Operational efficiency of decentralized Internet auction mechanisms

The recent consumer-to-consumer (C2C) Internet auction boom at eBay, Yahoo, Amazon, and other sites has added new theoretical challenges to the science of auctions. This paper uses experiments with economically-motivated human subjects to measure the operational efficiency of decentralized Internet auction mechanisms as they compare to centralized double auction mechanisms. Subjects are recruited randomly from the undergraduate population of a large urban university to be buyers or sellers in a simulated trading environment. They are randomly assigned costs and values for 10 units of a virtual product. During the experiment subjects trade these units through computer terminals in auctions similar to those held on eBay and generate profits, which subjects receive at the end of the experiment. The paper uses data from this experiment and previous laboratory experiments to compare operational efficiency and convergence pattern of prices to equilibrium levels in continuous double auctions versus online Internet auctions based on two variables: auction size and time. Experimental results suggest that, because of their decentralized nature, Internet auctions require a few more participants and more time to achieve operational efficiency levels than centralized markets which use continuous double auction mechanisms.     

Decentralized fault diagnosis approach without a global model for fault diagnosis of discrete event systems

Diagnosability property ensures that a predefined set of faults are diagnosable by a centralized diagnoser built using a global model of the system, while co-diagnosability guarantees that these faults are diagnosed in decentralized manner using a set of local diagnosers. A fault must be diagnosed by at least one local diagnoser by using its proper local observation of the system. The aim of using decentralized diagnosis approaches is to overcome the space complexity and weak robustness of centralized diagnosis approaches while at the same time preserving the diagnostic capability of a centralized diagnosis. However, co-diagnosability property is stronger than diagnosability property. If a system is co-diagnosable, then it is diagnosable, while a diagnosable system does not ensure that it is co-diagnosable. Therefore, the challenge of decentralized diagnosis approaches is to perform local diagnosis and to verify that it is equivalent to the centralized one without the need for a global model. In this paper, an approach is proposed to obtain co-diagnosable decentralized diagnosis structure of discrete event systems without the use of a global model. This approach is based on the synchronization of local diagnosis decisions in order to solve the ambiguity between local diagnosers. This synchronization allows obtaining local diagnosis equivalent to the global one without the use of a global model.     

An empirical investigation of information technology structure, control and corporate governance

In a study involving 72 large publicly-held organizations, the relationships between information technology (IT) structure, control, and corporate governance were examined. Despite the fact that some have called centralized systems electronic dinosaurs, many companies are turning to this traditional approach because it does offer one thing that decentralized systems generally lack, namely, control. The results indicate that information technology structure (centralized vs. decentralized) is strongly related to management equity ownership and to concentration of stock holding. For example, an organization where top executives and managers own a relatively high percentage of the company's stock possesses a more centralized information technology structure than an organization with a relatively lower percentage of management ownership. This means that user departments in the latter have more control over their IT functions than users in the former. Conversely, companies with a high level of stock concentration have a more decentralized IT structure. In addition, the research revealed a statistically significant relationship between a company's equity to debt ratio and the level of IT centralization.     

A decentralized approach for mining event correlations in distributed system monitoring

Nowadays, there is an increasing demand to monitor, analyze, and control large scale distributed systems. Events detected during monitoring are temporally correlated, which is helpful to resource allocation, job scheduling, and failure prediction. To discover the correlations among detected events, many existing approaches concentrate detected events into an event database and perform data mining on it. We argue that these approaches are not scalable to large scale distributed systems as monitored events grow so fast that event correlation discovering can hardly be done with the power of a single computer. In this paper, we present a decentralized approach to efficiently detect events, filter irrelative events, and discover their temporal correlations. We propose a MapReduce-based algorithm, MapReduce-Apriori, to data mining event association rules, which utilizes the computational resource of multiple dedicated nodes of the system. Experimental results show that our decentralized event correlation mining algorithm achieves nearly ideal speedup compared to centralized mining approaches.     

A futures market-based model for dynamic network resource allocation

Management of telecommunication network requires quick, continuous and decentralized allocation of network bandwidth to various sorts of demands. So as to achieve the efficient network resource allocation, this paper describes a market-based model combining futures market with the agent-based approach. That is, utilization time is divided into many timeslots, and futures markets in hereafter use of bandwidth are opened. In our model, all market participants (software agents) observe only market prices and decide to buy or sell bandwidth trying to maximize their utilities over time so that they can secure enough network resources. The authors discuss network resource allocation through simulation using the proposed model. Masayuki Ishinishi, Ph.D.: He graduated from National Defense Academy in 1995 and 2000. He received the B.E. (1995) and M.E.(2000) degrees in computer science from National Institution for Academic Degrees (NIAD). He received his Ph.D. degree from Tokyo Institute of Technology in 2003. He has been a communications officer at Air Communications and Systems Wing in Japan Air Self-Defence Force (JASDF) since 2003. His research interests include information assurance, agent-based modeling and simulation, multi-agent system and market-based control. He is a member of IEEJ, IPSJ and JSAI. Yuhsuke Koyama, Ph.D.: He received the B.Econ., M.Econ., and Ph.D. degrees in economics from Kyoto University, in 1996, 1998, 2002, respectively. He has been a research associate of Tokyo Institute of Technology since 2002. His research field is evolutionary economics, mathematical sociology and experimental economics. He is a member of JAFEE, JAMS, JASESS and JASAG. Hiroshi Deguchi, Ph.D.: He received his Ph.D. degree in systems science from Tokyo Institute of Technology, in 1986. He also received the Dr. Econ. degree from Kyoto University in 2001. He has been a Professor of Tokyo Institute of Technology since 2002. His research field is evolutionary economics, computational organization theory, agent-based modeling, social system theory, gaming simulation, and philosophy of science. He is a member of SICE, JAMS, IPSJ, PHSC, JASAG and JAFEE. Hajime Kita, Ph.D.: He received the B.E., M.E., and Ph.D. degrees in electrical engineering from Kyoto University, in 1982, 1984, 1991, respectively. He has been a Professor of Kyoto University since 2003, His research field is systems science/engineering, and his research interests are evolutionary computation, neural networks and socio-economic analysis of energy systems, and agentbased modeling. He is a member of IEEJ, IEICE, ISCIE, JNNS, JSER, ORSJ and SICE.     

Implementation of the conversation scheme in message-baseddistributed computer systems

Several different approaches for implementing conversations in message-based distributed computer systems (DCSs) are discussed. Two different exit control strategies (synchronous and asynchronous) and three different approaches to execution of the conversation acceptance test (centralized, decentralized, and semicentralized) are examined and compared in terms of system performance and implementation cost. An efficient approach to run-time management of recovery information based on an extension of the recovery cache scheme is also discussed. The two major types of conversation structures, name-linked recovery block and abstract data type conversations, are examined to analyze which execution approaches are the most efficient for each conversation structure. As a case study, an unmanned vehicle system is used to illustrate how the approaches can be used in a realistic real-time application     

LMI solutions for H-two and H-infinity decentralized controllers applied to an aerothermic process

In this paper, we present a linear matrix inequality (LMI)-based solution to implement H-two and Hinfinity decentralized robust control strategies. Appropriate parametrization of optimal H-two and H-infinity controllers is used. The general formulation of the decentralized control design leads to the optimal determination of both the state feedback gains and the observer gains of the decentralized controllers. This formulation is two folds: first, a centralized controller is obtained, and then, a simplified decentralized solution is derived by optimizing only the observer gains. The mathematical determination of these gains is formulated as an LMI optimization problem that can be easily solved using LMI solvers. As an experimental evaluation of these controllers, a real time application to an aerothermic process is carried out. A continuous-time model of the process obtained with a suitable direct continuous-time identification approach is elaborated. Results illustrating the real performance obtained from the H-two and H-infinity decentralized controllers are discussed and compared with the centralized ones.     

Design of reliable control systems

A methodology for the design of reliable centralized and decentralized control systems is developed. The resulting control systems are reliable in that they provide guaranteed stability and H _∞ performance not only when all control components are operational, but also for sensor or actuator outages in the centralized case, or for control-channel outages in the decentralized case. Reliability is guaranteed provided these outages occur only within a prespecified subset of control components. Strongly stabilizing designs are also developed, for both centralized and decentralized systems     

大系统的分散能控子空间与分散不能观测子空间

提出了大系统的分散能控子空间和分散不能观测子空间的概念.研究了它们分别与集中控制中的能控子空间和不能观测子空间之间的关系.研究中借用了几何控制理论中的方法.结果表明,分散能控子空间和分散不能观测子空间是集中控制中能控子空间和不能观测子空间在分散控制下的自然推广.利用这两个概念,可以从几何角度研究大系统分散控制的几个问题,比如时变分散控制下的系统镇定问题.     

Voltage regulation in distribution grids: A survey

Environmental and sustainability concerns have caused a recent surge in the penetration of distributed energy resources into the power grid. This may lead to voltage violations in the distribution systems making voltage regulation more relevant than ever. Owing to this and rapid advancements in sensing, communication, and computation technologies, the literature on voltage control techniques is growing at a rapid pace in distribution networks. In particular, there is a paradigm shift from traditional offline centralized approaches to distributed ones leveraging increased and varied types of actuators, real-time sensing, fast and efficient computations, and an overall distributed situational awareness. This paper reviews state-of-the-art voltage control algorithms, summarizes the underlying methods, and classifies their coordination mechanisms into local, centralized, distributed, and decentralized. The underlying solution methodologies are further classified into two categories, open-loop and feedback-based. Two specific example workflows are provided to illustrate these solutions for voltage regulation.     

用多Agent系统分配具有启动成本的有限资源

提出了基于多Agent系统分配具有启动成本的有限资源的两种方法：集中式求解方法和分布式求解方法。在分布式求解方法中,给出了分布式的连续双向拍卖协议算法,其Agent采用零智慧增强学习策略。实验结果表明：相对于集中式求解方法,使用分布式求解方法是以降低部分效率为代价的,但是当市场的总需求逐渐接近所有卖者能提供的最大资源数时,市场平均效率呈现逐渐递增的趋势。     

Personal Rapid Transit in an open-control framework

Over the last decade, authorities have begun inquiring about the use of safe, comfortable, ecological vehicles for operation in an urban context as an alternative to private cars. Several on-demand transport projects have emerged with new automated vehicles known as cybercars or Personal Rapid Transit (PRT). Our state-of-the-art survey of the literature about automated On-Demand Transport (ODT) control solutions highlighted the desirability of a decentralized approach, although centralized approaches do have some advantages. In order to benefit from the advantages of both centralized/hierarchical and decentralized/heterarchical control approaches, we propose a new concept of control: open-control. In this paper, the context is intelligent transportation, where vehicles (e.g., PRTs) can be seen as autonomous decisional entities that are part of a transport system. In this context, the open-control concept is used to support two solutions to PRT routing with uncertainty and perturbations. This open-control concept, developed in our lab, exhibits the traditional explicit control, as well as an innovative type of control called implicit control, which allows system entities to be influenced via an Optimization Mechanism (OM). After introducing the open-control paradigm, we illustrate two applications of the implicit control of a PRT fleet, one based on a stigmergic method and the second based on an embedded version of the Dijkstra’s algorithm. We present a real implementation of the second approach applied to an experimental PRT network. We describe our experimental platform for PRT control and report our first experimental results. These experiments clearly show the reactivity of the control faced with unpredictable events, such as path perturbation or dynamic insertion of PRT in the network.     

贝加莱工控产品在污水处理厂自动化系统中的应用

本文介绍了贝加莱紧凑智能型Power Panel和分布式X20 I／O在污水处理厂自动化系统中的应用。基于Ethemet POWERLINK现场总线（一种工业实时以太网）和工控组态软件Power Vision的分布式控制系统,实现了对污水处理厂的自动控制和监控。整个系统采用上位机集中管理和现场控制站分散控制的方式,其上位机与现场站之间组网简单、运行速度快、通信稳定性强,操作可视化程度高,这些特点有利于提高污水处理设备的运行效率和管理水平。运行实践表明该系统具有较高的推广应用价值。     

ATM 网络虚通道资源动态分配的分散控制

建立了ＡＴＭ网络ＶＰ资源分配控制模型,并对其进行了分散化处理,使用动态规划求出分散模型下的控制律,仿真结果表明,尽管在性能上比集中控制稍有下降,但其可实现性和扩展性远远高于集中控制;与简单的反馈方案相比,降低了呼叫损失,提高了网络利用率。     

Nonlinear Model Predictive Formation Flight

This correspondence paper presents the validation of a formation-flight control technique with obstacle avoidance capability based on nonlinear model predictive algorithms. Control architectures for multiagent systems employed in this correspondence paper can be categorized as centralized, sequential-decentralized, and fully decentralized methods. Centralized methods generally have better performance than decentralized methods. However, it is well known that the performance of the centralized methods for formation flight degrades when there exists communication failure among the vehicles, and they require more computation time than the decentralized method. This correspondence paper evaluates the control performance and the computation time reduction of the sequential-decentralized and fully decentralized methods in comparison with the centralized method and shows that the fully decentralized method can be made effective against short-term communication failure. The control inputs for formation flight are computed by nonlinear model predictive control (NMPC). The control input saturation and state constraints are incorporated as inequality constraints using Karush–Kuhn–Tucker conditions in the NMPC framework, and the collision avoidance can be considered in real time. The proposed schemes are validated by numerical simulations, which include the process and measurement noise for more realistic situations.      

The catallaxy approach for decentralized economic-based allocation in Grid resource and service markets

Efficient resource allocation in dynamic large-scale environments is one of the challenges of Grids. In centralized economic-based allocation approaches, the user requests can be matched to the fastest, cheapest or most available resource. This approach, however, shows limitations in scalability and in dynamic environments. In this paper, we explore a decentralized economic approach for resource allocation in Grid markets based on the Catallaxy paradigm. Catallactic agents discover selling nodes in the resource and service Grid markets, and negotiate with each other maximizing their utility by following a strategy. By means of simulations, we evaluate the behavior of the approach, its resource allocation efficiency and its performance with different demand loads in a number of Grid density and dynamic environments. Our results indicate that while the decentralized economic approach based on Catallaxy applied to Grid markets shows similar efficiency to a centralized system, its decentralized operation provides greater advantages: scalability to demand and offer, and robustness in dynamic environments.     

Plural: A decentralized business process modeling method

Top-down and centralized approaches prevail in the design and improvement of business processes. However, centralized structures pose difficulties for organizations in adapting to a rapidly changing business environment. Here we present the Plural method which can be used to guide organizations in performing process modeling in a decentralized way. Instead of a centralized group of people understanding, modeling and improving processes, our method allows individuals to model and improve their own processes to help in fulfilling their roles in the organization. An individual model depicts a set of activities performed by a role, which together result in a cohesive service within the organization. These individual models are then integrated as necessary to show the way the organization works. We applied the Plural method in a case study of a small-size software organization. We describe the method and its underlying principles and then discuss the findings of our case study, lessons learned, and limitations. The study thus provided evidence of Plural's utility and showed how an organization might exploit its strengths.     

基于贝加菜Power Panel和X20 I／O的污水处理厂自动化系统

本文介绍了贝加菜紧凑智能型PowerPanel和分布式X20I／O在污水处理厂自动化系统中的应用。基于EthernetPOWERLINK现场总线（一种工业实时以太网）和工控组态软件PowerVision的分布式控制系统,实现了对污水处理厂的自动控制和监控。整个系统采用上位机集中管理和现场控制站分散控制的方式,其上位机与现场站之间组网简单、运行速度快、通信稳定性强,操作可视化程度高,这些特点有利于提高污水处理设备的运行效率和管理水平。运行实践表明该系统具有较高的推广应用价值。     

分散式MPC经济性能评估

近年来,学术界对集中式模型预测控制 (Model predictive control, MPC) 性能评估进行了广泛的研究. 对于大规模化工过程, 工业现场通常采用分散式MPC的控制结构. 由于各子系统间存在复杂的耦合关系, 针对集中式MPC 的性能评估方法不能客观反映分散式MPC的性能. 本文基于线性矩阵不等式(Linear matrix inequality, LMI)的方法对分散式MPC进行经济性能评估. 首先提出了一种迭代方法求解分散式线性二次型调节器(Linear quadratic regulator, LQR)问题, 该方法显著降低了已有求解方法的保守性. 再利用LQR基准建立了一组随机优化命题对MPC进行经济性 能评估, 评估方法对集中式MPC与分散式MPC均适用, 评估结果可以指导MPC参数调整, 也可以为集中式与分散式MPC结构选择提供重要参考. 通过对重油分馏塔控制问题的仿真验证了本文方法的有效性与应用价值.     

Graph-theoretic approach to controllability and localizability of decentralized control systems

The controllability and the localizability problems are considered under the decentralized information structure using some concepts from graph theory. First of all, the information structure graph is introduced for a decentralized control system based upon local output controllability, matrices between stations when a local output feedback law is applied, and some invariant properties of the graph with respect to local output feedback laws are driven. Then by using these results, necessary and sufficient conditions of controllability under decentralized information structure are obtained. Also obtained are several conditions for localizability, of decentralized control systems, which is newly introduced in the present paper as an extended concept of a class of disturbance localization and decoupling for centralized systems.