• 한국품질경영학회:학술대회논문집
• /
• 한국품질경영학회 2004년도 품질경영모델을 통한 가치 창출
• /
• pp.148-151
• /
• 2004

This study presents a new intelligent yield management methodology which can forecast the yield level of a production unit based on features' behaviors. In this proposed methodology, we identify the existing features using C5.0 that are combination of nodes (i.e., variables) in the decision tree generated by C5.0, use SOM(Self-Organizing Map) neural networks in oder to extract the feature's patterns and classify, and then make features' control rules using C5.0.

• 한국정보처리학회논문지
• /
• 제5권11호
• /
• pp.2897-2907
• /
• 1998

학습 예(training examples)로 부터 규칙을 생성하는 문제는 큰 탐색 공간상에서 많은 지역최소치를 가지고 있는 최적화 문제로 귀결되므로 복잡하고 어려운 문제로 알려져 있다. 이러한 생성규칙을 만들기 위한 여러 가지 학습방법들이 제안되었으며, 그 중 한가지 학습방법이 유전자알고리즘을 연산모델로 사용하는 것이다. 그러나 전통적인 유전자알고리즘은 전역해 부근에서 수렴속도가 떨어지고, 추출된 규칙의 효율성에 문제가 있다. 본 논문에서는 유전자알고리즘의 학습과정에서 포착되는 염색체의 스키마를 분석하여 탐색공간을 부분해(subsolution)를 구할 수 있는 공간들로 분할함으로써, 보다 일반화된 분류 규칙집합을 찾는 방법을 제안하였다. 또한, 실험을 통하여 기존의 기계학습 방법을 사용한 경우와 효율을 상호 비교하여 제안한 방법을 타당성을 입증하였다.

• Industrial Engineering and Management Systems
• /
• 제9권3호
• /
• pp.196-203
• /
• 2010

Make-to-forecast production is a way to realize high customization and fast responsiveness. This study firstly investigates the effect of introducing a common component in a make-to-forecast production environment. The common component can eliminate a modification step, which is a major cost component in make-to-forecast production. It is illustrated, however, that introducing a versatile component that merely covers several variants is unattractive, and thus adding values to the common component is inevitable in this environment. Secondly, an order-matching rule under the condition that two partially overlapped delivery lead time intervals exist is proposed. The rule considers the effect of matching orders to units that can cover both intervals. An alternative re-matching rule is also developed and examined. Numerical experiments clarify that the proposed rule generally realizes higher contribution ratio and lower percentages of orphans and rejected orders. The proposed re-matching rule increases the average contribution ratio at the expense of increased orphans and order rejections.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1996년도 하계학술대회 논문집 B
• /
• pp.793-796
• /
• 1996

This study proposes a new algorithm which performs a production simulation under various constraints and maintains computational efficiency. In order to consider the environmental and operational constraints, the proposed algorithm is based on optimization techniques formulated in LP form In the algorithm, "system characteristic constraints" reflect the system characteristics such as LDC shape, unit loading order and forced outage rate. By using the concept of Energy Invariance Property and two operational rules i.e. Compliance Rule for Emission Constraint, Compliance Rule for Limited Energy of Individual Unit, the number of system characteristic constraints is appreciably reduced. As a solution method of the optimization problem, the author uses Karmarkar's method which performs effectively in solving large scale LP problem. The efficiency of production simulation is meaningful when it is effectively used in power system planning. With the proposed production simulation algorithm, an optimal expansion planning model which can cope with operational constraints, environmental restriction, and various uncertainties is developed. This expansion planning model is applied to the long range planning schemes by WASP, and determines an optimal expansion scheme which considers the effect of supply interruption, load forecasting errors, multistates of unit operation, plural limited energy plants etc.

• 대한조선학회논문집
• /
• 제31권3호
• /
• pp.154-164
• /
• 1994

본 연구에서는 경제적이고, 품질이 우수한 용접 조립보의 생산을 위해 요구되는 조건을 컴퓨터 시뮬레이션 방법 및 최적화 방법을 이용해서 찾고자 한다. 이를 위해 용접 조건과 가열 조건 등 조립보의 생산 과정을 지배하는 변수를 선정하고, 이 변수에 따라 조립보 생산비를 추정할 수 있는 모델을 제시하여, 생산에 따른 제한 조건을 만족시키면서, 생산비가 최소가 되는 생산 변수를 구하였다. 더 나아가서 설계 및 생산의 통합 제어를 위해 용접 조립보의 생산에 따른 총 비용을 산출할 수 있는 추정식을 제시하고, 선급 규정에 따라 설계하는 과정까지 포함하여 최적화를 수행하였다. 또한 곡직가열이 불가능한 경우를 가정하여, 가중치법에 기초한 다목적 최적화 기법을 도입하고 최소 비용과 최소 변형의 절충해를 얻는 방법을 제시하였다.

• 반도체디스플레이기술학회지
• /
• 제20권3호
• /
• pp.57-64
• /
• 2021

The domestic semiconductor industry can produce various products that will satisfy customer needs by diversifying assembly parts and increasing compatibility between them. It is necessary to improve the production line as a method to reduce the work-in-process inventory (WIP) in the assembly line, the idle time of the worker, and the idle time of the process. The improvement of the production line is to balance the capabilities of each process as a whole, and to determine the timing of product input or the order of the work process so that the time required between each process is balanced. The purpose of this study is to find the optimal WIP and buffer size through SA (Simulated Annealing) that minimizes lead time while matching the number of two parts in a parallel assembly line with bottleneck process. The WIP level and buffer size obtained by the SA algorithm were applied to the CONWIP and DBR systems, which are the existing production systems, and the simulation was performed by applying them to the new hybrid production system. Here, the Hybrid method is a combination of CONWIP and DBR methods, and it is a production system created by setting new rules. As a result of the Simulation, the result values were derived based on three criteria: lead time, production volume, and work-in-process inventory. Finally, the effect of the hybrid production method was verified through comparative analysis of the result values.

• 한국융합학회논문지
• /
• 제9권3호
• /
• pp.53-59
• /
• 2018

본 논문에서는 스마트 공장에서 의사결정 모델을 이용한 순차 마이닝 기반 제조공정을 제안한다. 제안하는 모델은 소규모의 제조공정에서 순차 마이닝 의사결정 모델을 적용하여 제조 효율을 높이는 방법이다. 제조 단계 중 제품 제조 과정에서 나타나는 데이터를 입력 변수들로 구성하고, 시간당 제조량과 불량률을 출력 변수로 구성한다. t-검정을 통해 유의수준이 높은 변수만을 사용하여 GSP 알고리즘과 REPTree 알고리즘을 이용한 규칙과 모델을 생성한다. 의미있는 순차 규칙과 의사결정 모델은 정확도, 민감도, 특이성, 예측도를 통해 유의미함을 확인한다. 결과적으로, 실제 제조에 적용한 결과 불량률은 0.38%가 개선되었고, 시간당 제조량은 평균 1.89/h 증가되었다. 이는 소규모 제조 공정에서 데이터 마이닝 분석을 통한 제조 효율을 높이기 위한 의미있는 결과를 나타낸다.

• Management Science and Financial Engineering
• /
• 제20권1호
• /
• pp.1-9
• /
• 2014

In multi-objective scheduling problems, the objectives are usually in conflict. To obtain a satisfactory compromise and resolve the issue of NP-hardness, most existing works have suggested employing meta-heuristic methods, such as genetic algorithms. In this research, we propose a novel data-driven approach for generating a single solution that compromises multiple rules pursuing different objectives. The proposed method uses a data mining technique, namely, random forests, in order to extract the logics of several historic schedules and aggregate those. Since it involves learning predictive models, future schedules with the same previous objectives can be easily and quickly obtained by applying new production data into the models. The proposed approach is illustrated with a simulation study, where it appears to successfully produce a new solution showing balanced scheduling performances.

• 전자공학회논문지B
• /
• 제30B권12호
• /
• pp.51-58
• /
• 1993

This paper presents a design of the fuzzy decision maker which infers set value for fuel rate in the rotary kiln of making CaO. The fuzzy decision maker proposed are divided into two groups whose functions are different each other. The one operates when production demand is constant. The other deals with the status of varying production demand. We have chosen several variables used for composing condition and action part by investigating ingerent features of the rotary kiln and skilled operators`manual method of inferring fuel rate. Membership function of each variable was designed by analyzing experimental data and field data collected during two months. On-line operation with fuzzy rules suggested was done safely like human operators' action.

• 한국세라믹학회지
• /
• 제41권12호
• /
• pp.873-879
• /
• 2004

In all major industries ranging from powder industries and advanced ceramics, to the food and pharmaceutical manufacture powder industries, the main production process is the spray dryers. In this paper, a systematic approach is used and six rules are obtained for the basis of the fuzzy model. A fuzzy model is based on the past behavior of the target system and expected to be able to reproduce the behavior of the target system. The output of the developed fuzzy model shows, graphically and statistically, a high level of face validity. Therefore, it is concluded that the developed fuzzy model mimics the actual process and can be considered, with confidence, as a reliable model to study, analyze, and improve the existing process.