• Journal of Korean Institute of Industrial Engineers
• /
• v.40 no.4
• /
• pp.404-414
• /
• 2014

Yield prediction is important to manage semiconductor quality. Many researches with machine learning algorithms such as SVM (support vector machine) are conducted to predict yield precisely. However, yield prediction using SVM is hard because extremely imbalanced and big data are generated by final test procedure in semiconductor manufacturing process. Using SVM algorithm with imbalanced data sometimes cause unnecessary support vectors from major class because of unselected support vectors from minor class. So, decision boundary at target class can be overwhelmed by effect of observations in major class. For this reason, we propose a under-sampling method with minor class based SVM (MCSVM) which overcomes the limitations of ordinary SVM algorithm. MCSVM constructs the model that fixes some of data from minor class as support vectors, and they can be good samples representing the nature of target class. Several experimental studies with using the data sets from UCI and real manufacturing process represent that our proposed method performs better than existing sampling methods.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.41 no.2
• /
• pp.56-64
• /
• 2018

In recent years, imbalanced data is one of the most important and frequent issue for quality control in industrial field. As an example, defect rate has been drastically reduced thanks to highly developed technology and quality management, so that only few defective data can be obtained from production process. Therefore, quality classification should be performed under the condition that one class (defective dataset) is even smaller than the other class (good dataset). However, traditional multi-class classification methods are not appropriate to deal with such an imbalanced dataset, since they classify data from the difference between one class and the others that can hardly be found in imbalanced datasets. Thus, one-class classification that thoroughly learns patterns of target class is more suitable for imbalanced dataset since it only focuses on data in a target class. So far, several one-class classification methods such as one-class support vector machine, neural network and decision tree there have been suggested. One-class support vector machine and neural network can guarantee good classification rate, and decision tree can provide a set of rules that can be clearly interpreted. However, the classifiers obtained from the former two methods consist of complex mathematical functions and cannot be easily understood by users. In case of decision tree, the criterion for rule generation is ambiguous. Therefore, as an alternative, a new one-class classifier using hyper-rectangles was proposed, which performs precise classification compared to other methods and generates rules clearly understood by users as well. In this paper, we suggest an approach for improving the limitations of those previous one-class classification algorithms. Specifically, the suggested approach produces more improved one-class classifier using hyper-rectangles generated by using Gaussian function. The performance of the suggested algorithm is verified by a numerical experiment, which uses several datasets in UCI machine learning repository.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.25 no.5
• /
• pp.31-38
• /
• 2002

In this paper, we model a two-server queueing system with priority, to which we put a restriction on the number of servers for each customer class. customers are divided into two different classes. Class 1 customers have non-preemptive priority over class 2 customers. They are served by both servers when available but class 2 customers are served only by a designated server. We use a method of generating function depending on the state of servers. We find the generating function of the number of customers in queue, server utilization, mean queue length and mean waiting time for each class of customers.

• Industrial Engineering and Management Systems
• /
• v.11 no.4
• /
• pp.331-338
• /
• 2012

This research proposes a Pareto-based multi-objective optimization approach to class-based storage warehouse design, considering a two-block warehouse that operates under the class-based storage policy in a low-level, picker-to-part and narrow aisle warehousing system. A mathematical model is formulated to determine the number of aisles, the length of aisle and the partial length of each pick aisle to allocate to each product class that minimizes the travel distance and maximizes the usable storage space. A solution approach based on multiple objective particle swarm optimization is proposed to find the Pareto front of the problems. Numerical examples are given to show how to apply the proposed algorithm. The results from the examples show that the proposed algorithm can provide design alternatives to conflicting warehouse design decisions.

• Journal of Korean Institute of Industrial Engineers
• /
• v.16 no.1
• /
• pp.99-106
• /
• 1990

With two-class-based storage assignment policy and dual command cycle in Automated Storage/Retrieval Systems(AS/RS), the problem of determining the region dedicated for class-one item is considered. First, the expected travel time of the S/R machine is derived when the boundary of the class-one region is square. Secondly, a heuristic procedure is proposed which determines sequentially the class-one region in a discrete rack. An application of the procedure generates leaf shape region which confirms that the L-shape partition is not necessarily optimal.

• Journal of Korean Institute of Industrial Engineers
• /
• v.41 no.5
• /
• pp.439-446
• /
• 2015

Recently, the importance of one-class classification problem is more increasing. However, most of existing algorithms have the limitation on providing the information that effects on the prediction of the target value. Motivated by this remark, in this paper, we suggest an efficient one-class classifier using hyper-rectangles (H-RTGLs) that can be produced from intervals including observations. Specifically, we generate intervals for each feature and integrate them. For generating intervals, we consider two approaches : (i) interval merging and (ii) clustering. We evaluate the performance of the suggested methods by computing classification accuracy using area under the roc curve and compare them with other one-class classification algorithms using four datasets from UCI repository. Since H-RTGLs constructed for a given data set enable classification factors to be visible, we can discern which features effect on the classification result and extract patterns that a data set originally has.

• Proceedings of the KIEE Conference
• /
• 2008.10c
• /
• pp.174-176
• /
• 2008

HYOSUNG manufactured and tested 1kw class PEFC systems to generate electrical and thermal energy for each residential usage. In particular, HYOSUNG developed Power conditioning system that performs about 91% electrical conversion ratio specified in 1kW class PEFC systems. Prior to system integration, we tested each performances of components to derive control issues from it. In addition, we have been developing the adequate simulator to describe and predict system performance. In this paper, we verified HYOSUNG's 1kW class PEFC systems are valid for residential energy sureces by testing the characteristics of systems and performances of main component.

• Journal of Korean Institute of Industrial Engineers
• /
• v.25 no.1
• /
• pp.141-149
• /
• 1999

In this paper, we consider multiple-class queueing systems in which the server starts a set-up as soon as the number of customers in the "start-up class" reaches threshold N. After the set-up the server starts his service. We obtain the Laplace-Stieltjes transform and the mean of the waiting times of each class of customers for FCFS and non-preemptive priority disciplines.

• Journal of Korean Institute of Industrial Engineers
• /
• v.14 no.1
• /
• pp.31-42
• /
• 1988

This paper studies class-based turnover assignment rule in terms of the expected travel time in automated storage/retrieval systems (AS/RS). With the rack face not necessarily in square in time, travel time models of two and three-class systems are developed, from which class partition values are determined for single command cycle. Also, the effects of the system parameters such as the rack shape factor and the skewness of the ABC curve are evaluated on the travel time through example problems.

• Journal of Korean Institute of Industrial Engineers
• /
• v.30 no.3
• /
• pp.190-196
• /
• 2004

In this paper, we address a layout design problem, PTL[3], for determining an optimal 3-class-based dedicated linear storage layout in a class of unit load storage systems. Our objective is to minimize the expected single command travel time. We analyze PTL[3] to derive a fundamental property that an optimal solution to PTL[3] is one of the partitions based on the PAI(product activity index)-nonincreasing ordering. Using the property and partial enumeration, we construct an efficient exact algorithm with O $(n\;{\lceil}\;log\;n\;{\rceil}\;)$ for solving PTL[3].