In the context of intensifying global climate change and the steady advancement of the “carbon peak and carbon neutrality” strategy, the electric car industry is rapidly reshaping the traditional automotive industrial landscape, emerging as a pivotal force driving energy transition and green development. With the deepening implementation of the dual-carbon strategy and the continuous refinement of industrial policy systems, the sales of new electric cars in China have been consistently rising, reaching 12.866 million units in 2024, a year-on-year increase of 35.5%, accounting for 40.9% of total new car sales. Technological advancements and market expansion have formed a virtuous interactive trend, driving the gradual expansion and deepening of the industrial chain. Simultaneously, as technological complexity increases and industrial division of labor deepens, new challenges in the electric car industry chain, such as innovation resource allocation, upstream-downstream collaborative efficiency, and system integration capabilities, are gradually emerging. Existing research has explored the development and collaborative mechanisms of the electric car industry from multiple dimensions. Some studies focus on the evolution of technological pathways and innovation diffusion mechanisms, revealing key technological nodes and evolutionary trends; others analyze the structural evolution characteristics and collaborative efficiency of the industrial chain from a regional industrial cluster perspective; and yet others address issues such as upstream-downstream collaborative barriers and resource allocation efficiency in the electric car sector. Overall, the existing literature tends to emphasize static analyses from a single dimension, lacking systematic research from the micro-level perspective of innovation networks to uncover the collaborative optimization mechanisms of the electric car industry chain, particularly in terms of patent cooperation network structures, identification of technology-industry interaction pathways, and their impact mechanisms on collaborative efficiency. Shandong Province is a significant agglomeration area for China’s electric car industry, boasting a relatively complete industrial chain foundation and strong innovation vitality. Therefore, this study takes Shandong Province’s electric car industry as the research object, based on patent cooperation and industrial development data from 2016 to 2023, constructs a technological innovation network for electric cars, and explores the optimization pathways of the electric car industry chain from the micro-level of knowledge flow and cooperative interaction, providing empirical support and policy recommendations for industrial collaborative development.
To systematically analyze the coupling relationship between technological innovation and industrial development in Shandong Province’s electric car industry chain and its impact mechanisms, a research framework of “technological innovation-industrial development-collaborative optimization” is constructed. This study develops an evaluation indicator system from two dimensions: technological innovation and industrial development, aiming to systematically reflect the synergy between innovation capabilities and development levels in the upstream, midstream, and downstream segments of Shandong Province’s electric car industry chain. The indicator system design is based on industrial chain synergy theory and technological innovation network theory, grounded in Porter’s value chain synergy concept and Freeman’s innovation network structure model, emphasizing the coupling relationship between the structural characteristics of technological innovation and the functional performance of industrial development. Concurrently, incorporating policy documents such as the “14th Five-Year Plan for National Strategic Emerging Industries,” “New Energy Vehicle Industry Development Plan (2021-2035),” and “Shandong Province New Energy Vehicle Industry Development Plan (2021-2025),” the focus is on highlighting technological innovation as the core, promoting synergistic development across all segments of the industrial chain, and achieving industrial chain optimization and competitiveness enhancement. Empirically, referencing related research findings by scholars such as Xu Xiurui and Pan Sunan, and combining authoritative data resources like the Global Patent Intelligence System and the China Automotive Industry Yearbook, the indicator design ensures scientific rigor, policy orientation, and operability.
Technological innovation indicators primarily examine the cooperative network characteristics and node influence of the upstream, midstream, and downstream segments, including quantitative indicators such as the number of patents in each segment, patent cooperation network density, network standardized centrality, and average patent citation counts. Industrial development indicators are evaluated from six dimensions: material supply, cost control, production scale, cost efficiency, market performance, and infrastructure, encompassing key indicators such as lithium battery production-sales ratio, import prices of key materials, and annual production of electric cars. Negative indicators have been inversely processed to ensure uniformity in indicator direction. Indicator weights are determined using the CRITIC method, which comprehensively considers the variability and conflict of indicators, effectively reflecting the importance of each indicator and ensuring the rationality and scientificity of weight allocation (CR < 0.1). The specific indicator system is presented in Table 1.
| Primary Indicator | Secondary Indicator | Tertiary Indicator | Indicator Attribute | Weight (%) |
|---|---|---|---|---|
| Technological Innovation | Upstream Cooperation Network Characteristics | X1: Upstream Patent Count | + | 15.15 |
| X2: Upstream Patent Cooperation Network Density | + | 5.61 | ||
| Upstream Node Influence | X3: Upstream Network Standardized Centrality | + | 6.69 | |
| X4: Upstream Average Patent Citations | + | 7.30 | ||
| Midstream Cooperation Network Characteristics | X5: Midstream Patent Count | + | 8.35 | |
| X6: Midstream Patent Cooperation Network Density | + | 11.22 | ||
| Midstream Node Influence | X7: Midstream Network Standardized Centrality | + | 5.70 | |
| X8: Midstream Average Patent Citations | + | 7.00 | ||
| Downstream Cooperation Network Characteristics | X9: Downstream Patent Count | + | 6.14 | |
| X10: Downstream Patent Cooperation Network Density | + | 5.65 | ||
| Downstream Node Influence | X11: Downstream Network Standardized Centrality | + | 7.38 | |
| X12: Downstream Average Patent Citations | + | 13.81 | ||
| Industrial Development | Material Supply | Y1: Lithium Battery Production-Sales Ratio | + | 15.89 |
| Y2: Lithium Battery Market Share | + | 9.47 | ||
| Cost Control | Y3: Key Material Import Price | – | 15.40 | |
| Production Scale | Y4: Annual Production of Electric Cars | + | 16.80 | |
| Y5: Number of Newly Established Electric Car Enterprises | + | 7.88 | ||
| Cost Pressure | Y6: Capacity Utilization Rate of Key Enterprises | + | 9.07 | |
| Market Performance | Y7: Annual Sales of Electric Cars | + | 9.43 | |
| Y8: Industry User Satisfaction | + | 8.81 | ||
| Infrastructure | Y9: Number of Public Charging Piles | + | 7.24 |
The coupling coordination model is a method used to measure the degree of interaction between two or more systems. Referencing the extension of the concept of capacity coupling in physics and the capacity system model, the coupling coordination model is constructed as follows:
$$ C = n \left\{ \frac{(u_1 \times u_2 \times \cdots \times u_n)}{(u_1 + u_2 + \cdots + u_n)^n} \right\}^{1/n} $$
where \( C \) is the coupling degree; \( T \) is the comprehensive coordination index; \( D \) is the coupling coordination degree; \( a_1 \) to \( a_n \) are the weights in the indicator system; and \( n \) is the number of coupled systems or elements. The comprehensive coordination index \( T \) is calculated as:
$$ T = a_1 u_1 + a_2 u_2 + \cdots + a_n u_n $$
And the coupling coordination degree \( D \) is given by:
$$ D = \sqrt{C \times T} $$
Referencing existing research results from scholars, the coupling coordination degree levels for the electric car industry chain are classified as shown in Table 2.
| Coordination Level | Coordination Degree | Coupling Coordination Degree Interval |
|---|---|---|
| 1 | Extreme Dysregulation | (0.0, 0.1) |
| 2 | Severe Dysregulation | [0.1, 0.2) |
| 3 | Moderate Dysregulation | [0.2, 0.3) |
| 4 | Mild Dysregulation | [0.3, 0.4) |
| 5 | On the Verge of Dysregulation | [0.4, 0.5) |
| 6 | Barely Coordinated | [0.5, 0.6) |
| 7 | Primary Coordination | [0.6, 0.7) |
| 8 | Intermediate Coordination | [0.7, 0.8) |
| 9 | Good Coordination | [0.8, 0.9) |
| 10 | High-Quality Coordination | [0.9, 1.0) |
In the study of collaborative optimization of Shandong Province’s electric car industry chain, by constructing an obstacle factor diagnosis model, the degree of influence of various factors on the coupling coordination degree can be quantified. The obstacle degree \( O_{ij} \) of the j-th indicator in the i-th year is determined by the following formula:
$$ O_{ij} = \frac{(1 – x_{ij}) \times w_j}{\sum_{j=1}^{n} [(1 – x_{ij}) \times w_j]} \times 100\% $$
where \( x_{ij} \) is the standardized indicator value, \( 0 \leq x_{ij} \leq 1 \); \( w_j \) is the indicator weight determined by the CRITIC method; and \( (1 – x_{ij}) \) reflects the deviation of the indicator value from the ideal state.
Regarding the truncation characteristic of the dependent variable (industrial chain synergy degree \( D_{it} \)), where \( 0 < D_{it} < 1 \), traditional linear regression models may lead to biased parameter estimates due to the asymmetry of the data distribution. Therefore, the study employs a panel Tobit regression model to address the statistical inference problem of limited dependent variables, while quantifying the nonlinear impact mechanism of technological innovation network characteristics on the synergy degree.
$$ D_{it}^* = \beta x_{it} + \alpha_i + \varepsilon_{it}, \quad \varepsilon_{it} \sim N(0, \sigma^2) $$
$$ D_{it} = \begin{cases}
0 & \text{if } D_{it}^* \leq 0 \\
D_{it}^* & \text{if } 0 < D_{it}^* < 1 \\
1 & \text{if } D_{it}^* \geq 1
\end{cases} $$
where \( D_{it}^* \) is the latent variable, representing the theoretical synergy degree; \( D_{it} \) is the observed value (limited dependent variable); \( x_{it} \) is the set of explanatory variables; \( \beta \) is the coefficient to be estimated; \( \alpha_i \) represents the individual fixed effect; and \( \varepsilon_{it} \) is the error term.
The research primarily selects development data of Shandong Province’s electric car industry from 2016 to 2023, constructing a complete analysis dataset through multi-source data integration. For technological innovation network data, it is mainly based on the Global Patent Intelligence System and UCINET social network analysis tool. Using IPC classification numbers (including B60L, H01M, etc.) combined with keywords such as “electric car” and “power battery” for systematic retrieval, detailed information including patent application numbers, patent types, and applicant information is obtained. The time span is from 2016 to 2023, finally resulting in 19,122 patents in Shandong Province’s electric car industry. Subsequently, UCINET is used to process and analyze the data. By modeling the cooperative relationships between patent applicants as a network, key indicators such as cooperation network density and centrality in upstream, midstream, and downstream technological fields are calculated. Industrial development data mainly come from three sources: first, official statistical materials, including macroeconomic indicators from the China Automotive Industry Yearbook; second, enterprise operation data, obtained through the “Electric Car Industry Operation Monitoring Report” released by the Shandong Provincial Department of Industry and Information Technology and annual reports of key enterprises; third, market monitoring data, mainly citing the “Electric Car Market Data Report” from the China Automotive Technology and Research Center and annual statistics from the China Electric Vehicle Charging Infrastructure Promotion Alliance. Additionally, the research collects policy texts such as Shandong Provincial Government bulletins and various city industrial planning documents, as well as auxiliary data like the annual Shandong Province Electric Car User Satisfaction Survey Report. To ensure data quality, cross-validation from different data sources ensures accuracy, and for a small amount of missing data (less than 3%), reasonable interpolation is performed using the geometric average growth rate. For outlier issues, data points beyond the ±3σ range are traced and verified, and after confirmation, the Winsorize method is used for processing.
During the period from 2016 to 2023, the cooperation network density in the upstream, midstream, and downstream of the electric car industry chain all showed a逐年 declining trend. Overall, the midstream cooperation network density is significantly higher than that of the upstream and downstream, indicating its more active position in the industrial chain’s innovation cooperation. However, the continuous decline in network density across all segments reflects a gradual weakening of the overall industrial chain’s internal collaborative innovation activity. Specifically, the midstream cooperation network density peaked in 2016, then declined year by year, with a slight rebound in 2019, but the overall trend remained downward, dropping to about 0.04 by 2023. The upstream and downstream cooperation network densities were close in initial values, both around 0.04–0.05, but the decline was significant, both falling to about 0.01 by 2023. The downstream segment experienced a brief rebound after 2018 but subsequently also trended downward. This change indicates that although the midstream remains the core of the industrial chain’s innovation cooperation, the overall innovation collaborative relationships across the upstream, midstream, and downstream are weakening, possibly due to multiple factors such as intensified market competition, differentiation in technological pathways, and optimization adjustments in resource allocation. To promote the overall innovation efficiency of the industrial chain in the future, efforts are needed to enhance vertical collaboration and resource integration capabilities among all segments.
Between 2016 and 2023, the coupling coordination degree among the upstream-downstream, upstream-midstream, and midstream-downstream of the electric car industry chain underwent significant changes, overall showing an evolution trend of “high level—decline—fluctuation.” From 2016 to 2017, the coordination degrees at all levels were generally maintained at high levels, close to or above 0.95, especially in 2017, when the upstream-midstream, midstream-downstream, and upstream-downstream coordination degrees all reached peaks near 0.99. During this stage, under the strong guidance of national policies (such as financial subsidy policies for the promotion and application of electric cars, and the construction of access standard systems), innovation activities were in step, and upstream-downstream collaboration in the industrial chain was close, forming a highly coupled collaborative innovation network. However, starting in 2018, the inter-level coupling coordination degree generally experienced a cliff-like decline. Among them, the upstream-downstream coordination degree dropped most significantly, from 0.985 in 2017 to 0.242 in 2018; the upstream-midstream and midstream-downstream also experienced declines to varying degrees. This change mainly stems from several aspects: First, the gradual phase-out of subsidy policies for electric cars weakened the policy binding relationship between upstream and downstream enterprises, reducing willingness to collaborate. Second, as the competition among power battery technology routes intensified, different enterprises diverged in their choice of technological directions, leading to the breakdown of the original collaborative system. Additionally, changes in market demand drove rapid iteration of end products, and整车 manufacturing enterprises put forward higher and more diverse requirements on the component supply chain, further exacerbating the inconsistency in rhythm within the industrial chain. International environmental instability and sharp fluctuations in raw material prices also made supply chain relationships tense, increasing the difficulty and cost of industrial chain collaboration. It is worth noting that from 2020 to 2022, the coordination degree of some segments showed a recovery trend, especially in the upstream-midstream segment. This phenomenon indicates that leading enterprises, through resource integration, technology alliances, and other means, have to some extent repaired the collaborative relationships between some nodes of the industrial chain. However, the upstream-downstream coordination degree recovered slowly, indicating that there is still a large disconnect between infrastructure support, market-end demand, and production-end supply. This shows that to improve the overall coupling coordination degree of the industrial chain in the future, it is necessary to rely on standard unification, improvement of the supply chain system, and optimization of the innovation ecosystem, strengthen communication and collaboration between different segments, and promote the overall leap in the collaborative innovation capability of the industrial chain.
The study calculated the coupling coordination degree between technological innovation and industrial development, and the results show phased evolution characteristics. From the time series evolution, the coupling coordination degree increased continuously from 0.391 (mild dysregulation) in 2016 to 0.920 (high-quality coordination) in 2023. This evolution process can be divided into three typical stages: The first stage (2016–2018) was the policy-driven period. Under the promotion of policies such as the “Shandong Province Electric Car Industry Development Plan,” the coordination degree had an average annual growth rate of 20.5%, and broke through the coordination threshold of 0.5 in 2018 (D=0.567). The second stage (2019–2021) was affected by the subsidy phase-out policy, and the growth rate fell to 7.3%, but it still maintained a coordinated state. The third stage (2022–2023), with breakthroughs in core technologies such as solid-state batteries, the coordination degree achieved a leap of 21.7%, finally reaching a high-quality coordination level. From the perspective of coordination level division, the system experienced a “dysregulation-coordination” state transition during the research period. Among them, the coordination degree broke through the 0.7 boundary in 2021 (D=0.700), marking the transition of the technology-industry system from “primary coordination” to a new stage of “benign interaction.” Further analysis found that the evolution of coordination degree has three key driving mechanisms: First, the policy leverage effect is manifested as a significant improvement in coordination degree within 1–2 years after the introduction of major policies, such as the implementation of the industrial plan in 2017 leading to a 26% growth in coordination degree in 2018. Second, the technological innovation inflection point effect is prominent. The breakthrough in key technologies by leading enterprises in 2020 prompted an average annual growth of 11.9% in coordination degree from 2021 to 2023. Third, the market regulation effect is obvious. The subsidy phase-out in 2019 caused a阶段性回落 in the growth rate of coordination degree. This triple driving mechanism of “policy-technology-market” jointly shaped the dynamic path of technology-industry collaborative development in Shandong Province.
Based on the obstacle degree model calculation results, this study systematically identified the key obstacle factors for the collaborative development of technological innovation and industry in Shandong Province’s electric car industry from 2016 to 2023. Through the obstacle degree analysis of 12 technological innovation indicators (X1~X12) and 9 industrial development indicators (Y1~Y9), it was found that the obstacle factors in different industrial chain segments show significant differences and have dynamic evolution characteristics. The obstacles in the upstream segment are mainly concentrated on patent quality, among which X4 rose rapidly to 12.57% in 2022-2023, reflecting the increasingly prominent problem of insufficient innovation quality in basic research. The obstacle degree of X3 increased from 8.11% in 2019 to 10.24% in 2023, indicating that the construction of hub nodes in the upstream innovation network is relatively lagging. It is worth noting that the obstacle degree of X1 continued to decrease from 11.20% in 2016 to 7.84% in 2021, showing that the number of patents is no longer a major constraint. The obstacle characteristics of the midstream segment show phased changes: from 2016 to 2019, X6 was the main obstacle (9.13% in 2016), indicating insufficient collaborative innovation among component enterprises; after 2020, the obstacle degree of X8 increased significantly (12.05% in 2023), indicating that the quality of applied technology innovation has become a new bottleneck. The obstacle degree of X5 remained at 7%–9% from 2019 to 2021, showing that the patent quantity problem has been partially alleviated. The obstacles in the downstream segment have rapidly become prominent in recent years. X11 and X9 reached obstacle degrees of 12.71% and 10.57% respectively in 2023, revealing obvious shortcomings in the innovation network construction and patent reserves of整车 enterprises. This echoes with the fact that Y7 was the main obstacle from 2018 to 2020 (7.04%–7.42%), reflecting the mismatch between end product innovation and market demand. Focusing on the obstacle evolution of the industrial development system, it presents three characteristic stages: from 2016 to 2018, dominated by Y1 and Y3 (combined obstacle degree 23.14%); from 2019 to 2021, shifted to Y4 and Y6 as the main ones (combined obstacle degree 21%–23%); from 2022 to 2023, prominently manifested as insufficient Y9 (obstacle degree 10.57%). This evolution reflects that the industrial bottleneck has gradually shifted from initial production capacity problems to insufficient infrastructure support. In summary, the dynamic analysis of obstacle factors reveals three important trends: First, the focus of obstacles has shifted from “quantitative” indicators (patent number, production volume) to “qualitative” indicators (patent citations, network centrality). Second, obstacles in different segments of the industrial chain show a relay-like alternation characteristic. Finally, infrastructure (Y9) and market (X11) obstacles have significantly increased recently. These findings provide an empirical basis for formulating precise industrial chain collaborative policies. At the same time, the evolution of obstacle factors has an obvious response relationship with industrial policy adjustments. The policy tools mainly based on financial subsidies from 2016 to 2018 had a significant effect on resolving material-end obstacles (Y1, Y3); but after the subsidy phase-out in 2019, the collaborative problems in the midstream manufacturing segment began to emerge; and the “chain leader system” implemented in 2021, although to some extent improved network connectivity, failed to effectively enhance innovation quality indicators. This difference in policy effects suggests that in the future, it is necessary to establish a more precise obstacle diagnosis and targeted governance mechanism, especially to strengthen breakthroughs in new bottlenecks such as downstream service networks and innovation quality.
Based on the previous obstacle factor diagnosis results and the characteristics of the electric car industry chain, this study constructs a panel Tobit regression model to focus on examining the impact mechanism of technological innovation network characteristics and industrial development indicators on the coupling coordination degree. Referring to existing research and combining the actual development situation of Shandong Province’s electric car industry chain and the obstacle factor analysis results, the annual technology-industry coupling coordination degree is used as the explained variable, and upstream innovation capability (UPS), midstream collaboration level (MDS), downstream transformation efficiency (DNS), supply chain cost pressure (SCP), production scale effect (PSE), and market driving capability (MDC) and other related indicators are used as explanatory variables, measured by upstream patent count, midstream patent cooperation network density, downstream average patent citations, key material import price, annual production of electric cars, and annual sales of electric cars, respectively. Stata software is used to perform the regression of the panel Tobit model. The regression results are shown in Table 4.
| Item | Regression Coefficient1 |
|---|---|
| Intercept | 8.601*** (2.912) |
| UPS | 0.021** (2.407) |
| MDS | 0.003** (2.362) |
| DNS | 0.007** (2.437) |
| SCP | -0.007*** (-2.937) |
| PSE | 0.054** (2.427) |
| MDC | 0.017*** (2.598) |
1 **P<0.05; ***P<0.01; z-values in parentheses.
From Table 4, in the technological innovation network dimension, the regression coefficient of UPS is 0.021 (P<0.05), indicating that for every 1% increase in upstream patent count, the coupling coordination degree increases by 0.021 units. This result verifies the positive impact of technological accumulation in upstream segments such as battery materials in Shandong Province on industrial chain collaboration. The coefficient of MDS is 0.003 (P<0.05), indicating that the increase in patent cooperation network density among midstream enterprises can effectively promote technological collaboration, but the degree of impact is relatively small, and the synergy effect still has room for improvement. The coefficient of DNS is 0.007 (P<0.05), reflecting that the market recognition of downstream technological achievements has a significant promoting effect on the overall coordination of the industrial chain, highlighting the importance of the marketization transformation of technological innovation. In the industrial development dimension, the coefficient of PSE is as high as 0.054 (P<0.05), indicating that the annual production of electric cars has the most significant positive impact on the coupling coordination degree, confirming the key position of economies of scale in industrial chain collaboration. The coefficient of market driving capability MDC is 0.017 (P<0.01), indicating that end market demand is an important driving force for pulling the coordinated development of the industrial chain. It is worth noting that the coefficient of SCP is -0.007 (P<0.01), which means that for every 1% increase in the import price of key materials, the coupling coordination degree will decrease by 0.007 units, revealing the constraining effect of supply chain security on the sustainable development of the electric car industry chain. In addition, there are obvious differences in the impacts of technological innovation network characteristics and industrial development indicators: the impact coefficients of upstream innovation capability and production scale effect are the largest, indicating that source technological innovation and industrial scale expansion are the two core driving forces for promoting coupled and coordinated development; the impact of midstream collaboration level is relatively limited, reflecting that the collaborative efficiency of the midstream segment of the current industrial chain still has room for improvement; the significant negative effect of supply chain cost pressure warns that excessive reliance on imported key materials may weaken the coordination stability of the industrial chain.
Based on the data of Shandong Province’s electric car industry chain from 2016 to 2023, this paper constructs a two-dimensional evaluation system of technological innovation and industrial development, and uses social network analysis, coupling coordination model, obstacle degree model, and panel Tobit regression model to systematically analyze the impact mechanism of technological innovation network characteristics on the collaborative optimization of the industrial chain, and draws the following main conclusions. First, there are obvious inter-level differences and a downward trend in the innovation network cooperation density within the industrial chain. From 2016 to 2023, the cooperation network density of upstream, midstream, and downstream all decreased year by year, among which the midstream density was always higher than that of upstream and downstream, dropping to 0.04 in 2023, while upstream and downstream dropped to about 0.01, indicating that the overall collaborative innovation activity of the industrial chain significantly weakened. Although the midstream is the center of cooperation, the overall collaborative activity of the industrial chain is weakening, and internal vertical collaboration needs to be strengthened. Second, the coupling coordination degree of upstream, midstream, and downstream of the industrial chain experienced phased fluctuations. The coupling coordination degree of upstream, midstream, and downstream experienced an evolution process of “high level—decline—fluctuation.” In 2017, the coordination degree of upstream, midstream, and downstream reached a peak (about 0.99), but starting in 2018, there was a cliff-like decline, for example, the upstream-downstream coordination degree dropped from 0.985 to 0.242, reflecting the significant impact of policy phase-out, technological differentiation, and market fluctuations on the collaborative structure. Although the midstream coordination recovered from 2020 to 2022, the overall collaborative pattern gradually transformed into a local collaborative network with leading enterprises as the core. Third, the collaborative level between technological innovation and industrial development continues to improve. The coupling coordination degree between technological innovation and industrial development increased from 0.391 (mild dysregulation) in 2016 to 0.920 (high-quality coordination) in 2023, showing a triple mechanism drive of “policy-technology-market,” especially after 2021, the technological inflection point effect significantly increased, with an average annual growth rate of 11.9%, driving the system into a benign interaction stage. Fourth, the collaborative development obstacle factors show dynamic evolution and chain distribution characteristics. The obstacle factors shifted from “quantitative” (such as patent number) to “qualitative” (such as network centrality). In 2023, the obstacle degree of upstream X4 reached 12.57%, downstream X11 reached 12.71%, and industrial development Y9 reached 10.57%, highlighting that the current collaborative optimization faces the dual challenges of quality improvement and infrastructure shortcomings. At the same time, the focus of obstacles shows a “relay” rotation among different segments of the chain, reflecting that collaborative governance requires precise policies in stages and levels. Fifth, the panel Tobit empirical results verify the significant role of key influencing factors. Technological network structure characteristics and industrial development have significant impacts on collaborative development. Upstream innovation capability (UPS, β=0.021, P<0.05) and production scale effect (PSE, β=0.054, P<0.05) are the strongest positive factors, highlighting the core role of source innovation and scale effect. At the same time, supply chain cost pressure (SCP, β=–0.007, P<0.01) is the only significant negative factor, suggesting that fluctuations in raw material prices seriously constrain collaborative development, and the construction of industrial chain resilience needs to be highly valued. In summary, the collaborative optimization of the electric car industry chain shows dynamic evolution characteristics of “structural differentiation—coupling leap—obstacle transfer.” To improve the system coordination level in the future, it should focus on enhancing the original innovation capability of the upstream, optimizing the efficiency of the midstream collaborative network, and the transformation of downstream achievements and market adaptation, while establishing a triple support mechanism of policy guidance, technological breakthrough, and market response.
Based on the research conclusions, the following countermeasures and suggestions are proposed. First, strengthen the vertical collaboration mechanism of upstream, midstream, and downstream. Leading enterprises should be used as the牵引 to build innovation collaboration platforms for upstream, midstream, and downstream, promote整车 enterprises to establish closer joint research and development mechanisms with battery, component, and basic material suppliers, and improve the overall collaborative innovation level of the industrial chain. Second, improve the innovation quality and network construction of the upstream. The obstacles in the upstream segment are mainly concentrated on patent quality and innovation network construction. It should focus on the original innovation and independent controllable capability of key components such as batteries, motors, and electronic controls, promote research institutions, universities, and enterprises to jointly build innovation consortia, and enhance the technological supply capability and network hub status of the upstream segment. It is recommended to establish a “strengthening and supplementing the chain” special fund to enhance the continuous investment capability of upstream enterprises in technological breakthroughs. Third, optimize the market transformation and service network of the downstream segment. Encourage downstream enterprises to actively closely connect with upstream research and development segments, form a demand-oriented innovation closed loop, and improve the matching between end products and market demand; at the same time, increase the support for the promotion and application of electric cars, and enhance the market activity of the user end; strengthen the construction of infrastructure such as public charging piles, and optimize the downstream service network. Fourth, strengthen the precision and dynamic adjustment of policies. Policies have played an important role in promoting the coupling coordination between technological innovation and industrial development, but the policy effects in different stages are different. It is recommended to formulate more precise policy tools according to the obstacle characteristics and dynamic evolution trends of different segments of the industrial chain. For example, increase policy support for weak links such as upstream basic research and downstream infrastructure construction; at the same time, adjust the direction and intensity of policies in time according to the dynamic changes of market and technology development, to avoid policy lag or excessive intervention. Fifth, enhance supply chain security and independent controllable capability. Supply chain cost pressure has a significant negative impact on the coupling coordination degree, indicating that excessive reliance on imported key materials may weaken the coordination stability of the industrial chain. It is recommended to reduce dependence on a single imported resource through strategic reserves, localization substitution of raw materials, and international diversified layout, and enhance risk resistance capability. At the same time, reduce production costs and improve the overall competitiveness of the industrial chain by optimizing supply chain management. Sixth, build a dynamic obstacle factor monitoring and collaborative governance mechanism. In response to the phased changes of obstacle factors, a coupling system obstacle database and early warning model should be constructed to promote the transformation of policies from inclusive support to differentiated and precise governance, especially in terms of infrastructure support and downstream service system construction to achieve targeted support and investment. At the same time, it is recommended to build a collaborative governance mechanism involving the government, enterprises, research institutions, and social organizations, and through multi-party cooperation, jointly promote the collaborative development and high-quality development of Shandong Province’s electric car industry chain.