• 지능정보연구
• /
• 제17권4호
• /
• pp.213-226
• /
• 2011

글로벌 공급사슬 환경에서 글로벌하게 분산 조달, 생산, 유통하게 됨에 따라 전체 공급사슬의 스케줄을 최적화하기 위해서 공급사슬상의 개별 기업 혹은 공장의 스케줄링 최적화뿐만 아니라 각 개별 기업 혹은 공장의 스케줄을 긴밀하게 연계하는 것이 필요하게 되었다. 이는 경쟁과 협력을 동시에 하는 개별 기업 혹은 공장을 개별 에이전트로 보고 각 에이전트간 커뮤니케이션을 통해 개별 에이전트가 관할하는 스케줄러의 스케줄을 조정함으로써 가능해진다. 하지만 전통적인 스케줄링 연구는 개별 스케줄러의 최적화에 집중되어 있고, 에이전트 연구는 스케줄링 도메인에 적용한 예가 제한적이며 이 예도 개별 스케줄러 내의 최적화에 적용하거나 실제 현장 문제가 아닌 실험실 문제 수준에 그치고 있다. 따라서 본 연구에서는 전체 글로벌 공급사슬 스케줄의 최적화를 위해 개별 기업 혹은 공장 스케줄러의 스케줄링을 연계하는 경쟁협력 스케줄링을 위한 에이전트 기반 플랫폼을 구축하였다. 글로벌 공급사슬에서 경쟁협력 스케줄링을 위한 에이전트 기반 플랫폼을 구축하기 위해 첫째, 경쟁협력 스케줄링 분류 체계를 확립하고, 둘째, 경쟁협력 스케줄링을 위한 에이전트를 설계하고, 셋째, 경쟁협력 스케줄링을 위한 지식기반 의사결정 모델을 개발한 후, 넷째 조선산업에 적용 가능한 프로토타입 시스템을 개발했다. 이를 통해 글로벌 공급사슬상의 전체 스케줄의 품질과 에이전트간 커뮤니케이션의 노력에 대한 균형점을 찾을 수 있다. 이를 통해 공급사슬내 개별 기업 혹은 공장의 부분 최적화를 극복할 수 있는 대안을 제시할 것으로 기대한다.

• Journal of Information Technology Applications and Management
• /
• 제19권1호
• /
• pp.37-46
• /
• 2012

The scheduling problem of large products like ships, airplanes, space shuttles, assembled constructions, and automobiles is very complex in nature. To reduce inherent computational complexity, we often design scheduling systems that the original problem is decomposed into small sub-problems, which are scheduled independently and integrated into the original one. Moreover, the steep growth of communication technology and logistics makes it possible to produce a lot of multi-nation corporation by which products are produced across more than one plant. Therefore vertical and lateral coordination among decomposed scheduling systems is necessary. In this research, we suggest an agent-based coordinating mechanism for multi-level scheduling systems in supply chain. For design of a general coordination mechanism, at first, we propose a grammar to define individual scheduling agents which are responsible to their own plants, and a meta-level coordination agent which is engaged to supervise individual scheduling agents. Second, we suggest scheduling agent communication protocols for each scheduling agent topology which is classified according to the system architecture, existence of coordinator, and direction of coordination. We also suggest a scheduling agent communication language which consists of three layers : Agent Communication Layer, Scheduling Coordination Layer, Industry-specific Layer. Finally, in order to improve the efficiency of communication among scheduling agents we suggest a rough capacity coordination model which supports to monitor participating agents and analyze the status of them. With this coordination mechanism, we can easily model coordination processes of multiple scheduling systems. In the future, we will apply this mechanism to shipbuilding domain and develop a prototype system which consists of a dock-scheduling agent, four assembly-plant-scheduling agents, and a meta-level coordination agent. A series of experiment using the real-world data will be performed to examine this mechanism.

• 대한전기학회논문지
• /
• 제41권8호
• /
• pp.831-840
• /
• 1992

This paper presents an efficient short-term generation scheduling method using a rule-based expert/consulting system approach to assist electric energy system operators and planners. The expert system approach is applied to improve the Dynamic Programming(DP) based generation scheduling algorithm. In the selection procedure of the feasible combinations of generating units at each stage, automatic consulting on the manipulation of several constraints such as the minimum up time, the minimum down time and the maximum running time constraints of generating units will be performed by the expert/consulting system. In order to maximize the solution feasibility, the aforementioned constraints are controlled by a rule-based expert system, that is, instead of imposing penalty cost to those constraint violated combinations, which sometimes may become the very reason of no existing solution, several constraints will be manipulated within their flexibilities using the rules and facts that are established by domain experts. In this paper, for the purpose of implementing the consulting of several constraints during the dynamic process of generation scheduling, an expert system named STGSCS is developed. As a building tool of the expert system, C Language Integrated Production System(CLIPS) is used. The effectiveness of the proposed algorithm has been demonstrated by applying it to a model electric energy system.

• 산업경영시스템학회지
• /
• 제20권41호
• /
• pp.231-242
• /
• 1997

This paper describe an approximation method for solving the minimum makespan problem of job shop scheduling with fuzzy processing time. We consider the multi-part production scheduling problem in a job shop scheduling. The job shop scheduling problem is a complex system and a NP-hard problem. The problem is more complex if the processing time is imprecision. The Fuzzy set theory can be useful in modeling and solving scheduling problems with uncertain processing times. Lee-Li fuzzy number comparison method will be used to compare processing times that evaluated under fuzziness. This study propose heuristic algorithm solving the job shop scheduling problem under fuzzy environment. In This study the proposed algorithm is designed to treat opinions of experts, also can be used to solve a job shop environment under the existence of alternate operations. On the basis of the proposed method, an example is presented.

• Journal of Communications and Networks
• /
• 제12권4호
• /
• pp.346-357
• /
• 2010

Opportunistic scheduling provides the capability of resource management in wireless networks by taking advantage of multiuser diversity and by allowing delay variation in delivering data packets. It generally aims to maximize system throughput or guarantee fairness and quality of service (QoS) requirements. In this paper, we develop an extended proportional fair (PF) scheduling policy that can statistically guarantee three kinds of QoS. The scheduling policy is derived by solving the optimization problems in an ideal system according to QoS constraints. We prove that the practical version of the scheduling policy is optimal in opportunistic scheduling systems. As each scheduling policy has some parameters, we also consider practical parameter adaptation algorithms that require low implementation complexity and show their convergences mathematically. Through simulations, we confirm that our proposed schedulers show good fairness performance in addition to guaranteeing each user's QoS requirements.

• Current Optics and Photonics
• /
• 제2권1호
• /
• pp.39-46
• /
• 2018

Visible-light communication (VLC) combined with advanced illumination can be expected to become an integral part of next-generation communication networks. One of the major concerns in VLC implementation is developing resource-allocation schemes in a multi-user scenario. However, the scheduling for heterogeneous quality of service (QoS) traffic has not been studied so far, for the indoor VLC downlink system. In this paper, we creatively introduce effective-bandwidth and effective-capacity theory into the multi-user scheduling (MUS) problem, to guarantee the user's statistical delay QoS. We also take account of the aggregate interference (AI) in the indoor VLC downlink system, and analyze its impact on the user-centric MUS problem for the first time. Simulations show that the AI has a nonnegligible influence on the scheduling result, and that the proposed scheduling scheme could guarantee the user's QoS requirement under the premise of ensuring sum capacity.

• 한국통신학회논문지
• /
• 제29권5A호
• /
• pp.574-582
• /
• 2004

본 논문에서는 패킷 데이터 서비스를 지원하는 DS/CDMA 시스템에서 역방향 링크의 대기 패킷 버스트들이 겪는 평균 전송 시간을 최소화하기 위한 여러 가지 패킷 스캐줄링 방식을 비교 분석하였다. 모든 활성화된 데이터 사용자들이 동시에 전송을 하는 경우의 최적의 스케줄링 방식에 대하여 논하고 기존의 여러 가지 다른 전송률 스케줄링 방식들과의 수학적인 비교 분석을 통하여 평균 전송 지연을 고려한 전송률 스케줄링 방식의 효율적인 지표를 제시하였다.

• 제어로봇시스템학회논문지
• /
• 제11권2호
• /
• pp.167-174
• /
• 2005

This paper experimentally validates the efficacy of scheduling method that satisfies the performance requirement of periodic data in the Foundation Fieldbus. The scheduling method allocates periodic data traffic to a bandwidth-limited fieldbus medium. The scheduling method generates schedule list that records schedule starting time and schedule period of each periodic data. This study developed an experimental model of a Foundation Fieldbus network system. Using the experimental model, this study showed that the scheduling method can be easily implemented in a practical Foundation Fieldbus network system. The results obtained from the experimental evaluation showed that the scheduling method restricts the delay of periodic data to a pre-specified bound.

• ETRI Journal
• /
• 제29권2호
• /
• pp.201-211
• /
• 2007

This paper focuses on the scheduling problem with the objective of maximizing system throughput, while guaranteeing long-term quality of service (QoS) constraints for non-realtime data users and short-term QoS constraints for realtime multimedia users in multiclass service high-speed uplink packet access (HSUPA) systems. After studying the feasible rate region for multiclass service HSUPA systems, we formulate this scheduling problem and propose a multi-constraints HSUPA opportunistic scheduling (MHOS) algorithm to solve this problem. The MHOS algorithm selects the optimal subset of users for transmission at each time slot to maximize system throughput, while guaranteeing the different constraints. The selection is made according to channel condition, feasible rate region, and user weights, which are adjusted by stochastic approximation algorithms to guarantee the different QoS constraints at different time scales. Simulation results show that the proposed MHOS algorithm guarantees QoS constraints, and achieves high system throughput.

• 대한간호학회지
• /
• 제25권1호
• /
• pp.80-87
• /
• 1995

The critical problem of nurse scheduling in the hospital is determining the day-to -day shift assignments for each nurse for the specified period in a way that satisfies the given requirements of the hospital. As nurse scheduling involves many factors and requirements, manual scheduling requires much time and effort to produce an adequate schedule. A PC - based decision support system, developed in Turbo - C/sup ++/, for nurse scheduling was introduced. The system is composed of 4 sub-systems : 1) Entering basic information for each nursing unit : 2) Generation of an appropriate initial schedule and revised schedule for a given period, maximally satisfying each nurse's duty requests : 3) Provision of variety of statistical information, and 4) Help messages for each modular function. Icons and a mouse are used for easier graphic user interface and reducing the need for typing efforts. This system can help nurses develop quick and easy schedule generation and allow more time for the patient care.