Tkachenko Sergey А.

Candidate of Sciences (Economics), Associate Professor, Pro-rector of the International Technological University "Mykolaiv Polytechnic", Ukraine

Approaches to build database structures of economic monitoring in the strategic adjustment of the regional structure and territorial organization of the agro-food sector

The effectiveness of the system of strategic control of the regional structure and territorial organization of the agro-food sector data bank function of economic monitoring depends not only on the correct determination of the composition of the database, but also from the relevant organization of its structure, under which, in general, can be understood as a set of basic records, considered in their interconnection. Practice shows that the formation of the structure of the database subsystem economic monitoring should be based: 1. Build on the rules of abstract structures, which are characterized by a generalized representation of information and finding common processes of its transformation, distribution and retrieval of data in the information fund; 2. On the study of real-life relationships (relations) between the different groups and individual performance indicators; 3. To meet the requirements of the user.

In general, all the approaches to the construction of structures of databases in the strategic management of the regional structure and territorial organization of the agro-food sector can be reduced to three basic ones: the hierarchical, network and relational. To answer the question of which of these approaches best meets the requirements of efficient organization of the database function of economic monitoring, will give a brief description of each of them.

A hierarchical approach is still the most common. It is used in many foreign and domestic database systems. Hierarchical approach in many respects provides an adequate simulation of real-life hierarchical structures. The basic principle of this approach is that each item of data is considered and would only make sense in conjunction with the previous hierarchy, without which it can not exist. The spread of the hierarchical approach to a certain extent contributed to the recent use of memory structures in which the data processing based on a purely sequential media storage, and between the model and the structure of the data memory held only minor differences.

A disadvantage of the hierarchical approach is that, firstly, with the number of types of data segments, that is, with increasing complexity of the hierarchy structure, programs realizing user requests significantly complicated, causing additional development time, debugging and maintenance programs. Secondly, reducing economic efficiency in hierarchical structures is due to various types of anomalies associated with the performance of memory operations (addition, deletion, correction). Usually these anomalies occur when memory is necessary to reflect the relationship “many – to many”, that is the relationships that are common to most economic indicators, while the hierarchy of the classical definition of a relationship is implementing “one – to many”.

Network modeling approach provides a more general structures than hierarchical. The main elements of the network data structures are two types of entries, namely, nodes and links. Each node depending on the nature of real-life relationships can have any number of directly subordinate units, and directly elders against him nodes, ie network approach is more likely to meet the requirement of realization of relations between the data of the “many – to many”. This is its advantage over hierarchical approach from this point of view, the use of a network approach in building a database of economic monitoring function is preferred in view of the adequate reflection of networked relationships between economic indicators, which are characterized by attitude “many – to many”.

The main drawback of the network approach is the difficulty of programming in the database due to excessive proximity of the network data model to memory structure and the need for sufficient detailed knowledge of the user numerous chains of links between elements of the database. When the network approach as the hierarchical, barriers remain necessary changes in the database, such as its growth, as this can cause disturbances in the logical database structure, and hence the distortions in the application programs.

More effective in the construction of the database structure of economic monitoring function seems relational approach based on the mathematical theory of relations. According to this theory, the relationship between the data can be presented in the form of two-dimensional tables, which is one of the most natural ways of presenting data to the user-non-programmers. This process is performed step by step for each relationship between the data in the database is called normalization. From a logical point of view, a relational database – a set of two-dimensional tables with the operations of extraction and combination of columns. The flexibility of a relational database is that the algebra of relations – the system of operations used to manipulate relationships allows, if necessary relationship to cut and glue, and so programmers can receive a variety of files in their correct form. With the help of algebra relations can construct a simple and flexible data manipulation language. In addition to flexibility, the merits of relational databases include: ease of working with databases for untrained or inexperienced personnel; accuracy due to the fact that the relations in nature have a more precise meaning, and lend themselves to precise mathematical manipulation techniques using tools such as algebra and calculus relations relations; connectivity as a relational view gives a clear picture of the relationship attributes of different relationships and files; data independence is manifested in the fact that the construction of the base in a normalized form of an independent software reconstruction of the data does not require changes in application programs; ease of implementation, resulting from the fact that the physical location in memory relational flat file structure is much simpler than the placement of hierarchical and network structures, and so on. Abroad, there are already some development of relational databases. Work in this area is carried out, and in our country. To implement the relational database, you can use some of the existing software options.

Thus, when you create a system of strategic control of the regional structure and territorial organization of the agro-food sector database functions economic monitoring the most suitable and reliable in terms of practical implementation at the moment is the network approach. In the future, preference should be given the relational approach, which in the near future will be the predominant.