Research on development trends of transport vehicle structures by forecasting time series of patent data

Abstract

This article presents the main approaches to forecasting developments in freight wagon designs based on the analysis of time series of patent information, which is the most informative source of data for long-term technical forecasting due to its systematic nature, chronological completeness and detailed classification according to the headings of the International Patent Classification. Purpose. The aim of the article is to justify approaches to forecasting the development of freight wagon designs based on patent information, to refine the mathematical framework for forecasting patent activity, and to determine the critical level of innovation quality at which patenting is economically viable. Methods. Modern methods of processing patent information are analysed: thematic modelling using the LDA algorithm, network analysis and time series forecasting using the ARIMA model. An improved econometric procedure for forecasting the number of patent applications is proposed, which takes into account research and development expenditure, intellectual capital, the number of people employed in the scientific sector, and the level of intellectual property rights protection. Results were obtained: a five-factor econometric model of the number of patent applications was developed; relationships were derived for the discounted value of future profits in the presence of a patent; and a critical level of innovation quality was substantiated. Based on data for the Ukrainian railway carriage manufacturing sector, the number of applications stands at ≈409, and as the quality of innovation increases, this figure decreases in accordance with an inverse exponential relationship. Scientific novelty. A shift has been observed from increasing metal content towards composites, modular platforms and digital diagnostic systems. The scientific novelty of the results lies in the development of an improved econometric model for forecasting the number of patent applications, which comprehensively takes into account R&D expenditure, intellectual capital, human resources and the level of intellectual property rights protection, and also enables the critical level of innovation quality to be substantiated. Practical importance of the research lies in the possibility of using the model to formulate scientific and technical strategies for railway wagon manufacturing enterprises and to justify the feasibility of patenting new technical solutions.