Jerzy Martyna , Design and Building Problems of Distributed Systems for Time-Constrained Environments,
pp.7-20.

Abstract. In this paper, we present a number of issues that arise in the design and building of  distributed systems for time-constrained environments. These issues include: a guarantee mechanism for time-critical tasks, the possibilities of task execution in the order defined by the precedence graph, non-blocking of resources, and interprocess communication. The building of hard real-time distributed system is related to the implementation of these systems in the  local area networks. Each implementation of the distributed hard real-time  system regards the problem of what must be controlled by the operating system or by the real-time application. In this paper, we also present other issues associated with the implementation of these systems, namely: task scheduling, resource management, real-time synchronization. We examine some examples of the existing distributed systems, describe their structuring and implementation, and compare their principal features.

Jerzy Martyna , Distributed Hard Real-Time System Project,
pp.21-31.

Abstract. An architecture of a distributed system for time-constrained environments is presented. Hardware fault tolerance of this system is attained by the duplication of some elements of computer systems. Data replications, loose coupling, an automated restart capability and some functions resident on a system supervisor node provide the architecture software fault tolerant.

Jerzy Martyna , A Bibliography on Theory and Applications of Distributed Hard Real-Time Systems,
pp.33-58.

Abstract. This bibliography contains a list of references concerned with the distributed hard real-time systems that are either available as published papers, books, technical reports, or on-line. The bibliography is divided in some sections. In each section the list is in alphabetical order by author name(s).

Adrian Horzyk , A Neural Network Used for the Colour Conversion from RGB to CMY,
pp.59-73.

Abstract. The use of the neural network in the solution of the RGB-to-CMY colour conversion problem is discussed in the paper. Classically, a colour conversion   problem is solved in an approximate way. The architecture of neural networks,   which gives the ability to get satisfactory results is presented. The method of learning the network based on the back propagation method with a controlled  process of the change of learning parameters to shorten its time is constructed.

Jarosław Smagłowski , Petri Net as a Tool for Diagnosing Rule-based Expert Systems,
pp.75-89.

Abstract. The paper concerns theoretical and applicational aspects of the so-called Petri Rule Net (PRN) model. The problem of a "rules' infinite looping" is analyzed with the method proposed.

Piotr Wójtowicz, A Syntactic Analysis of IE-graphs for Constructing Expert Systems,
pp.91-106.

Abstract. According to sophisticated concepts of an expert systems' design, complex phenomena/processes should be described at two levels: a semantic one (ie. parameters, measurements, etc) and a structural level. Production rules are suitable to encode knowledge which is then used to reason (at a semantic level) over amorphous data - the so called working memory. However, it is hard to express structural information in pure Rule-Based Knowledge Sources. A way of solving this problem is presented in this paper. An extension of the rule-based approach is proposed, which makes it possible to take into consideration structural information (represented by IE-graphs) during reasoning.

Paweł Oleksik , Syntax-Directed Translation Schema of IE-Graphs Representing Solids,
pp.107-119.

Krzysztof  Janowski, Some Extensions of ETL(1) Grammatical Inference Algorithm,
pp.121-146.

Abstract. The paper concerns theoretical and applicational aspects of an extension of the algorithm of ETL(1) graph grammars inference. This extension let us to make a computer system which automatically creates an ETPL($k$) graph grammar for the set of IE-graphs, and then the system checks whether any IE-graph on its input belongs to the language generated by the grammar.

Mariusz Flasiński , Mark H. Lee, The Use of Graph Grammars for Model-based Reasoning in Diagnostic Expert Systems,
pp.147-165.

Abstract. The new approach to a construction of model- and qualitative reasoning- based expert systems is presented in the paper. The main idea of the approach is the use of formalisms of the theory of graph languages and automata for defining reasoning schemes that process pre-defined graph-like models of complex systems diagnosed. The main advantage of the approach is the efficiency of such reasoning schemes due to the use of the ETPL(k) graph languages with a polynomial membership problem (M.Flasiński,   Theoretical Computer Science, 201 (1998), pp.189-231).