Cooperation under uncertainty in distributed expert systems

The causes of deadlocks in distributed expert systems are examined and classified, and two methods of processing them are proposed. In particular, this problem is discussed within the framework of a working prototype distributed expert system, Hecodes (heterogeneous cooperative distributed expert system) which consists of a control system, a blackboard system, and a group of expert systems whose structures and functions can be different. The two approaches for processing any deadlock are termed deadlock avoidance, and deadlock detection and destruction. Only one of the two can be chosen for a particular run of the system. The detailed working and effects of using each of the proposed deadlock resolution methods are discussed.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Application of distributed expert systems to the advanced control of smart houses

A model of horizontal cooperation under uncertainty in distributed expert systems is presented. This model uses the synthesis of solutions under uncertainty and decision-making to fulfil horizontal cooperation. Firstly, a Boolean algebra is used in order to represent the solution synthesis hypothesis space in the model. Then, a generalization of evidence theory is used as a mathematical model to synthesize all conclusions which come from other expert systems. Finally, a competing mechanism is introduced to make the decisions.

In the paper is presented an idea of utilisation of expert systems as an aid in decision-making in the scope of measurement quality assurance in the process of diagnosis and measurement quality control management of complex electronic equipment. The solution of the problem involves cooperation between distributed expert system, distributed measurement & testing net (with test & measurement systems separated in space) and communication software modules. As a result, there is developed and experimentally verified a new tool for computer supported calibration, verification and diagnosis of complex distributed measurement systems.

A distributed expert system that uses a simulation-based optimization methodology for queuing networks and whose architecture permits parallel simulation of multiple configurations is presented. A knowledge-based search that is a randomized combination of steepest descent and branch and bound algorithms in which the generating function of new states uses qualitative reasoning and the gradient of the objective function is estimated using a heuristic score function method is shown to drive the performance of the network. A random search based on the relative order of the performance gradient components is shown to be a powerful qualitative reasoning technique. The system is implemented as a loosely coupled expert system with components written in Prolog, Simscript, and C. The efficacy of the method is demonstrated through an example from the domain of Jackson (1963) queuing networks. >

In this paper, two neural network mechanisms for synthesis of solutions in non-conflict cases in distributed expert systems (DESs) are proposed. The ideas are: inputs of the neural network are different solutions for the same problem from different expert systems in DESs; outputs of the neural network are the final solutions for the problem after combining different solutions which should be the same as the human experts' final solutions. The first point is to set up the architecture of the neural network and train the neural network by adjusting weights of links to match the outputs of the neural network against the human experts' solutions for all patterns. The second point is that the neural network mechanism proposed in this paper can accommodate the variable number of inputs and outputs without changing neural network architecture.

In this paper, a general procedure of distributed problem solving in distributed expert systems (DESs) is formally described, solution synthesis and solution composition in DESs are compared, and the relationship between solution synthesis in DESs and conflict resolutions in distributed AI is analyzed. Furthermore, general methodologies used for solution synthesis in DESs are introduced and compared.

The trains are considered tasks with specified temporal behaviors that have to fulfill their deadlines. The solutions are found with distributed expert systems. The control signals are implemented and verified using temporal Rule Nets, which are a formalism that seeks to express an automatism in a similar way to a human being: “IF conditions THEN actions”. The results obtained through simulations show that the proposed distributed controllers solve the control problems and can be used for large scale implementation. An algorithm for calculate all possible paths is presented.

In the past, expert systems exploited mainly the EMYCIN model and the PROSPECTOR model to deal with uncertainties. In other words, a lot of stand-alone expert systems which use these two models are available. If we can use the Internet to couple them together, their performance will be improved through cooperation. This is because the problem-solving ability of expert systems is greatly improved by the way of cooperation among different expert systems in a distributed expert system. Cooperation between different expert systems with these two heterogeneous uncertain reasoning models is essentially based on the transformations of uncertainties of propositions between these two models. In this paper, we discovered the exactly isomorphic transformations uncertainties between uncertain reasoning models, as used by EMYCIN and PROSPECTOR.

If different expert systems use different uncertain reasoning models in a distributed expert system, it is necessary to transform the uncertainty of a proposition from one model to another when they cooperate to solve problems. This paper looks at ways to transform uncertainties between the EMYCIN model and the Bayesian network. In the past, the uncertainty management scheme employed the most extensively in expert systems was the EMYCIN model. Now the scheme is turning towards the Bayesian network. If we can combine, by means of the Internet, pre-existing stand-alone expert systems that use these two models into a distributed expert system, the ability of these individual expert systems in their real applications will be greatly improved. The work described in this paper is an important step in this direction.

It is an important direction that introducing Distributed Expert System into Decision Support System to form Intelligent Decision Support System in this research field. But Current resolutions does almost not involve in the heterogeneity of Expert System all. For this, the objective of this paper is to study the organizational pattern between Distributed Expert System and Decision Support System, and design an Intelligent Decision Support System supported by heterogeneous Distributed Expert System. Firstly, a unified descriptive method of heterogeneous knowledge is presented. And then, an architecture integrating heterogeneous Distributed Expert System and Decision Support System is provided. Finally, a hierarchical approach to cooperate and manage Expert Systems have different degrees of capability is given. The result of analysis shows that the architecture given in the paper can improve the scope and precision of Intelligent Decision Support System to resolve problems.

The main directions of development of artificial intelligence systems are considered. Fundamental research in this area served as the basis for the development of applied areas of artificial intelligence, one of which is the development of knowledge-based systems, that is, expert systems. It is shown that the further development of intelligent systems of this class lies within the framework of the paradigm of distributed artificial intelligence, the main idea of which is the cooperative interaction of local intelligent agents. The necessity of including a specialized module in the distributed expert system - a coordinating node, which is entrusted with the functions of organizing the interaction of local expert systems in accordance with the logic of cooperative solution of complex problems and ensuring the efficiency of the computing process in a distributed environment is substantiated. Within the framework of the theoretical and logical approach to the consideration of the cooperative interaction of intelligent agents, the types of logical structures for solving problems that can be formed in the process of functioning of a distributed expert system are distinguished.

The TERESA system (Trouble Evaluation and Resolution via Expert System Application) is a distributed expert system that provides automated diagnoses of transmission problems in AT&T’s digital communication network. Its domain of operation is challenging, requiring collection and analysis of large amounts of information from equipment and systems in disparate locations and cooperation among the processors in the distributed system. Through its distributed design, TERESA secures reliable local communications with operations support systems throughout the network and can serve users in a variety of locations. TERESA employs its wide area networking capability to obtain global alarm correlation information and to perform cooperative trouble diagnosis from both ends of a circuit. The trouble shooting application program is an expert system knowledge base consisting of rules garnered from human experts.

Distributed expert systems as a new tool in analytical chemistry

This paper presents a framework for fault diagnosis in distributed expert systems. We investigate computational methods for evidential reasoning in expert systems based on the Guan-Bell approach. Matrices and their products are used to express the reasoning algorithms. This makes the algorithms simpler and perhaps more efficient to implement. To make the description of our method clear, we trace through the analysis of the simplified problem of an example in the context of fault diagnosis in a distributed vehicle monitoring system. >