Data–Driven Technology Foresight: Integrating ARIMA Forecasting and Network Analysis for Patent–Based Roadmapping in Korea’s Railway Industry

Authors

  • Yong-Jae Lee *

    Ai Lab, Cloud Group, TOBESOFT, Seoul 06083, South Korea

  • Tae-Seong Lee

    Department of Business, Chung-Ang University, Seoul 06974, South Korea

DOI:

https://doi.org/10.55121/tdr.v3i1.572

Keywords:

Technology Roadmapping, Technology Forecasting, ARIMA Time-Series Analysis, Social Network Analysis, Railway Industry

Abstract

Traditional expert–driven technology foresight is often limited by subjectivity, low replicability, and high costs. To address these challenges, this study proposes a novel, data–driven framework for technology foresight that offers an objective, scalable, and multi–dimensional approach to identifying and prioritizing emerging technologies in the railway industry. Our tripartite methodology systematically integrates three analytical dimensions: (1) Temporal Forecasting using Autoregressive Integrated Moving Average (ARIMA) models to project the growth trajectories of technology keywords; (2) Structural Analysis using Social Network Analysis (SNA) to evaluate the systemic importance and influence of technologies within the innovation network; and (3) Semantic Analysis using BERT–based contextual embeddings to track the longitudinal evolution of technological concepts. By applying this framework to 4,199 patents from the Korean Intellectual Property Office (1990–2023), we demonstrate its efficacy. The results identify a portfolio of high–priority technologies, such as “sensor,” “signal,” and “control,” which exhibit both strong growth momentum and high network centrality. Critically, our semantic drift analysis reveals a significant paradigm shift, exemplified by
the term “device,” which has evolved from representing simple mechanical components to denoting complex digital control systems. This integrated framework provides a robust, transparent, and replicable methodology for strategic technology roadmapping, offering actionable intelligence that moves beyond static analysis to capture the dynamic nature of technological evolution.

References

[1] Pietrobelli, C., Puppato, F., 2016. Technology foresight and industrial strategy. Technological Forecasting and Social Change. 110, 117–125. DOI: https://doi.org/10.1016/j.techfore.2015.10.021

[2] Khabarov, V.I., Volegzhanina, I.S., 2020. Digital Railway as a precondition for industry, science and education interaction by knowledge management. IOP Conference Series: Materials Science and Engineering. 918(1), 012189. DOI: https://doi.org/10.1088/1757–899X/918/1/012189

[3] Phusakulkajorn, W., Núñez, A., Wang, H., et al., 2023. Artificial intelligence in railway infrastructure: current research, challenges, and future opportunities. Intelligent Transportation Infrastructure. 2, liad016. DOI: https://doi.org/10.1093/iti/liad016

[4] Love, P.E.D., Zhou, J., Matthews, J., et al., 2018. Managing rail infrastructure for a digital future: Future–proofing of asset information. Transportation Research Part A: Policy and Practice. 110, 161–176. DOI: https://doi.org/10.1016/j.tra.2018.02.014

[5] Meyer, T., Von Der Gracht, H.A., Hartmann, E., 2022. Technology foresight for sustainable road freight transportation: Insights from a global real‐time Delphi study. FUTURES & FORESIGHT SCIENCE. 4(1), e101. DOI: https://doi.org/10.1002/ffo2.101

[6] Terk, E., Hunt, T., Saar, I., 2025. Future shifts in the competitiveness of freight transport: a Delphi method–based study. Estonian Discussions on Economic Policy. pp. 192–214. DOI: https://doi.org/10.15157/TPEP.V32I1–2.25523

[7] Xie, J., Huang, J., Zeng, C., et al., 2020. Systematic Literature Review on Data–Driven Models for Predictive Maintenance of Railway Track: Implications in Geotechnical Engineering. Geosciences. 10(11), 425. DOI: https://doi.org/10.3390/geosciences10110425

[8] Moretto, S.M., Palma, A.P., Moniz, A.B., 2012. Constructive technology assessment in railway: The case of high–speed train industry. International Journal of Railway Technology. 1(3), 73–95.

[9] Lee, Y.–J., Han, Y.J., Kim, S.–S., et al., 2024. Identifying the Technology Opportunities and the Technology Taxonomy for Railway Static Inverters With Patent Data Analytics. IEEE Access. 12, 17389–17403. DOI: https://doi.org/10.1109/ACCESS.2024.3360138

[10] Kwon, K., Jun, S., Lee, Y.–J., et al., 2022. Logistics Technology Forecasting Framework Using Patent Analysis for Technology Roadmap. Sustainability. 14(9), 5430. DOI: https://doi.org/10.3390/su14095430

[11] Feldhans, R., Wilke, A., Heindorf, S., et al., 2021. Drift Detection in Text Data with Document Embeddings. In: Yin, H., Camacho, D., Tino, P., et al. (eds.). Intelligent Data Engineering and Automated Learning – IDEAL 2021, Lecture Notes in Computer Science. Springer International Publishing: Cham, Switzerland. pp. 107–118. DOI: https://doi.org/10.1007/978–3–030–91608–4_11

[12] Archibugi, D., 1992. Patenting as an indicator of technological innovation: a review. Science and Public Policy. 19(6), 357–388. DOI: https://doi.org/10.1093/spp/19.6.357

[13] Chakraborty, M., Byshkin, M., Crestani, F., 2020. Patent citation network analysis: A perspective from descriptive statistics and ERGMs. PLOS ONE. 15(12), e0241797. DOI: https://doi.org/10.1371/journal.pone.0241797

[14] Song, K., Kim, K., Lee, S., 2018. Identifying promising technologies using patents: A retrospective feature analysis and a prospective needs analysis on outlier patents. Technological Forecasting and Social Change. 128, 118–132. DOI: https://doi.org/10.1016/j.techfore.2017.11.008

[15] Sokmen, N., Petrov, A.E., 2021. Forecasting of Radar Technologies Using Patent Information. In Proceedings of the 15th International Conference on Application of Information and Communication Technologies (AICT), Baku, Azerbaijan, 13 Oectober 2021; pp. 1–5. DOI: https://doi.org/10.1109/AICT52784.2021.9620514

[16] Linares, I.M.P., De Paulo, A.F., Porto, G.S., 2019. Patent–based network analysis to understand technological innovation pathways and trends. Technology in Society. 59, 101134. DOI: https://doi.org/10.1016/j.techsoc.2019.04.010

[17] Li, S., Garces, E., Daim, T., 2019. Technology forecasting by analogy–based on social network analysis: The case of autonomous vehicles. Technological Forecasting and Social Change. 148, 119731. DOI: https://doi.org/10.1016/j.techfore.2019.119731

[18] Sarica, S., Luo, J., Wood, K.L., 2020. TechNet: Technology semantic network based on patent data. Expert Systems with Applications. 142, 112995. DOI: https://doi.org/10.1016/j.eswa.2019.112995

[19] Kalaivani, E.R., Marivendan, E.R., 2021. The effect of stop word removal and stemming in datapreprocessing. Annals of the Romanian Society for Cell Biology. 25(6), 739–746. Available from: http://annalsofrscb.ro/index.php/journal/article/view/5490 (cited 24 March 2025).

[20] Jun, S., Sung Park, S., Sik Jang, D., 2012. Technology forecasting using matrix map and patent clustering. Industrial Management & Data Systems. 112(5), 786–807. DOI: https://doi.org/10.1108/02635571211232352

[21] Chen, D.–Z., Lin, C.–P., Huang, M.–H., et al., 2012. Technology Forecasting Via Published Patent Applications and Patent Grants. Journal of Marine Science and Technology. 20(4). DOI: https://doi.org/10.6119/JMST–011–0111–1

[22] Xue, D., Hua, Z., 2016. ARIMA Based Time Series Forecasting Model. Recent Advances in Electrical & Electronic Engineering (Formerly Recent Patents on Electrical & Electronic Engineering). 9(2), 93–98. DOI: https://doi.org/10.2174/2352096509999160819164242

[23] Gilsing, V., Nooteboom, B., Vanhaverbeke, W., et al., 2008. Network embeddedness and the exploration of novel technologies: Technological distance, betweenness centrality and density. Research Policy. 37(10), 1717–1731. DOI: https://doi.org/10.1016/j.respol.2008.08.010

[24] Harrigian, K., Dredze, M., 2022. The Problem of Semantic Shift in Longitudinal Monitoring of Social Media: A Case Study on Mental Health During the COVID–19 Pandemic. In Proceedings of the 14th ACM Web Science Conference 2022, Barcelona Spain, 26 June 2022; pp. 208–218. DOI: https://doi.org/10.1145/3501247.3531566

[25] Miaschi, A., Dell’Orletta, F., 2020. Contextual and Non–Contextual Word Embeddings: an in–depth Linguistic Investigation, In the Proceedings of the 5th Workshop on Representation Learning for NLP, Association for Computational Linguistics, Online, 17 April; pp. 110–119. DOI: https://doi.org/10.18653/v1/2020.repl4nlp–1.15

[26] Giorgi, S., Nguyen, K.L., Eichstaedt, J.C., et al., 2022. Regional personality assessment through social media language. Journal of Personality. 90(3), 405–425. DOI: https://doi.org/10.1111/jopy.12674

[27] Ponta, L., Puliga, G., Manzini, R., 2021. A measure of innovation performance: the Innovation Patent Index. Management Decision. 59(13), 73–98. DOI: https://doi.org/10.1108/MD–05–2020–0545

[28] Smith, M., Agrawal, R.A., 2015. Comparison of Time Series Model Forecasting Methods on Patent Groups. MAICS. 1353, 167–173. Available from: https://ceur–ws.org/Vol–1353/paper_13.pdf (cited 24 March 2025).

[29] Noh, H., Jo, Y., Lee, S., 2015. Keyword selection and processing strategy for applying text mining to patent analysis. Expert Systems with Applications. 42(9), 4348–4360. DOI: https://doi.org/10.1016/j.eswa.2015.01.050

[30] Xiong, Q., Ma, Z., 2025. The Multi–dimensional Evaluation System Framework for New Equipment in Rail Transit. In: Zeng, X., Xie, X., Sun, J., et al. (eds.). International Symposium for Intelligent Transportation and Smart City (ITASC) 2025 Proceedings, Lecture Notes in Electrical Engineering. Springer Nature Singapore: Singapore, pp. 109–117. DOI: https://doi.org/10.1007/978–981–96–4702–6_12

[31] Youn, S.J., Lee, Y.–J., Han, H.–E., et al., 2024. A Data Analytics and Machine Learning Approach to Develop a Technology Roadmap for Next–Generation Logistics Utilizing Underground Systems. Sustainability. 16(15), 6696. DOI: https://doi.org/10.3390/su16156696

[32] Lewis, R., 2015. A semantic approach to railway data integration and decision support [Doctoral Dissertation]. University of Birmingham: Birmingham, UK. Available from: http://etheses.bham.ac.uk/id/eprint/5959 (cited 24 March 2025).

[33] Tutcher, J., 2016. Development of semantic data models to support data interoperability in the rail industry [Doctoral Dissertation]. University of Birmingham: Birmingham, UK. Available from: https://etheses.bham.ac.uk/id/eprint/6774/1/Tutcher16PhD.pdf (cited 24 March 2025).

[34] Jägare, V., 2022. A challenge–driven framework for innovations in railways [Doctoral Dissertation]. Luleå University of Technology: Luleå, Sweden. Available from: https://www.diva–portal.org/smash/get/diva2:1701543/FULLTEXT02.pdf (cited 24 March 2025).

[35] Ariyachandra, M.R.M.F., Wen, Y., 2025. Building a holistic smart technology–centric value assessment framework for rail infrastructure. University of Strathclyde Publishing: Glasgow, Scotland. DOI: https://doi.org/10.17868/STRATH.00093275

[36] Hansen, C., Daim, T., Ernst, H., et al., 2016. The future of rail automation: A scenario–based technology roadmap for the rail automation market. Technological Forecasting and Social Change. 110, 196–212. DOI: https://doi.org/10.1016/j.techfore.2015.12.017

[37] Unger, S., Heinrich, M., Scheuermann, D., et al., 2023. Securing the Future Railway System: Technology Forecast, Security Measures, and Research Demands. Vehicles. 5(4), 1254–1274. DOI: https://doi.org/10.3390/vehicles5040069

[38] Karasev, O., Beloshitskiy, A., Shitov, E., et al., 2022. Integral Assessment of the Level of Innovative Development of the Railway Industry Companies. The Open Transportation Journal. 16(1), e187444782203141. DOI: https://doi.org/10.2174/18744478–v16–e2203141

Downloads