• 한국산학기술학회논문지
• /
• 제10권7호
• /
• pp.1702-1710
• /
• 2009

본 논문은 rework 발생확률을 고려하는 병렬기계 스케줄링 문제를 위해 문제공간기반 탐색 알고리즘을 제안한다. 각 기계와 작업유형별로 rework 발생확률이 존재하며 이것은 자동화된 공정에서 과거데이터로부터 산출가능하다. 스케줄링 문제의 데이터 벡터 (가공시간, 납기, 셋업시간, rework확률)를 교란시킴으로써 이웃해를 생성하고 이로부터 도출된 해는 EDDR이라는 효과적인 휴리스틱을 이용하여 평가한다. 제안된 알고리즘은 납기지연의 최대값과 rewok 발생 작업수로 평가함으로써 제품의 품질과 납기수준을 동시에 고려할 수 있도록 한다.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회/대한산업공학회 2005년도 춘계공동학술대회 발표논문
• /
• pp.506-516
• /
• 2005

This paper deals with an investment scheduling problem of maximizing net present value of dividend with reinvestment allowed, where each investment has certain capital requirement and generates deterministic profit. Such deterministic profit is calculated at completion of each investment and then allocated into two parts, including dividend and reinvestment, at each predetermined reinvestment time point. The objective is to make optimal scheduling of investments over a fixed planning horizon which maximizes total sum of the net present values of dividends subject to investment precedence relations and capital limit but with reinvestment allowed. In the analysis, the scheduling problem is transformed to a kind of parallel machine scheduling problem and formulated as an integer programming which is proven to be NP-complete. Thereupon, a depth-first branch-and-bound algorithm is derived. To test the effectiveness and efficiency of the derived algorithm, computational experiments are performed with some numerical instances. The experimental results show that the algorithm solves the problem relatively faster than the commercial software package (CPLEX 8.1), and optimally solves the instances with up to 30 investments within a reasonable time limit.

• 대한산업공학회지
• /
• 제38권1호
• /
• pp.31-40
• /
• 2012

Distributed arrival time control (DATC) is a distributed feedback control algorithm for real-time scheduling problems in dynamic operational environment. Even though DATC has provided excellent performance for dynamic scheduling problems, it can be improved by considering the following considerations. First, the original DATC heavily depends on the quality of initial solution. In this paper, well-known dispatching rules are incorporated DATC algorithm to enhance its performance. Second, DATC improves its solution with adjusting virtual arrival times of jobs to be scheduled in proportion to the gap between completion time and due date iteratively. Since this approach assigns the same weight to all gaps generated with iterations, it fails to utilize significantly more the latest information (gap) than the previous ones. To overcome this issue we consider exponential smoothing which enable to assign different weight to different gaps. Using these two consideration This paper proposes A-DATC (Advanced-DATC). We demonstrate the effectiveness of the proposed scheduling algorithm through computational results.

• 대한산업공학회지
• /
• 제36권2호
• /
• pp.125-137
• /
• 2010

This study focuses on the problem of scheduling wafer lots of several product families in the deposition workstation in a semiconductor wafer fabrication facility. There are multiple identical parallel machines in the deposition workstation, and two types of setups, record-dependent setup and family setup, may be required at the deposition machines. A record-dependent setup is needed to find optimal operational conditions for a wafer lot on a machine, and a family setup is needed between processings of different families. We suggest two-phase heuristic algorithms in which a priority-rule-based scheduling algorithm is used to generate an initial schedule in the first phase and the schedule is improved in the second phase. Results of computational tests on randomly generated test problems show that the suggested algorithms outperform a scheduling method used in a real manufacturing system in terms of the sum of weighted flowtimes of the wafer lots.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2010년도 춘계학술대회 논문집
• /
• pp.1177-1190
• /
• 2010

The high-speed railway track occurs train operating result track irregularity, subsidence of the track, ballast abrasion. This is the unusual condition. High-speed railway track maintenance task is the behavior which repairs unusual section by using the human resource or machine resource. The resource used to maintenance task is restrictive. A resource can be efficiently used if the high-speed railway track maintenance scheduling is used. So the more task can be performed in the fit time. In conclusion, this manages the unusual condition of a track efficiently. So additional expenses is minimized cause by deteriorating unusual condition. And it offers comfortable ride to passenger. However, maintenance scheduling has to reflect well practical situation and environment. That's maintenance scheduling is used. We gather the opinions of the hands-on workers. So in this paper define field situation and condition. And suggest mathematical model about this. And we developed the track maintenance scheduling software engine using ILOG.

• 한국통신학회논문지
• /
• 제28권1B호
• /
• pp.27-32
• /
• 2003

본 논문에서는 고속 네트웍 기반의 분산 병렬 시스템에서 다양한 내재 병렬 형태를 갖는 프로그램의 효과적인 수행을 위한 유전자 알고리즘 기반의 태스크 스케쥴링 방법(Genetic Algorithm based Task Scheduling GATS)을 제안한다. 분산병렬 시스템은 고속 네트웍을 통해 연결되어진 다수의 범용, 병렬, 벡터 컴퓨터들로 구성되어진다. 분산병렬 처리의 목적은 다양한 내재 병렬 형태를 갖는 연산 집약적인 문제들을 다수의 고성능 및 병렬 컴퓨터들의 각기 다른 능력을 최대한 이용하여 해결함에 있다 분산병렬 시스템에서 스케쥴링을 통하여 더 많은 속도향상을 얻기 위해서는 시스템간의 부하 균형보다는 태스크와 병렬 컴퓨터간의 병렬특성의 일치가 주의 깊게 다루어져야 하며 태스크의 이동으로 인한 통신 오버헤드가 최소화되어야 한다 본 논문에서는 유전자 알고리즘의 동작이 병렬 특성을 감안하여 이루어질 수 있도록 초기화 방법과 지식 기반의 mutation 방법을 제안한다.

• ETRI Journal
• /
• 제45권2호
• /
• pp.318-328
• /
• 2023

Recently, embedded systems, such as mobile platforms, have multiple processing units that can operate in parallel, such as centralized processing units (CPUs) and neural processing units (NPUs). We can use deep-learning compilers to generate machine code optimized for these embedded systems from a deep neural network (DNN). However, the deep-learning compilers proposed so far generate codes that sequentially execute DNN operators on a single processing unit or parallel codes for graphic processing units (GPUs). In this study, we propose PartitionTuner, an operator scheduler for deep-learning compilers that supports multiple heterogeneous PUs including CPUs and NPUs. PartitionTuner can generate an operator-scheduling plan that uses all available PUs simultaneously to minimize overall DNN inference time. Operator scheduling is based on the analysis of DNN architecture and the performance profiles of individual and group operators measured on heterogeneous processing units. By the experiments for seven DNNs, PartitionTuner generates scheduling plans that perform 5.03% better than a static type-based operator-scheduling technique for SqueezeNet. In addition, PartitionTuner outperforms recent profiling-based operator-scheduling techniques for ResNet50, ResNet18, and SqueezeNet by 7.18%, 5.36%, and 2.73%, respectively.

• 정보과학회 컴퓨팅의 실제 논문지
• /
• 제23권9호
• /
• pp.572-577
• /
• 2017

가상화 기술은 물리머신의 자원을 다수의 가상머신이 공유하여 유휴 자원을 활용할 수 있도록 한다. 가상화 기술의 핵심은 가상머신 간의 효율적인 자원 분배에 있다. 그러나 현재의 가상화 환경에서는 가상머신에서 수행되는 작업의 특성을 파악할 수 없기 때문에 효율적인 자원 분배가 어렵다. 비효율적인 자원 분배는 가상머신 간의 성능 간섭을 발생시켜 가상머신의 성능 저하를 야기한다. 가상머신 간 성능 간섭 현상을 해결하기 위해 다양한 연구들이 진행되었다. 본 논문은 가상머신 성능 간섭 현상 해결을 위한 대표적인 방법인 피드백 제어를 활용하여 I/O 성능을 보장하는 CPU 스케줄링 기법을 소개하고, 특히 모델기반 피드백 제어 방식과 동적 피드백 제어 방식을 비교 분석한다.

• 한국정보과학회논문지:시스템및이론
• /
• 제35권9_10호
• /
• pp.417-420
• /
• 2008

우리는 이기적이고 비협조적인 사용자들이 이용하는 시스템의 성능을 최적화하는 문제를 다룬다. 사용자들이 요구하는 작업들은 각각의 속도함수를 가지고 있는 기계들에 스케줄 된다. 여기서 속도함수는 기계에 할당된 작업량에 반비례한다. 시스템의 성능은 기계들이 할당된 작업들의 수행을 끝내는 완료시간의 최대 값으로 평가한다. 이기적인 사용자들은 자신의 작업이 수행될 기계를 고를 수 있고 현재 가장 빠른 기계를 고른다. 그러나 이러한 스케줄은 시스템의 성능을 최적화하지 못한다. 사용자들의 이기적인 행동으로 발생되는 시스템의 성능 저하를 측정하는 기준으로서 Price of Anarchy(PoA)가 소개되었다. 이것은 내쉬 평형의 비용과 최적의 비용의 비율로 정의된다. 이 논문에서 우리는 위 스케줄링 문제에 대한 PoA를 평가한다.

• 한국정밀공학회지
• /
• 제32권1호
• /
• pp.25-37
• /
• 2015

This paper suggests a decision tree based approach for flexible job shop scheduling with multiple process plans. The problem is to determine the operation/machine pairs and the sequence of the jobs assigned to each machine. Two decision tree based scheduling mechanisms are developed for static and dynamic flexible job shops. In the static case, all jobs are given in advance and the decision tree is used to select a priority dispatching rule to process all the jobs. Also, in the dynamic case, the jobs arrive over time and the decision tree, updated regularly, is used to select a priority rule in real-time according to a rescheduling strategy. The two decision tree based mechanisms were applied to a flexible job shop case with reconfigurable manufacturing cells and a conventional job shop, and the results are reported for various system performance measures.