Filatov V.O., Dr.Sci., Prof., Chala O. S.

Kharkiv National University of Radio Electronics

APPROACH TO THE KNOWLEDGE REPRESENTATION OF DISCRETE-EVENT PROCESSES

Discrete-event processes include sequences of actions that change at discrete instants of time when certain events occur. Each event occurs as result of influence of external factors or process actions. Knowledge workers change the process when events occur. They use their experience to choose a rational sequence of actions to complete the task.

Existing approaches to the description of the processes include a set of action sequences. They have a number of drawbacks/

1) Depending on the events that occurred, the sequence of actions may change. However, only the basic workflows are usually included in the process model.

2) Actions consist of elementary operations that can affect its result. This means that the process model consists of several levels. However, usually only one level is considered.

3) The influence of the executors of the process on the sequence of its implementation is not considered. However, performers can change the process based on their experience and their goals [1].

The problem of changing the sequence of actions is as follows. With the existing approaches to constructing the process model, the actions that must be performed to achieve the process goal are usually specified at the stage of the description building. As a result, the change in the description of the process after the introduction of the model is fraught with considerable difficulties. In other words, the traditional model defines the actions to be taken during the implementation of the process, but does not provide an opportunity to determine the actions that can be performed to achieve the process goal.

The problem of insufficient detailing of actions in the model is due to the fact that in practice the actions of performers are generalized in the model. Each action is treated as a separate complex procedure. The performance of the process is affected by the result of the action, and not the details of its implementation. At the same time, each performer performs work at a much more detailed level of individual operations. And these operations can influence the result of the action as a whole.

The problem of incompleteness of information in the process model is due to the fact that part of the context data in the process model construction is simply implied and not considered. However, these contextual data are often used by executors to modify the process [2,3].

If uncontrollable external disturbances occur during the execution of a discrete-event process, then the state of the data necessary for performing operations and actions changes. This requires the adaptation of the structure of the process. Such an adaptation is performed on the basis of knowledge of the interrelations between the context and the actions of the process. Thus, the development of a knowledge representation for the actions of the process is topical.

The proposed approach is based on the integration of the model of such processes with the knowledge graph. The knowledge graph is based on the same ideas as semantic networks. It connects entities and their properties through a set of relationships. The knowledge graph is used for knowledge representation, storage and reasoning [4]. But usually the knowledge graph contains static knowledge that does not reflect the dynamic processes in which the actions are performed in time.

It is suggested to consider the actions of the process as the entities. These actions have temporal and spatial properties, as shown in Figure.

The proposed approach allows us to adapt the process model using the knowledge graph.

Figure – Action structure

References:

1. Gronau, N. (2012). Modeling and Analyzing knowledge intensive business processes with KMDL: Comprehensive insights into theory and practice (English). Gito, 522.

2. Brockmann E. N., Anthony W. P., (1998): The Influence of Tacit Knowledge and Collective Mind on Strategic Planning, Journal of Managerial Issues, X, 2, 204-222.

3. Polanyi, M. (1958) Personal Knowledge: Towards a Post-Critical Philosophy. University of Chicago Press, 493.

4. Pan, J.Z. Exploiting Linked Data and Knowledge Graphs in Large Organisations / J.Z. Pan, G. Vetere, J.M. Gomez-Perez, H. Wu, . – Springer, 2017. – 266 p.