• Journal of Korea Society of Industrial Information Systems
• /
• v.27 no.2
• /
• pp.61-69
• /
• 2022

A multi-agent system is a system that aims at achieving the best-coordinated decision based on each agent's local decision. In this paper, we consider a multi agent-multi task assignment problem. Each agent is assigned to only one task and there is a completion probability for performing. The objective is to determine an assignment that maximizes the sum of the completion probabilities for all tasks. The problem, expressed as a non-linear objective function and combinatorial optimization, is NP-hard. It is necessary to design an effective and efficient solution methodology. This paper presents an approximation algorithm using submodularity, which means a marginal gain diminishing, and demonstrates the scalability and robustness of the algorithm in theoretical and experimental ways.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10b
• /
• pp.85-87
• /
• 2001

다수의 이익-추구형 에이전트들 간의 자동화된 협상 과정을 통해 합리적으로 작업 분배가 이루어질 수 있도록 조정하는 것은 대부분의 다중 에이전트 시스템에서 매우 중요한 문제이다. 본 연구에서는 대표적인 다중 에이전트 작업 재할당 문제의 하나로서 다중 에이전트 외판원 여행 문제를 설명하고, 이 문제를 풀기 위한 조정 메커니즘으로 Vickrey 경매 프로토콜의 적용을 제안하였다. 대표적인 시장기반의 조정 메커니즘인 Vickrey 경매의 적용을 위해 본 논문에서는 각 에이전트의 이익, 협상의 최종 목표, 경매대상, 각 에이전트의 입찰전략, 전체적인 경매 진행과정 등을 정의하였다. 이와 같이 설계된 시장기반의 조정메커니즘의 장점은 각 에이전트의 이익만을 고려함으로써 문제를 단순화할 수 있고, 복잡도가 큰 중앙-집중식 제어 메커니즘이 필요치 않으며, 따라서 각 에이전트의 자율성을 최대한 보장할 수 있고, 큰 협상비용을 들이지 않고 최적의 할당에 도달할 수 있다는 점등이다. 본 논문에서는 하나의 다중 에이전트 외판원 여행문제에 대한 풀이과정을 통해 앞서 설계된 Vickrey 경매프로토콜의 진행방식을 설명하고 그 효과와 효율성을 예시하였다.

• Journal of Korea Society of Industrial Information Systems
• /
• v.27 no.4
• /
• pp.37-45
• /
• 2022

In this paper, a multi agent-multi task assignment problem, which is a representative problem of combinatorial optimization, is presented. The objective of the problem is to determine the coordinated agent-task assignment that maximizes the sum of the achievement rates of each task. The achievement rate is represented as a concave down increasing function according to the number of agents assigned to the task. The problem is expressed as an NP-hard problem with a non-linear objective function. In this paper, to solve the assignment problem, we propose a hybrid cross-entropy algorithm as an effective and efficient solution methodology. In fact, the general cross-entropy algorithm might have drawbacks (e.g., slow update of parameters and premature convergence) according to problem situations. Compared to the general cross-entropy algorithm, the proposed method is designed to be less likely to have the two drawbacks. We show that the performances of the proposed methods are better than those of the general cross-entropy algorithm through numerical experiments.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.1
• /
• pp.27-34
• /
• 2023

The optimal assignment problem between agents and tasks is known as one of the representative problems of combinatorial optimization and an NP-hard problem. This paper covers multi agent-multi task assignment problems with uncertain completion probability. The completion probabilities are generally uncertain due to endogenous (agent or task) or exogenous factors in the system. Assignment decisions without considering uncertainty can be ineffective in a real situation that has volatility. To consider uncertain completion probability mathematically, a mathematical formulation with stochastic programming is illustrated. We also present an algorithm by using the sample average approximation method to solve the problem efficiently. The algorithm can obtain an assignment decision and the upper and lower bounds of the assignment problem. Through numerical experiments, we present the optimality gap and the variance of the gap to confirm the performances of the results. This shows the excellence and robustness of the assignment decisions obtained by the algorithm in the problem with uncertainty.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10c
• /
• pp.502-504
• /
• 2000

최근 높은 컴퓨팅 파워를 요구하는 응용문제 처리를 위해 고 비용의 슈퍼컴퓨터 대신 인터넷상에 분산된 다수의 일반 컴퓨터들을 이용하는 병렬처리에 관한 연구가 진행되고 있다. 본 연구에서는 다중 에이전트 시스템을 기반으로 하여 이질적인 성능을 갖는 컴퓨터들을 병렬 컴퓨팅 환경으로 구성하고 각 호스트의 성능측정 결과에 따라 효율적으로 작업을 분산하는 기법을 제안한다. 또한 본 연구에서 제안한 방법을 다중 에이전트 시스템인 IBM의 AgletsTM을 사용하여 실험적으로 성능을 평가하고, 기존 연구와 비교한다.

• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.4
• /
• pp.411-422
• /
• 2002

Recently, there have been active researches about the VPCS (Virtual Parallel Computing System) based on multiple agents. The PVCS uses personal computers or workstations that are dispersed all over the internet, rather than a high-cost supercomputer, to solve complex problems that require a huge number of calculations. It can be made up with either homogeneous or heterogeneous computers, depending on resources available on the internet. In this paper, we propose a new method in order to distribute worker agents and work packages efficiently on the VPCS based on the IBM Aglets. The previous methods use mainly the master-slave pattern for distributing worker agents and work packages. However, in these methods the workload increases dramatically at the central master as the number of agents increases. As a solution to this problem, our method appoints worker agents to distribute worker agents and workload packages. The proposed method is evaluated in several ways on the VPCS, and its results are improved to be worthy of close attention as compared with the previous ones.

• Journal of KIISE:Computer Systems and Theory
• /
• v.35 no.9_10
• /
• pp.417-420
• /
• 2008

We consider the problem of optimizing the performance of a system shared by selfish non-cooperative users. In this problem, small jobs which the users request should be scheduled on a set of shared machines with their speed functions, each of which dependson the amount of jobs allocated on a machine. The performance of the system is measured by the maximum of the completion times when the machines complete the jobs allocated on them. The selfish users can choose a machine on which their jobs are executed, and they choose the fastest machine. But it typically results in suboptimal system performance. The Price of Anarchy(PoA) was introduced as a measure of the performance degradation due to the user's selfish behavior. The PoA is the worst-case ratio of the cost of a Nash equilibrium to the optimal cost. In this paper, we estimate the PoA for the above scheduling problem.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.1 s.33
• /
• pp.113-124
• /
• 2005

The work of investigation in the laboratory medicine includes various kinds of investigations and microbes and it is too complicated to draw needed results in time. So, we aim to improve work performance of the laboratory medicine. For this study, we implemented the scheduling system in microbe investigation using agent environment and the workflow management system to manage the schedule of investigation, and the query system to check the schedule. And preliminary report and final report of microbe investigation can be announced automatically using agent. The scheduling system implemented could identify the lack or waste of resources and thus enable efficient management and distribution of resources. The query system could check the schedule and retrieve the Processing status in short time, enabled the automated report, and reduced possible interrupts and the delay of work that can be occurred in confirming process. It also enables users to access from local and remote sites. Also, this system can reduce the conflicts among People that may occur in unexpected situations because it enables doctors to confirm those situations such as the malfunction of instrument and the lack of agar or reagent, and the efficiency of work process can be expected.