In previous sections of the book, the possibility of calculating the physical properties of polymers on the basis on the chemical structure of the repeat unit of linear polymers or the repeating fragment of polymer networks was demonstrated. At the present time, the above-mentioned calculations are performed in a computer. Therewith, main direct and reverse tasks and secondary tasks may be solved.

The direct task concludes in calculating the polymer properties from the data on the chemical structure of the repeat unit of the polymer or a fragment of network.

The reverse task (which is more complicated) consists of predicting the structural formula of the repeat unit of the polymer which would have a single or several required characteristics of the physical properties.

A solution of both tasks is described in refs. [9, 17, 26, 126].

When setting and solving problems, two border approaches and a series of intermediate approaches are possible. The first border approach concludes in recording in computer memory a large number of polymers with different chemical structure for which the physical characteristics mentioned above were previously calculated. Then, computer's role will conclude only in the search for necessary structural formulas in the database and layout of the required information, when both direct and reverse tasks are solved. In the case of such approach, the accuracy of layout characteristics is close to 100%, but the predictive power for structures absent in the database equals zero. In fact, such approach enables just to create a database.

The second...