글로버메뉴 바로가기 본문 바로가기 하단메뉴 바로가기

논문검색은 역시 페이퍼서치

대한설비관리학회지검색

JOURNAL OF THE KOREAN INSTITUTE OF PLANT ENGINEERING


  • - 주제 : 공학분야 > 산업공학
  • - 성격 : 학술지
  • - 간기: 계간
  • - 국내 등재 : KCI 등재
  • - 해외 등재 : -
  • - ISSN : 1598-2475
  • - 간행물명 변경 사항 :
논문제목
수록 범위 : 26권 1호 (2021)

야간 조명설비의 조광제어에 의한 소비전력 절감에 관한 연구

김명호 ( Myung Ho Kim )
4,500
초록보기
While local autonomous' night lighting facilities are constantly increasing, control system for them are staying at basic level such as on-off control or a system where every two street lights are turned on, which significantly increases operation and management expense. Thus, this study controlled night lighting facilities by controlling the intensity of illumination to each 0%, 50% and 100% according to the frequency of road usage in the previous controlling system, thereby enabling remote bidirectional control between the administrator and the light pole in terms of lonworks power line communication. As a result, by controlling the intensity of illumination to 0%, 5% and 100% according to frequency of road usage, approximately 40% of energy consumption was saved compared to the previous control system. At the same time, rather than dealing with situations after citizens’ report, it became possible to prevent complaints and safety accidents in advance through remote control.
5,500
초록보기
This paper addresses the relative pre-positioning point of storage/retrieval vehicle to minimize the expected response time when the items in storage bin of the rack in an automated storage/retrieval system are retrieved as the divided batch size. The retrievals for the items begin when the number of the order reaches the plan batch size. The plan batch size is divided as the retrieval batch sizes and the items are retrieved as the retrieval batch size. When a storage/retrieval vehicle is idle within aisle, the vehicle stays at the relative pre-positioning point that minimizes the expected response time. The expected response time is the travel time from the pre-positioning point to the first retrieval point of the plan batch size. When each column of storage location comes to the first retrieval location of each retrieval batch size, the retrieval probability is analyzed. The relative pre-positioning point that minimizes the expected response time is obtained through these retrieval probabilities. The solution procedure is developed, and the numerical examples are shown to explain the solution procedure.

특수기 소포우편 물동량 예측시스템 개발

김은혜 ( Eunhye Kim ) , 정훈 ( Hoon Jung )
5,100
초록보기
In the postal logistics service, rapid change in mail volume and requirements for shortening delivery time make operational optimization a fundamental concern. The mail traffic forecasting is essential for optimizing the resource operation of major postal infrastructures such as post offices, sorting centers, exchange centers, and delivery stations. In particular, in extreme event periods where the volume of traffic increases by about 1.4 times compared to normal, it is important to establish a resource operation plan at the logistics infrastructures for ensuring delivery service quality to the public. This paper addresses an extreme event prediction problem for Korea Post's parcel logistics service. The main purpose of this paper is to describe the development of extreme event prediction system and to introduce an application of the prediction system to postal parcel service site. The extreme event prediction system for Korea Post's parcel logistics service develops a hybrid prediction model based on internal and external factor-based features and ensemble models, and enables flexible prediction in consideration of traffic patterns by applying input features and model selection and parameter adjustment functions.

반도체 EDS 검사 공정의 설비 용량 계획에 관한 최적화 모델

김해중 ( Hae Joong Kim ) , 임대은 ( Dae-eun Lim )
5,200
초록보기
In this paper, a mixed-integer programming (MIP) based capacity planning model for the electrical die sorting (EDS) process where probe cards are used as consumables in semiconductor facilities is presented. The model is designed to satisfy the demand while maximizing the utilization of testers and probe cards in hand. As a result, the level of resource procurement can stay at the minimum level in which directly can be translated into positive profitability. This research employs the decomposition methodology to convert the MIP formulation into a “solver recognizable” format. By doing so, the decomposed model can provide near-optimal solutions instantaneously for practical application and the positive outcome is projected to be over several million dollars.

음식물 쓰레기 수거에서 거래처 방문 일정과 차량 경로 동시 결정 문제

양병학 ( Byoung-hak Yang ) , 김보송 ( Bo-song Kim )
5,200
초록보기
Food waste collection company should decide weekly visiting scheduling and daily vehicle routing for pick up points during a week. A mathematical model to minimize travelling distance was introduced. The mathematical model shows that the characteristics of problem was similar to the travelling sales man and the scheduling problem. Four evolutionary algorithm with penalty cost function to handle constraints were developed to solve our problem. A feasibility search operation was used to increase the feasibility on the solution set. Experimental results show that the average objective value dropped by 9% by using the feasibility search operation, but the calculation time increased by 12%.

지진 위험 지대에서 건물 침하 분석을 위한 GNSS 기반 건물 침하감지의 데이터 활용 침하 경향 분석

김성호 ( Sungho Kim ) , 황상호 ( Sang-ho Hwang ) , 권오언 ( Oeon Kwon ) , 석수영 ( Soo-young ) , 천승만 ( Seung-man Chun )
4,500
초록보기
In this paper, the settlement data was collected using the settlement detector developed in the previous study, and a settlement trend analysis method was presented based on the collected data. The settlement trend analysis method was largely composed of three mathematical steps. The first formula measures an error range based on latitude and longitude, and the second formula determines whether to use the data for the error range. The last equation was analyzed through a method that can filter the distributed altitude information by intermediate values. Through these results, in this paper, it was possible to confirm the information on settlement through three equations and analysis results for the error of the building where the building settlement occurs through the settlement detector developed in the previous study.

인간친화형 홈로봇 UI/UX 개발에 관한 연구

김지관 ( Jiekwan Kim )
5,100
초록보기
In this paper, we proposes a development process that utilizes collective intelligence based on the Kansei engineering product development theory and explains the UI/UX development process of a home robot using it. In order to develop a home robot, it is necessary to first grasp the requirements of the market. However, there is a need to redefine the definition of home robots in a situation where AI speakers, cleaning robots and pet robots are currently classified as home robots. To solve the this problem, we introduce the Delphi method in which experts in the field related to home robot development establish the concept as a group into the function deployment type product development process. And this study exemplifies the entire process of product development from the composition of the project team for home robot development to the establishment of Zero-Concept(0-level Kansei) of the product and the technology development roadmap related to the user experience scenario. It also introduces the process of determining the product development roadmap by arranging the quality characteristics of each target customer through Kano analysis. The case of home robot product concept development will serve as the basis for a product development process that can preoccupy the market by supplying products in a timely manner in line with the market formation in the development of new products that have not formed a market.
1