• Journal of Korean Institute of Industrial Engineers
• /
• v.39 no.6
• /
• pp.577-586
• /
• 2013

In a shipyard, it is hard to predict block movement due to the uncertainty caused during the long period of shipbuilding operations. For this reason, block movement is rarely scheduled, while main operations such as assembly, outfitting and painting are scheduled properly. Nonetheless, the high operating costs of block movement compel task managers to attempt its management. To resolve this dilemma, this paper proposes a new block movement analysis framework consisting of the following operations: understanding the entire process, log clustering to obtain manageable processes, discovering the process model and detecting exceptional processes. The proposed framework applies fuzzy mining and trace clustering among the process mining technologies to find main process and define process models easily. We also propose additional methodologies including adjustment of the semantic expression level for process instances to obtain an interpretable process model, definition of each cluster's process model, detection of exceptional processes, and others. The effectiveness of the proposed framework was verified in a case study using real-world event logs generated from the Block Process Monitoring System (BPMS).

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.4
• /
• pp.88-95
• /
• 2022

Detecting anomalies in human movement is an important task in industrial applications, such as monitoring industrial disasters or accidents and recognizing unauthorized factory intruders. In this paper, we propose an abnormal worker movement detection system based on data stream processing and hierarchical clustering. In the proposed system, Apache Spark is used for streaming the location data of people. A hierarchical clustering-based anomalous trajectory detection algorithm is designed for detecting anomalies in human movement. The algorithm is integrated into Apache Spark for detecting anomalies from location data. Specifically, the location information is streamed to Apache Spark using the message queuing telemetry transport protocol. Then, Apache Spark processes and stores location data in a data frame. When there is a request from a client, the processed data in the data frame is taken and put into the proposed algorithm for detecting anomalies. A real mobility trace of people is used to evaluate the proposed system. The obtained results show that the system has high performance and can be used for a wide range of industrial applications.

• Journal of Intelligence and Information Systems
• /
• v.15 no.2
• /
• pp.61-74
• /
• 2009

With an exponential growth of blogs, lots of important data have appeared on blogs. However, since main topics mentioned in blog pages are quite different from general web pages, there are problems which can't be solved by general search engines. Therefore, many researchers have studied searching methods only for blogs to help users who want to have useful information on blog. We also present a blog classifying method based on similarity of titles. First, we analyze blogs and blog search engines to find problems and solution of current blog search. Second, applying our similarity algorithm on blog titles, we discuss a way to develop clustering method only for blog. Finally, by making a prototype system of our algorithm, we evaluate our algorithm's effectiveness and show conclusion and future work. We expect this algorithm could add its power to current search engine.

• KIEE International Transactions on Electrophysics and Applications
• /
• v.5C no.3
• /
• pp.138-142
• /
• 2005

In this paper, we compared recognition rates between NN(neural networks) and clustering method as a scheme of off-line PD(partial discharge) diagnosis which occurs at the stator coil of traction motor. To acquire PD data, three defective models are made. PD data for classification were acquired from PD detector. And then statistical distributions are calculated to classify model discharge sources. These statistical distributions were applied as input data of two classification tools, BP(Back propagation algorithm) and ANFIS(adaptive network based fuzzy inference system) pre-processed FCM(fuzzy c-means) clustering method. So, classification rate of BP were somewhat higher than ANFIS. But other items of ANFIS were better than BP; learning time, parameter number, simplicity of algorithm.

• Journal of Korean Institute of Industrial Engineers
• /
• v.23 no.3
• /
• pp.451-467
• /
• 1997

Tabu search (TS) is a useful strategy that has been successfully applied to a number of complex combinatorial optimization problems. By guiding the search using flexible memory processes and accepting disimproved solutions at some iterations, TS helps alleviate the risk of being trapped at a local optimum. In this article, we propose TS-based heuristics for solving a class of clustering problems, and compare the relative performances of the TS-based heuristic and the simulated annealing (SA) algorithm. Computational experiments show that the TS-based heuristic with a long-term memory offers a higher possibility of finding a better solution, while the TS-based heuristic without a long-term memory performs better than the others in terms of the combined measure of solution quality and computing effort required.

• Journal of Korean Institute of Industrial Engineers
• /
• v.36 no.2
• /
• pp.94-100
• /
• 2010

The reliable multicast MAC (Medium Access Control) protocol is needed to guarantee the recipients' nonerroneous reception of the multicast data frames, which can be transmitted by the AP (Access Point) in infrastructure mode IEEE 802.11 wireless LANs. Enhancing the BMMM (Batch Mode Multicast MAC) protocol, in the literature, the connectivity-based reliable multicast MAC protocol was proposed to reduce the RAK (Request for ACKnowledgement) frame transmissions and enhance the multicast MAC performance. However, the number of necessary RAK frame transmissions increases as the number of multicast recipients increases. To alleviate the problem of the larger number of RAK frame transmissions with the larger number of multicast recipients, we propose the clustering algorithm for partitioning the recipients into a small number of clusters, so that the recipients are connected each other within the same clusters. Numerical examples are presented to show the reliable multicast MAC performance improvement by the clustering algorithm.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.29 no.1
• /
• pp.26-33
• /
• 2006

The potential needs as well as visible needs of customer should be considered in order to research and analyze of the customer data. The methods to analyze customer data is classified into customer segmentation, clustering analysis model, forecasting customer response probability model, analysis of the customer break rate model and new customer analysis model by the purpose. In this study, we developed the CW-CLV (Correlation Weight Customer Lifetime Value)method that used AHP(Analytic Hierarchy Process)rule for enhance the reliability of customer data and quantitative analysis of the customer segmentation, based on CLV(Customer Lifetime Value). We suggest to new variables and methodology from determined CW-CLV coefficients, because all of companies respect to the diversified customers classification and complexity of consumers needs. Finally, we unfolded any company's scheduling added new methodology using simulation and leaded conclusion about the new methodology.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.35 no.3
• /
• pp.187-191
• /
• 2012

The methods for constructing network of observational locations for measuring water quality in water reservoirs have been widely proposed, but they had some limitations to be applied to river areas, which lie in awkward clustering and finding representative observational locations among locations within each clustering. In this paper, a statistical approach to detect anomaly locations which were significantly different in important measurements for the water quality from the previous locations and construct observational network with them was proposed. Anomaly was detected with the sampling distribution of each primary principal component score, sum of primary PCs, or sum of residual PCs. The empirical study with the data of Nakdong Dam for guiding how to use our proposed approach and showing limitations of previous studied was described.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.4
• /
• pp.2370-2375
• /
• 2014

Wireless sensor networks (WSNs) have a big potential for distributed sensing for large geographical area. The improvement of the lifetime of WSNs is the important research topic because it is considered to be difficult to change batteries of sensor nodes. Clustering communication protocols are energy-efficient because each sensor node can send its packet to the cluster head near from itself rather than the sink far from itself. In this paper, we present an energy-efficient clustering communication protocol, which chooses cluster heads based on the expected residual lifetime of each sensor node. Simulation results show that our proposed scheme increases average lifetimes of sensor nodes as much as 20% to 30% in terms of the traffic quantity and as much as 30% to 40% in terms of the scalability compared to the existing clustering communication protocol, LEACH.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.12
• /
• pp.3675-3681
• /
• 2009

One of the most important factors in the lifetime of WSN(Wireless Sensor Network) is the limited resources and static control problem of the sensor nodes. In order to achieve energy efficiency and network utilities, sensor nodes can be well organized into one cluster and selected head node and normal node by dynamic conditions. Various clustering algorithms have been proposed as an efficient way to organize method based on LEACH algorithm. In this paper, we propose an efficient clustering algorithm using information entropy theory based on LEACH algorithm, which is able to recognize environmental differences according to changes from data of sensor nodes. To measure and analyze the changes of clusters, we simply compute the entropy of sensor data and applied it to probability based clustering algorithm. In experiments, we simulate the proposed method and LEACH algorithm. We have shown that our data balanced and energy efficient scheme, has high energy efficiency and network lifetime in two conditions.