experts in particular fields and the weighing of opin- 

 ions as to predominant probabilities in science and 

 technology. The method of modeling consists of build- 

 ing information models, games models, mathematical 

 models, and other systems of logic incorporating pre- 

 sent and future technical and economic characteris- 

 tics in particular fields of R&D." 



In general, methods of collective expertise and 

 evaluation are most frequently used, particularly 

 when broad or nebulous questions are under examina- 

 tion, such as prospective advance in an area of fun- 

 damental science. Modeling is least used. This is 

 certainly in part due to the heavy demands placed on 

 extensive and consistent data panels and on careful 

 specification of parameters. Modeling is more ame- 

 nable to such tasks as the projection of performance 

 characteristics for certain categories of machinery 

 and the development of branch-of-the-economy fore- 

 casts. 



On the procedural side, the forecasting of R&D of 

 national or interbranch scope is directed and moni- 

 tored by the "Big Four" planning agencies: the USSR 

 Academy of Sciences, the GKNT, Gosplan, and Gosstroy 

 (the USSR State Committee for Construction Affairs) . 

 The Academy and the GKNT, in particular, are the main 

 agencies in this activity. Each maintains an elabo- 

 rate structure of special problem councils and expert 

 groups which separately and jointly conduct forecast- 

 ing studies. S&T forecasting that is limited to an 

 intrabranch focus is the province of relevant branch 

 ministries, though subject to constraints imposed by 

 central forecasts. Branch-wide forecasts selected by 

 the ministries are also submitted to the "big four" 

 agencies for review and approval. ->5 



It needs to be noted, however, that the whole area 

 of scientific forecasting and technology assessment 

 continues to suffer from serious deficiencies. Long- 

 range planning and forecasting are still relatively 

 undeveloped on the branch level. Some branches, Nol- 

 ting points out, do not even bother to draw up fore- 

 casts or perspective plans of intrabranch R&D. Fore- 



112 





