In English. Summaries in Estonian

Proceedings of the Estonian Academy of Sciences.



Volume 11 No. 1 March 2005


Challenges for real-time systems engineering. Part 1: State of the art; 3–17

Leo Mõtus, Robertus A. Vingerhoeds and Merik Meriste

Abstract. This two-part paper addresses characteristic properties of time-critical software-intensive systems (e.g. embedded, real-time, remote monitoring and diagnosis systems). These properties lie in the interface between systems engineering and software engineering. The first part of the paper pinpoints existing problems, the second one discusses potential resolution routes. The first part starts from observations of the current systems development process with the focus on stages for specification of requirements and design. The key problem, in addition to the increase of the overall complexity of systems, is the increasing role of emergent (e.g. non-predictable) system behaviour, its causes and methods for maintaining better control over the system emergent behaviour.

Key words: real-time systems, time-critical systems, proactive components, interactive computing.

Challenges for real-time systems engineering. Part 2: Towards time-aware technology; 18–30

Leo Motus, Robertus A. Vingerhoeds and Merik Meriste

Abstract. This part of the paper discusses evolution trends of the theory and technology in time-aware interaction-centred models of computation and in time-aware multiagent systems that foster the emergence of a multidisciplinary environment, capable to support analysis of design decisions at the early development stages of time-critical software-intensive systems.

Key words: real-time systems, time-critical systems, proactive components, interactive computing.

Using agent and UML technologies in modelling organizations: the case of a vehicle theft; 31–45

Raul Savimaa

Abstract. The paper considers modelling the emergent behaviour of human multifunctional organiza­tions. A combined modelling methodology is introduced that unites UML, the Q-model and multi-agent approach. UML is used for modelling the processes, the Q-model for analysing timing characteristics and multi-agent model for simulating interactions between actors. Although this approach needs further refinement, it can already be used for modelling the time-dependent emergent behaviour in organizations. As an example, the case of the theft of a vehicle has been considered.

Key words: modelling of organizations, multi-agent systems, simulation, emergent behaviour, UML.

On micromechanical problems of erosive wear of particle reinforced composites; 46–58

Irina Hussainova

Abstract. Particle reinforced composites is a success story from the viewpoint of their numerous applications. The main objective of this paper is to review the contributions of solid mechanics research to the development of composite materials and structures and to better understanding of the multiphase material behaviour in a tribological environment. The effect of microstructural features on the wear of multiphase materials is analysed for hard carbide particle reinforced composites.

Key words: particle reinforced composites, residual stress, micromechanics, wear.

A method for securing working mining block stability in Estonian oil shale mines; 59–68

Jüri-Rivaldo Pastarus and Sergei Sabanov

Abstract. This paper deals with the stability analysis and design of working mining blocks in Estonian oil shale mines, where the room-and-pillar mining with blasting is used. The pillars are arranged in a singular grid. Calculations are performed using the concepts of critical area and conditional thickness, and the sliding rectangle method. The results are presented in the form of conditional thickness contours. For practical applications, the criterion of the conditional thickness difference is used. It guarantees safety and minimal losses of the mining block. The method is applicable in different geological conditions, where the room-and-pillar mining is used.

Key words: working mining block, room-and-pillar mining, numerical modelling, stability, critical area, conditional thickness.

The relative merits of EEG and MEG; 69–82

Jaakko Malmivuo

Abstract. After the first detection of the magnetoencephalogram, MEG, it was believed that it has important benefits over the electroencephalogram, EEG. We have, however, earlier shown that the planar gradiometer MEG has measurement sensitivity distribution very similar to bipolar EEG. Here we show that, unlike initially believed, MEG is not complementary to EEG but gives only partially independent information. We have also earlier shown that MEG does not have better spatial resolution than EEG. Based on the newest information on the resistivity of the skull, we have recalculated the spatial resolution of EEG. These calculations show that EEG has even better spatial resolution than MEG.

Key words: EEG, MEG, lead field, spatial resolution.

Instructions to authors; 83–85

Copyright Transfer Agreement; 86