantage of the unmatched training images, to successfully transfer the style of defective images from old products to defect-free images of new products. However, CycleGAN requires a large number of images for training, so this study primarily focuses on rare sample categories. We first obtained the defect mask through a semantic segmentation model and then separated the foreground defect from the wafer background using digital image processing techniques. We then copied and pasted the separated defect onto a new wafer background to generate fake defect images. Finally, a generative adversarial network architecture was used to perform image blending to make the fake defect images more natural and realistic. The effectiveness of the data augmentation method was verified through a convolutional neural network model. Through the proposed method in this study, the number of defect images in new products was successfully increased, which helps to deploy a defect identification classification model for new products quickly.
- 28 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 93012-93018
Open AccessData Augmentation With CycleGAN to Build a Classifier for Novel Defects From the Dicing Stage of Semiconductor Package Assembly
Industry 4.0, a concept first proposed by Germany, has resulted in an increasing number of companies adopting a mass customization strategy. This strategy is widely used across various industries, enabling the production of small batches of diversified products to meet the diverse needs of customers. It encompasses the business process of providing customized goods that best fulfill individual customer requirements, thereby necessitating small-scale production of multiple products. Therefore, the product life cycle of mass customization is much shorter than other production strategies. When product line changes are frequent and customized products have high yield rates, accurately detecting potential defects from a limited number of images is a daunting challenge. If the defect identification classification model needs to maintain a certain level of identification accuracy and the model needs to be deployed quickly, it is impossible to wait until a large number of defect images are collected before deploying an accurate model for new defects. Obtaining a high-precision defect identification classification model is crucial. In this study, we employed the style transfer method of CycleGAN, which takes advantage of the unmatched training images, to successfully transfer the style of defective images from old products to defect-free images of new products. However, CycleGAN requires a large number of images for training, so this study primarily focuses on rare sample categories. We first obtained the defect mask through a semantic segmentation model and then separated the foreground defect from the wafer background using digital image processing techniques. We then copied and pasted the separated defect onto a new wafer background to generate fake defect images. Finally, a generative adversarial network architecture was used to perform image blending to make the fake defect images more natural and realistic. The effectiveness of the data augmentation method was verified through a convolutional neural network model. Through the proposed method in this study, the number of defect images in new products was successfully increased, which helps to deploy a defect identification classification model for new products quickly.
- Open Access
- 6 June 2022
- Published by Taylor & Francis in Cogent Engineering
- Vol. 9 (1)
Quality prediction through machine learning for the inspection and manufacturing process of blood glucose test strips
Although machine learning for quality prediction of manufacturing processes has attracted attention in the literature, there is a significant lack of case studies from industry, especially in medical sector. This paper proposes a data-driven approach to infer the batch quality of blood glucose test strips. Once the low quality of work in process is detected, unnecessary process waste could be eliminated. Starting from data pre-processing, which consists of Synthetic Minority Over-sampling TEchnique and Random Over-Sampling Example, this project tries to balance the ill-distributed data first. Followed by machine learning aims to classify and predict the quality of blood glucose test strips. Different models are evaluated by the Receiver Operating Characteristic curve and Area Under Curve. Computational results show that the decision tree and random forest after SMOTE perform better than the counterpart of ROSE method. Ensemble learning, such as random forest, out-wins base learner decision tree. To sum up, random forest with SMOTE is the suggested model for accurately predicting the quality of blood glucose test strips. There is a 30% improvement in error rate under random forest and SMOTE for NG class that could be of top concern for prognosis. Several factors, including the direction of applying test reagent onto test strips and the position where the strips are located, that affect quality of test strips have been identified. Explanations in terms of inspection and manufacturing are discussed subsequently. Finally, the prognosis of quality can be attained through big data and statistical machine learning.
- 6 June 2022
- Published by Taylor & Francis in Cogent Engineering
- Vol. 9 (1)
Open AccessQuality prediction through machine learning for the inspection and manufacturing process of blood glucose test strips
Although machine learning for quality prediction of manufacturing processes has attracted attention in the literature, there is a significant lack of case studies from industry, especially in medical sector. This paper proposes a data-driven approach to infer the batch quality of blood glucose test strips. Once the low quality of work in process is detected, unnecessary process waste could be eliminated. Starting from data pre-processing, which consists of Synthetic Minority Over-sampling TEchnique and Random Over-Sampling Example, this project tries to balance the ill-distributed data first. Followed by machine learning aims to classify and predict the quality of blood glucose test strips. Different models are evaluated by the Receiver Operating Characteristic curve and Area Under Curve. Computational results show that the decision tree and random forest after SMOTE perform better than the counterpart of ROSE method. Ensemble learning, such as random forest, out-wins base learner decision tree. To sum up, random forest with SMOTE is the suggested model for accurately predicting the quality of blood glucose test strips. There is a 30% improvement in error rate under random forest and SMOTE for NG class that could be of top concern for prognosis. Several factors, including the direction of applying test reagent onto test strips and the position where the strips are located, that affect quality of test strips have been identified. Explanations in terms of inspection and manufacturing are discussed subsequently. Finally, the prognosis of quality can be attained through big data and statistical machine learning.
- Open Access
- 6 June 2013
- Published by Wiley in Journal of Applied Mathematics
- Vol. 2013, 1-8
Tradeoff Analysis for Optimal Multiobjective Inventory Model
Deterministic inventory model, the economic order quantity (EOQ), reveals that carrying inventory or ordering frequency follows a relation of tradeoff. For probabilistic demand, the tradeoff surface among annual order, expected inventory and shortage are useful because they quantify what the firm must pay in terms of ordering workload and inventory investment to meet the customer service desired. Based on a triobjective inventory model, this paper employs the successive approximation to obtain efficient control policies outlining tradeoffs among conflicting objectives. The nondominated solutions obtained by successive approximation are further used to plot a 3D scatterplot for exploring the relationships between objectives. Visualization of the tradeoffs displayed by the scatterplots justifies the computation effort done in the experiment, although several iterations needed to reach a nondominated solution make the solution procedure lengthy and tedious. Information elicited from the inverse relationships may help managers make deliberate inventory decisions. For the future work, developing an efficient and effective solution procedure for tradeoff analysis in multiobjective inventory management seems imperative.
- 6 June 2013
- Published by Wiley in Journal of Applied Mathematics
- Vol. 2013, 1-8
Open AccessTradeoff Analysis for Optimal Multiobjective Inventory Model
Deterministic inventory model, the economic order quantity (EOQ), reveals that carrying inventory or ordering frequency follows a relation of tradeoff. For probabilistic demand, the tradeoff surface among annual order, expected inventory and shortage are useful because they quantify what the firm must pay in terms of ordering workload and inventory investment to meet the customer service desired. Based on a triobjective inventory model, this paper employs the successive approximation to obtain efficient control policies outlining tradeoffs among conflicting objectives. The nondominated solutions obtained by successive approximation are further used to plot a 3D scatterplot for exploring the relationships between objectives. Visualization of the tradeoffs displayed by the scatterplots justifies the computation effort done in the experiment, although several iterations needed to reach a nondominated solution make the solution procedure lengthy and tedious. Information elicited from the inverse relationships may help managers make deliberate inventory decisions. For the future work, developing an efficient and effective solution procedure for tradeoff analysis in multiobjective inventory management seems imperative.
- Retracted
- 1 July 2010
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 3, 652-655
Notice of Retraction: Approximating tradeoff surfaces for inventory control through evolutionary multi-objective optimization
This article has been retracted by the publisher.
- 1 July 2010
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 3, 652-655
RetractedNotice of Retraction: Approximating tradeoff surfaces for inventory control through evolutionary multi-objective optimization
This article has been retracted by the publisher.
- 13 June 2008
- Published by Taylor & Francis in International Journal of Production Research
- Vol. 46 (14) , 3859-3874
Multi-objective inventory control using electromagnetism-like meta-heuristic
Most of the works on multi-objective inventory control unify the various objectives into a single objective such that lead to a compromise solution whose non-dominance is not guaranteed. This paper presents an algorithm based on Electromagnetism-like Mechanism (EM) to solve a multi-objective inventory control problem with cost and shortage minimization objectives. EM is a new meta-heuristic originated from the electromagnetism theory in physics; it simulates attraction and repulsion of charged particles in order to move towards the optimum. A framework, so called Multi-Objective EM (MOEM), is proposed to approximate the well-known efficient solutions of order size and safety stock without using any surrogate measure (e.g. service level or shortage cost) and prior preference information from decision-makers. To give a specific compromise solution, any outranking method can be implemented to prioritize the non-dominated solutions for decision-makers. Finally, this could be the first attempt to apply EM to multi-objective inventory control, even the inventory control problems.
- 13 June 2008
- Published by Taylor & Francis in International Journal of Production Research
- Vol. 46 (14) , 3859-3874
Multi-objective inventory control using electromagnetism-like meta-heuristic
Most of the works on multi-objective inventory control unify the various objectives into a single objective such that lead to a compromise solution whose non-dominance is not guaranteed. This paper presents an algorithm based on Electromagnetism-like Mechanism (EM) to solve a multi-objective inventory control problem with cost and shortage minimization objectives. EM is a new meta-heuristic originated from the electromagnetism theory in physics; it simulates attraction and repulsion of charged particles in order to move towards the optimum. A framework, so called Multi-Objective EM (MOEM), is proposed to approximate the well-known efficient solutions of order size and safety stock without using any surrogate measure (e.g. service level or shortage cost) and prior preference information from decision-makers. To give a specific compromise solution, any outranking method can be implemented to prioritize the non-dominated solutions for decision-makers. Finally, this could be the first attempt to apply EM to multi-objective inventory control, even the inventory control problems.
- 1 January 2008
- Published by Taylor & Francis in Journal of Statistics and Management Systems
- Vol. 11 (1) , 131-140
Multi-objective analysis of the inventory planning problem using particle swarm optimization
Traditional inventory models only involve single objective that relates to several cost items or service requirements. Even in its multi-objective formulation, most models have been solved with traditional optimization techniques by combining several objectives into a single one. The solutions obtained are unsatisfactory because their non-dominance is not guaranteed. This paper incorporates a local search and clustering mechanism into the multi-objective particle swarm optimization (MOPSO) algorithm to solve two bi-objective inventory planning models, both having a cost minimization objective along with the stockout occasions minimization objective (named as N-model) and the number of items stocked out minimization objective (named as B-model), respectively. The way of multiobjective analysis can determine the non-dominated solutions of order size and safety factor simultaneously. We compare the set coverage metric of both non-dominated solution sets by their expected relevant cost and the service level per order cycle. Our results show that even under the service level measure favorable to the N-model, the non-dominated solution set of the B-model are closer to the Pareto-optimal front than that of the N-model.
- 1 January 2008
- Published by Taylor & Francis in Journal of Statistics and Management Systems
- Vol. 11 (1) , 131-140
Multi-objective analysis of the inventory planning problem using particle swarm optimization
Traditional inventory models only involve single objective that relates to several cost items or service requirements. Even in its multi-objective formulation, most models have been solved with traditional optimization techniques by combining several objectives into a single one. The solutions obtained are unsatisfactory because their non-dominance is not guaranteed. This paper incorporates a local search and clustering mechanism into the multi-objective particle swarm optimization (MOPSO) algorithm to solve two bi-objective inventory planning models, both having a cost minimization objective along with the stockout occasions minimization objective (named as N-model) and the number of items stocked out minimization objective (named as B-model), respectively. The way of multiobjective analysis can determine the non-dominated solutions of order size and safety factor simultaneously. We compare the set coverage metric of both non-dominated solution sets by their expected relevant cost and the service level per order cycle. Our results show that even under the service level measure favorable to the N-model, the non-dominated solution set of the B-model are closer to the Pareto-optimal front than that of the N-model.
- Open Access
- 1 January 2007
- Published by National Library of Serbia in Yugoslav Journal of Operations Research
- Vol. 17 (1) , 95-105
The association between technology type and diffusion pattern under duopolistic competition
Many firms consider adopting new technologies as a means for enhancing competitive advantages. Therefore, the subject of technology diffusion has been studied by many researchers from different disciplines in order to explore the diffusion profiles throughout the industry or the country. The argument has frequently been made that the pattern of diffusion associated with most new technologies will typically have certain characteristics. In general, the diffusion pattern within an industry will depend on the competitive arena and technology characteristics. Based on a duopolistic game-theoretic model, this paper tries to explore the association between technology type and diffusion pattern. The results show that the cost-reducing technologies are adopted sequentially within a duopoly. On the other hard, strategic technologies diffuse over time or in a swarm. Although both technologies might be adopted sequentially, the rate for strategic technologies is faster than that for cost-reducing technologies. .
- 1 January 2007
- Published by National Library of Serbia in Yugoslav Journal of Operations Research
- Vol. 17 (1) , 95-105
Open AccessThe association between technology type and diffusion pattern under duopolistic competition
Many firms consider adopting new technologies as a means for enhancing competitive advantages. Therefore, the subject of technology diffusion has been studied by many researchers from different disciplines in order to explore the diffusion profiles throughout the industry or the country. The argument has frequently been made that the pattern of diffusion associated with most new technologies will typically have certain characteristics. In general, the diffusion pattern within an industry will depend on the competitive arena and technology characteristics. Based on a duopolistic game-theoretic model, this paper tries to explore the association between technology type and diffusion pattern. The results show that the cost-reducing technologies are adopted sequentially within a duopoly. On the other hard, strategic technologies diffuse over time or in a swarm. Although both technologies might be adopted sequentially, the rate for strategic technologies is faster than that for cost-reducing technologies. .
- 1 November 1995
- Published by Taylor & Francis in Production Planning & Control
- Vol. 6 (6) , 530-537
Optimal timing of investments in advanced manufacturing systems under competition
Large initial investment in advanced manufacturing systems (AMS) makes it imperative to develop economic justification methods. Once the new system has been properly justified, a critical issue facing the firm is the optimal timing of the installation in an industry. In this paper we use a game-theoretic model by which the optimal timing of the investments in AMSs is analysed. The existence of timing equilibria is demonstrated and simple decision rules of optimal strategy are provided. Economic interpretations of the decision rules are subsequently made clear. Finally, illustrative examples are used to demonstrate the correctness and usefulness of the analysis.
- 1 November 1995
- Published by Taylor & Francis in Production Planning & Control
- Vol. 6 (6) , 530-537
Optimal timing of investments in advanced manufacturing systems under competition
Large initial investment in advanced manufacturing systems (AMS) makes it imperative to develop economic justification methods. Once the new system has been properly justified, a critical issue facing the firm is the optimal timing of the installation in an industry. In this paper we use a game-theoretic model by which the optimal timing of the investments in AMSs is analysed. The existence of timing equilibria is demonstrated and simple decision rules of optimal strategy are provided. Economic interpretations of the decision rules are subsequently made clear. Finally, illustrative examples are used to demonstrate the correctness and usefulness of the analysis.
- 1 January 1993
- Published by Taylor & Francis in Production Planning & Control
- Vol. 4 (2) , 159-165
A chance-constraints linear programming model on the economic evaluation of flexible manufacturing systems
- 1 January 1993
- Published by Taylor & Francis in Production Planning & Control
- Vol. 4 (2) , 159-165
A chance-constraints linear programming model on the economic evaluation of flexible manufacturing systems