Search

In this study, a cloud-based real-time building health monitoring and prediction system using AI and IoT sensors was developed. To predict the building condition, which constitutes time-series data, statistical-based ARIMA and AI-based LSTM prediction models were designed, and the effectiveness of the proposed prediction models was experimentally verified using a 1/8-scaled miniaturized structure. The prediction accuracy in terms of MAPE (less than 1%) was experimentally confirmed to be satisfactory. Moreover, a method for analyzing dimensional structure deformation was developed by combining multiple sensor measurements, and its effectiveness was verified through the case study of a real earthquake-damaged building.

Rapid changes in the global economic environment and advancement of converging technology have been accelerating the digitalization of the manufacturing industry. In particular, digitalization is being magnified as one of the major contributors for manufacturers to secure competitive edge — such as better productivity and reduction of transaction cost — as well as the main culprit of gaps in growth. It is drastically changing the paradigm of business management. The manufacturing industry has been striving to cope with these changes by adopting some types of production management systems but having difficulties about how to collect 4M (Machine, man, materials and manufacturing resources) data effectively and how to integrate dispersed information systems following the expansion of production lines as well as the cost of adopting a new system. As the essential components of automation, HMI (Human machine interface) is designed to monitor the operation of equipment and collects the shop floor information. HMI relies heavily on equipment and should be different for different manufacturers as well as within same manufacture; thus inevitably imposing a heavy burden on HMI developers. This paper introduces a common-HMI that is capable of collecting data for heterogeneous equipment in the shop floor and presenting clear, visual information to the user. The common-HMI is reusable for different types of production equipment and can be deployed without any modifications; it is expected to save on system adoption cost and the proposed method can facilitates system expansion and maintenance effort allowing the expansion of the production line. It has proven its outstanding performance including reusability and reduction of system adoption cost in a pilot study wherein it was used for two different types of equipment.

The activities related to exploitation for oil and gas in the Arctic areas increase significantly. In order to transport increased resources in the Arctic areas, large Arctic commercial vessels such as gas carriers, oil tankers, bulk carriers, etc. are needed for mass transportation. In Arctic area, the ice load is the main factor of environmental load acting on Arctic vessel. The ice load is increased with the enlargement of vessel.The largest Arctic commercial vessel was built by DSME in 2016. The vessel was delivered after completion of ice trial in March 2017. The size of Arctic LNG carrier is larger than any other Arctic vessels have been constructed so far. The ice load monitoring system was installed for ice load measurement and structural safety of ice navigation of this large LNG carrier.This paper is concerned with comparison between estimated ice load for structural design and measured ice load for vessel navigation in Arctic area. Design ice load was calculated according to prescriptive rules of the Classification societies. Actual ice load during ice navigation was measured from ice load monitoring system. The arrangement of sensors in the monitoring system was determined for the precise measurement of ice induced loads acting on the hull. FE analyses were also carried out to compare between estimated ice load and measured ice load considering complex structural details in the Arctic LNG carrier.

A product production system consists of multiple fabrication processes which generate information such as the time, status of equipment and operators, and the process information related to the inconsistently changing production line status need to be collected and supervised. Existing data collection systems are expensive, complex, and difficult to extend, maintain, and integrate with high-level systems, when increasing the lines or applying new process. In this paper, we propose a direct equipment interface that is based on an information collection system. The proposed method can obtain information of shop floor using the direct equipment interface such as the PLC and SIB (Sensor-based I/O Board) and be flexibly applied regardless of any changes such as changes of processes or an increase of the lines. The PLC-based interface collects digital information by connecting lines to the facility PMC using a PLC’s digital module; and obtains additional information by connecting various sensors to a PLC’s analog module. The SIB interface collects information using sensors from the general-purpose facilities without a controller. We introduce the case study that develops the information collection system using the direct equipment interface and provides supervising information from shop floor, thereby reducing the cycle time, lead time, and Work-In-Process.

Workers are avoiding production/manufacturing sites due to the poor working environment and concern over safety. Small and medium-sized businesses introduce new equipment to secure safety in the production site or ensure effective process management by introducing the real-time monitoring technique for existing equipment. The importance of real-time monitoring of equipment and process in the production site can also be found in the ANSI/ISA-195 model. Note, however, that most production sites still use paper-based work slip as a process management technique. Data reliability may deteriorate because information on the present condition of the production site cannot be collected/analyzed properly due to manual data writing by the worker. This paper introduces the monitoring and process management technique based on a direct facility interface to secure safety in the field by improving the poor working environment and enhance there liability and real-time characteristics of the production data. Since the data is collected from equipment in real-time directly through the SIB-based interface and PLC-based interface, problems associated with workers` manual data input are expected to be solved; safety can also be improved by enhancing workers` attention to work by minimizing workers` injuries and disruption.

MES (Manufacturing Execution System) is a powerful production management tool supporting production optimization from the process initiation to the final shipment. It not only transmits the production information to production planning systems like ERP (Enterprise Resource Planning) and SCM (Supply Chain Management) in real-time, but also orders the planned production data to the production site. It is a production management system which plans and executes based on the production data in the field, thus it must be operated with a data acquisition system which gathers the accurate field data in real-time. This study deployed the data acquisition system and HMI (Human Machine Interface) system to collect real-time equipment data through the interface with the equipment to develop MES, and the developed system was applied to the machine tool for verification of its effectiveness. The test indicated that the equipment operating time and other operating data of the equipment could be processed without the interference of the workers. The developed data acquisition system and HMI can be applied to other equipment by using different sensors.

VR (virtual reality)-based fire training simulators provide the general public or inexperienced firefighters or commanders with wide-ranging second-hand experience so that they can make prompt decisions and safe and organized responses in actual fire situations. In order to effectively achieve this training goal, it is crucial to reliably express fire dynamics as realistic graphics. In the field of engineering, computational fluid dynamics (CFD) is widely used to precisely predict the behaviors of fluid phenomena. The resultant data, however, have structures and capacities that are not readily applied to real-time virtual reality systems. This study proposes a series of data conversion techniques and a real-time processing framework to develop a fire training simulator on the basis of a precise CFD simulation that is capable of calculating various invisible physical quantities such as toxic gases and heat as well as visible factors such as smoke and flame. By exploiting safety level-based visualization mapping, this study also proposes a new method to intuitively experience dangerous fire environments and perform training and evaluation. Lastly, this study implements a simulator that can undertake simple firefighting activities such as evacuation and rescue in fire situations at road tunnels; the functions and real-time performance of the simulator have been experimentally measured to verify the applicability of the proposed framework.

The MAGLEV (Magnetically Levitated Vehicle) system, which is under commercialization as a new transportation system in Korea, is operated by means of unmanned automatic control system. Therefore the plan of train operation should be carefully established and validated in advance. In general, when making the train operation plan, the statistically predicted traffic data is used. However, traffic wave can occur when real train service is operated, and the demand-driven simulation technology is required to review train operation plans and service qualities considering traffic wave. This paper presents a method and model to simulate the MAGLEV’s operation considering continuous demand changes. For this purpose, we employed the discrete event model which is suitable for modeling the behavior of railway passenger transportation, and modeled the system hierarchically using DEVS (Discrete Event System Specification) formalism. In addition, through the implementation and experiment using DEVSim++ simulation environment, we tested the feasibility of the proposed model and it is also verified that our demand-driven simulation technology could be used for the prior review of the train operation plans and strategies.

Abstract: The manufacturing execution system (MES) for product production handles different production processes according to the product characteristics and different types of data according to the process being considered. For efficiently providing the data pertaining to production equipment to production systems such as the MES, data collection through the equipment interface is required for obtaining the production data pertaining to field equipment. In this paper, a method is proposed for collecting the production data through the equipment interface in order to collect the various types of production-equipment data from the field. The proposed method is applied to a real manufacturing system to verify its efficiency. A more powerful MES can be constructed with a data acquisition system that acquires the status data at the shop-floor level.

There is much information of equipment in shop floor because the manufacturing processes are different as the equipment within the manufacturing process is varied. To provide effective process information to MES and other production systems, the DAS requires an equipment monitoring system that takes into account the characteristics of the equipment on the shop floor. In this study, we proposed some methods for collecting the required information about various equipments on a shop floor. The equipments such as CNC can be interfaced with the DAS by using a PLC-based method and a sensor-based interface board can be used to interface general equipments. The proposed methods can be used to collect information on the shop floor in real-time. Moreover these methods are very adaptive and can be easily modified according to the changes made to the shop floor. The information about a real shop floor acquired by employing these methods is saved in a database and the can be provided to a supervisor and MES so that they are aware of the status of the shop floor.

The processes for manufacturing a product differ depending on the characteristics of the product, and the information used or generated by the processes also varies. To implement a flexible and configurable Manufacturing Execution System (MES), a Data Acquisition System (DAS) that takes into consideration the characteristics of the manufacturing system is required. In this study, we design an information acquisition system that can process the information on equipments of a shop floor in real-time and that is adaptive to the changes in the shop floor. The system has a data parser module for flexible processing of the equipment status, a data mapper module to link the equipment status with a manufacturing process, and an SOA-based data integration module to transmit the processed information to other information systems such as MES and ERP. From the results of pilot study, its maintenance is easy even if new equipment or new manufacturing processes are adopted or if the equipments are rearranged.

For the fatigue design of a large scaled welded structure, the readily accessibility of various engineering tools is required for the static, fatigue, dynamic analysis and parametric study. In this paper, a multi-agent system is suggested for the development of distributed, open and intelligent design system. The developed design system is applied to assess cumulative fatigue damage on the welded bogie frame of railway vehicles. The engineering tools for fatigue durability analysis are chosen as I-DEAS, ANSYS and BFAP, in which iterative analysis is performed automatically caused by geometrical changes of transom support bracket attached in the bogie frame. The prototype of design system is implemented successfully for the assessment of cumulative fatigue damages of the welded bogie frame.

A special purposed program for ship hull strength analysis considering whipping phenomena is developed. In this program, the non-linear hydrodynamic impact force is considered using the momentum slamming theory and the hull girder is modeled as elastic body on the base of Timoshenko`s beam theory. The numerical verifications are conducted in the view points that the effect of slamming impact force, the effect of hydro-elastic formulation, and the effect of various design parameters such as ship speed, wave amplitude, wave length and others. By the application of a real ship design process, the availability of the program is proved. This program has a GUI function for many I/O data process as well as the function to show the 2-D ship motion in the graphic window, and has other available functions for the whipping analysis.

It can be said that the business sector is faced with a range of challenges–a rapidly changing business environment, an increase and diversification of customers- demands and the consequent need for quick response–for having in place flexible management and production info systems. As a matter of fact, many manufacturers have adopted production info management systems such as MES and ERP. Nevertheless, managers are having difficulties obtaining ever-changing production process information in real time, or responding quickly to any change in production related needs on the basis of such information. This is because they rely on poor production info systems which are not capable of providing real-time factory settings. If the manufacturer doesn-t have a capacity for collecting or digitalizing the 4 Ms (Man, Machine, Material, Method), which are resources for production, on a real time basis, it might to difficult to effectively maintain the information on production process. In this regard, this paper will introduce some new alternatives to the existing methods of collecting the 4 Ms in real time, which are currently comprise the production field., {"references":["Industry Directions, \"Reducing risk in semiconductor & electronics by\nadopting current MES\".","Lee, S. W., Lee, J. K., Nam S. J., Park J. K., Monitoring system of\nequipment in shop floor for embodying MES, International conference on\nComputers & Industrial Engineering 40th (CIE40), 2010, pp. 1-6.","Lei Ye and Henry C. B. Chan, \"RFID-based Logistics Control System for\nBusiness-to-Business E-commerce\", Proceedings of the International\nConference on Mobile Business, IEEE Computer Society Press, 2005,\npp. 630-636","MESA, \"MES Functionalities & MRP to MES Data Flow Possibilities\",","Kwark, W. Y., Kim, W. S. and Park, G. D., \"Design and Implementation\nof Equipment Monitoring System for Data Integration,\" Journal of the\nKorea Society of Computer and Information, Vol. 14, No.9, 2009, pp.\n115~126"]}

The increasing complexity of modern buildings, such as high-rise buildings and underground subway stations, has brought about an increase in the scale of potential hazards, which in turn requires changes in firefighter training. Due to cost, time, and safety requirements for the trainee, it is impossible for firefighters in training to experience a real fire. In additional, repeated training is impossible if real fire were used. We proposed a team-based firefighter training platform using the virtual environment for complex buildings. The platform provides training and evaluation of firefighting and mission-based team training. To achieve an immersive virtual environment, VR, AR and haptics technique is adopted. A numerical analysis method is used to model real fire phenomena.

Virtual reality based fire training enables the provision of emergency response training through a computer-created fire environment. Because the accuracy of fire behavior can play a very important role in the training and evaluation process, we can utilize a specialized numeric solver to generate fire dynamics data, and apply scientific visualization techniques to effectively display these data in 3D. With the exception of volume rendering, typical scientific visualization methods pose an obstacle for people to understand and experience fire behavior in an intuitive manner in the virtual reality environment. We present an approach that enables the use of scientific visualization for fire training and evaluation through the real-time volume rendering of various physical quantities in fire dynamics. We found that this enabled users to easily grasp the useful information of a fire, as well as to evaluate safety achievement during training activities.

In recent years, the use of large high-resolution display systems, usually implemented through tiled display walls, has been increasing in many areas of industry. In the virtual reality area, distributed visualization using PC clusters is generally used and it is necessary to design a synchronization procedure and to define appropriate simulation parameters to be synchronized within it. In this work, we developed a distributed visualization module to synchronize tiled scene contents generated by the open-source graphics engine, using the consistent shared-object mechanism which can be interoperable with the application-specific scene graph. We then applied this module to a couple of virtual reality applications using a DLP-Cube based tiled display system in order to evaluate its visualization result and real-time performance.

The manufacturing processes in a product are different according to the characteristics of the product and the information within the process is also varied. To implement the flexible MES, the Data Acquisition System (DAS) considering the characteristics of manufacturing system is needed. In this paper, we introduced the design of the equipment information acquisition system that can process the information of shop floor in real-time and adaptive to the changes of shop floor. The proposed system is the bridge between the shop floor and the Manufacturing Execution System (MES). It provides the MES the shop floor information between the work order and the results of work performance. The proposed system provides a data parser module to process equipment status in a flexible manner and a data mapper module to couple the equipment status to a manufacturing process and a SOA-based data integration module to link the processed shop floor information to upper system such as MES and ERP. The system is highly configurable system and easy to maintain even if new equipment or manufacturing process is inserted or re-arranged

The manufacturing execution system (MES) for product production handles different production processes according to the product characteristics, and different data according to the process in question. To efficiently provide the data of the production equipment to production systems like the MES, data collection by equipment interface is needed in consideration of the production data concerning the field equipment. This paper proposes a method of collecting production data through the equipment interface in order to collect the various types of production equipment data in shop floor. Production data can be collected by various interface systems such as the PLC-based interface, the sensor-based I/O interface, and the operator key-in interface. The data collection system secondarily processes the production data collected by the equipment interface using an H/W communication module and an operator key-in module, and can increase the flexibility of the collected data using a data parser.

In order to simulate and visualize natural phenomena, especially fluid behavior such as smoke and fire, many novel studies have recently been conducted. Usually these methods use CFD (computational fluid dynamics), which calculate Navier-Stokes equations in real-time to generate realistic fluid motion and interactions, as well as high-performance GPU technologies. We proposed a new approach to the visual simulation of fluid flow by combining the use of pre-calculated CFD data with the real-time processing of such data. As the domain-specialized CFD solver predicts detailed fluid dynamics to an accuracy of a guaranteed error range, we could provide nearly actual behaviors of a fire-driven fluid flow. Moreover, this CFD data includes physical quantities such as temperature distribution, which can provide useful information to the training evaluation process. However, the data-driven method requires appropriate data processing techniques to create and manage large data sets. In this study, we developed a firefighter training simulator to demonstrate our proposed methods and explore related research issues.

In order to improve productivity and quality of the products, there are many efforts to integrate and automate design/analysis processes based on various distributed engineering resources. KIMM (Korea Institute of Machinery and Materials) constructed SOA (Service Oriented Architecture)-based e-Engineering framework for flexible integration and automation of the engineering processes on the Web services environment. However, guaranteeing reliability in the engineering process is one of the new problems that must be solved in the e-Engineering framework. This paper presents two approaches to handle various exceptions that happen in the middle of the execution of the process. One is to introduce recovery activities into the engineering process modeling to handle exceptions caused by failures of the engineering resources; the other is to use log-based recovery techniques to recover from the system exceptions such as network failures or system crashes.

Product development in mechanical industry can be defined as an engineering process based on the collaboration between different experts using various engineering services. The e-Engineering framework aims to integrate engineering services and mediate between them for the processing of the engineering process. The collaboration within the framework is very similar to a business process, in terms of web service composition. However, the existing web service composition system shows some limitations when dealing with engineering services. This paper presents a QoS-based web service orchestration system for the e-Engineering framework by analyzing the characteristics of engineering services which usually perform numerical analysis. The characteristics of the services are expressed in the form of constraint and selection specification.

The Finite Element (FE) modeling support system enables easy and fast FE modeling to be conducted for static and durability analysis in the design of automobile suspension modules. It takes into account the entire design process verifying the index of fatigue durability, which is used to satisfy design criteria. To perform process automation and integrate the FE modeling support system, an integrated design system is implemented based on the JADE (Java Agent DEvelopment) platform to meet the standard specifications of FIPA (Foundation for Intelligent Physical Agent) and has various types of agents. As a case study, the design of the control arms in an automobile suspension module, several main independent parameters, i.e. width, height, fillet radius and thickness, are changed via PSG (Parametric Set Generator) for automatic generation of a parametric model, and the index of fatigue durability is investigated due to subsequent changes of these parameters. In comparison with a conventional fatigue durability analysis, this integrated design system brings a reduction of man-hours of as much as 27%.

A modeling support system that provides an easy and fast FE-modeling for the static and durability analysis of the lower arm in an automobile suspension module is described. The system can exclude human interactions such as meshing using pre-defined shell elements and can impose complicated boundary and loading conditions via a multi-point constraints technique. The platform of this system is implemented based on MSC.PATRAN, and MSC.NASTRAN and MSC.FATIGUE are utilized as the solver. The parametric geometry modeling tool utilizing CATIA V5 is linked with the modeling support system, which enables changes in 3D models, that occur frequently in an early stage of the design process, to be taken into account. The integration techniques using XML-based wrapper enable to integrate the overall process. This system was applied to an actual work process to calculate the fatigue durability of the lower arm in an automobile suspension module, and time and cost reduction for this analysis were validated.

Product development is a complex activity that involves the integration of design/analysis processes and distributed resources such as human beings, engineering tools, and data. This paper presents a multi-agent-based engineering framework and engineering service portal to support such complex engineering design problems. The engineering framework integrates various engineering processes such as design and analysis, which involve complex engineering processes, huge amounts of information processing and problem solving. The web-based engineering service portal provides a user interface for the engineering framework and shares the results between engineers who are participating in the engineering problem. Several engineering problems are applied to the framework and its usefulness is validated.

A web-based Computer-Aided Engineering (CAE) supporting system is integrated with a Product Data Management (PDM) system for Small and Medium Enterprises (SMEs) suffering from the lack of building hardware, software and related experts. An analysis of current business models and worksite requirements provides an improved process model and data to be shared between the PDM system and the CAE supporting system. Since all engineering tasks such as geometric modeling, mesh generation, static stress and vibration analysis, and fatigue durability analysis are automated in the CAE supporting system, the user in charge of the CAE have only to configure the concerned values of design variables and result data through the web page. The existing Change Management module of the PDM system is modified for seamless data exchange, i.e. sending the Engineering Change Order (ECO) data to the CAE supporting system and receiving the CAE result data back. The bi-directional data transfers between the PDM system and the CAE supporting system is made possible by adaptors based on the Simple Object Access Protocol (SOAP). The current approach will be very helpful for SMEs that only have the PDM system and have no adequate infrastructure for CAE.Copyright © 2007 by ASME

This paper presents the applicability of service-oriented architecture to an e-engineering framework. The existing e-engineering framework models product development as engineering processes based on the collaboration of different experts who are participated in the development. And it aims to provide an integrated design environment to support integrating personnel, design activities and engineering resources during product development process. The applicability of the service-oriented architecture and web services technologies enable the framework to utilize and integrate effectively various engineering resources on the geographically distributed computing environments. And intelligent agent technologies support cooperation and coordination mechanisms for the engineering resources and services. In this paper we describe an implementation of the framework in detail and the application of it to the performance evaluation system for centrifugal pumps as a case study.