This study begins with the observation that the global automobile industry is rapidly shifting from an internal combustion engine (ICE)-centered paradigm to future vehicles such as electric vehicles (EVs) and autonomous vehicles, accompanied by struct...
This study begins with the observation that the global automobile industry is rapidly shifting from an internal combustion engine (ICE)-centered paradigm to future vehicles such as electric vehicles (EVs) and autonomous vehicles, accompanied by structural changes in innovation and the industrial ecosystem. In order to overcome the limitations of previous research, which mainly focused on the analysis of local industry realities or changes in production networks, this study aims to elucidate the changes in technological knowledge creation activities and corresponding patterns of regional adaptation in the Korean automobile industry in response to changes in the industrial environment. Using patent big data from 1985 to 2022, both quantitative and qualitative analyses were conducted on the patterns of technological evolution and the changes in actors and regions involved in technological knowledge creation. The results of the study are as follows.
First, the Korean automobile industry is a representative comprehensive industry that is capital-, labor-, and knowledge-intensive, combining diverse sectors such as steel, machinery, electronics, and materials. In its early stages, the industry tended to rely on foreign technologies, but with the government's industrial promotion policies and proactive R&D investments led by the Hyundai Motor Group, it succeeded in the localization and diversification of parts technologies. While the Korean automobile industry has formed an oligopolistic market structure centered on the Hyundai Motor Group, the transition to future vehicles such as EVs has shifted the value chain from a vertical to a more horizontal relationship. In the past, production and R&D functions often coexisted in specific regions such as Ulsan; however, since the 2000s, R&D functions have increasingly concentrated in the capital region, including Seoul and Gyeonggi, while production functions are distributed across provincial regions, deepening the spatial division of labor.
Second, through IPC code and text-based analysis of patent data, it was revealed that technological knowledge in the Korean automobile industry has evolved in accordance with changes in the industrial environment. In addition, patents were classified using natural language processing techniques applied to textual data. Since the mid-1990s, the number of ICE-related patent applications has gradually decreased, whereas since the 2000s, patent applications related to autonomous vehicles and EVs have rapidly increased. According to IPC co-occurrence network analysis, the number of patents incorporating convergent technologies has increased, and the level of technological convergence has been continuously deepening. Unlike in the past, when traditional mechanical and parts technologies were central, recently, future vehicle-related technologies such as electrochemical cells, batteries, digital data processing, and sensors/instruments have emerged as core hubs within the network. The patent technology growth stage model also quantitatively confirmed that ICE technologies have entered a decline phase, while electric vehicles, hydrogen vehicles, and autonomous driving technologies are in growth and maturity phases.
Third, by analyzing the changes in actors and regions involved in technological knowledge creation over time, this study elucidated the co-evolutionary relationships among industry, technology, actors, and regions. In terms of actors, the early period was characterized by a single-hub structure with finished vehicle manufacturers such as Hyundai Motor Group leading innovation. However, with the transition to future vehicles, non-automotive companies such as Samsung Electronics and LG Chem have actively entered fields such as automotive electronics, secondary batteries, and advanced materials, diversifying the actor composition by increasing their share of patent applications. While collaboration among companies accounts for the largest share of actor cooperation, the frequency of industry-academia-research cooperation with non-corporate actors such as universities, research institutes, and public institutions has increased, evolving toward an open structure. Spatially, while manufacturing clusters in the past also served innovation functions, since the 2000s, southern Gyeonggi (Hwaseong, Suwon, Yongin, etc.) and Daejeon have emerged as innovation hubs based on diverse knowledge bases, intensifying the spatial concentration of technological knowledge creation. In particular, Ulsan has shown an adaptive pattern of weakening research functions and becoming a simple production base due to the relocation of core companies' research functions.
In conclusion, this study empirically examined that the patterns of adaptation of actors and regions to changes in the industrial environment vary depending on their industrial and knowledge bases. In the Korean automobile industry, it was found that the innovation capabilities of actors rooted in a region, such as companies, had a greater impact on regional adaptation than the role of the region itself.
This study has academic significance in that it addresses changes in innovation activities during industrial transitions from an evolutionary perspective, emphasizing the necessity of a multi-layered approach to technology, actors, and regions. Methodologically, it demonstrates the potential of big data analysis through the application of natural language processing techniques and network analysis in economic geography.