In the context of global energy transition and environmental challenges, the rapid development of electric cars has become a critical pathway for advancing new productive forces and achieving carbon neutrality goals. The resilience of the electric car supply chain serves as a foundational element for addressing complex risks and ensuring the stability and sustainability of the industrial chain. This article explores the evaluation of supply chain resilience for China EV, integrating game theory combination weighting and the extension cloud model to address uncertainties and randomness in the assessment process. The evaluation framework is built on resilience theory and dynamic capability theory, encompassing five dimensions: resistance and defense capability, adaptation and recovery capability, learning and transformation capability, sustainable development capability, and globalization capability. By combining subjective and objective weighting methods through game theory, and applying the extension cloud model for qualitative and quantitative analysis, this approach provides a robust tool for assessing and enhancing the resilience of the electric car supply chain.
The electric car industry, particularly in China EV markets, faces unique challenges due to its reliance on critical raw materials, rapid technological iterations, and globalized supply networks. Traditional evaluation methods often struggle with the fuzzy and stochastic nature of indicators, leading to potential inaccuracies. This study addresses these limitations by developing a comprehensive指标体系 that captures the multifaceted nature of supply chain resilience. The methodology involves determining indicator weights using a combination of trapezoidal fuzzy analytic hierarchy process (AHP) and improved Critic method, optimized through game theory to minimize deviations. The extension cloud model is then employed to evaluate resilience levels, leveraging cloud parameters to handle ambiguity and randomness. The results indicate that the overall resilience of the electric car supply chain is “relatively high,” with most primary indicators performing well, though globalization capability lags. Specific secondary indicators, such as supply chain digitalization and regional production capacity substitution, demonstrate high resilience, while geopolitical sensitivity and crisis diplomatic mediation require improvement. This evaluation aligns with practical scenarios and offers actionable insights for strengthening the electric car supply chain in China EV and beyond.
Introduction to Electric Car Supply Chain Resilience
The electric car industry has emerged as a pivotal sector in the global push for sustainable transportation, with China EV leading in production and adoption. Supply chain resilience, defined as the ability to anticipate, withstand, and recover from disruptions, is crucial for maintaining operational continuity in the face of risks such as resource scarcity, geopolitical tensions, and technological shifts. For electric cars, resilience encompasses not only traditional supply chain aspects but also specific factors like battery material sourcing, digital integration, and international collaboration. Existing research on supply chain resilience has employed various methods, including qualitative case studies, quantitative models, and comprehensive evaluations. However, these approaches often fall short in addressing the inherent模糊性 and randomness of indicators, particularly in the context of China EV supply chains. This study bridges this gap by introducing a novel framework that combines game theory for weight optimization and the extension cloud model for resilience evaluation, providing a more accurate and reliable assessment tool.
The importance of resilience in the electric car supply chain cannot be overstated, as disruptions can lead to significant economic and environmental repercussions. For instance, reliance on lithium, cobalt, and other critical minerals concentrated in specific regions makes the supply chain vulnerable to geopolitical and market fluctuations. Additionally, the fast-paced technological evolution in electric cars demands agility and adaptability from supply chain actors. The extension cloud model, which integrates matter-element theory with cloud modeling, effectively handles the transformation from qualitative descriptions to quantitative analysis, making it suitable for evaluating the resilience of China EV supply chains. By incorporating game theory combination weighting, this approach ensures that both expert knowledge and objective data are balanced, leading to more scientifically grounded results. The following sections detail the construction of the evaluation指标体系, the methodology for weight determination and resilience evaluation, and the findings and recommendations for enhancing electric car supply chain resilience.
Evaluation Indicator System for Electric Car Supply Chain Resilience
The evaluation of electric car supply chain resilience requires a multidimensional指标体系 that reflects the industry’s unique characteristics. Based on resilience theory and dynamic capability theory, this study constructs a system comprising five primary indicators and twenty secondary indicators. The primary indicators include resistance and defense capability, adaptation and recovery capability, learning and transformation capability, sustainable development capability, and globalization capability. Each of these dimensions captures essential aspects of resilience, from risk mitigation to long-term sustainability and global integration. The secondary indicators provide granular insights into specific capabilities, such as digitalization level, decision-making agility, and ethical governance. This指标体系 was developed through literature analysis and expert consultation, ensuring its relevance and comprehensiveness for assessing China EV supply chains.
The following table summarizes the evaluation指标体系 for electric car supply chain resilience, including descriptions of each secondary indicator. This framework serves as the foundation for the subsequent weight determination and resilience evaluation processes.
| Primary Indicator | Secondary Indicator | Description |
|---|---|---|
| Resistance and Defense Capability | Risk Management Culture Penetration | Extent of risk awareness and regular response drills within the organization. |
| Supply Chain Digitalization Level | Application breadth and integration of digital technologies like blockchain and IoT in procurement, production, and logistics. | |
| Smart Logistics Network Redundancy | Availability and rapid deployment of backup logistics resources during disruptions. | |
| Geopolitical Sensitivity | Ability to anticipate and prepare for geopolitical changes affecting the supply chain. | |
| Adaptation and Recovery Capability | Agile Decision-Making Level | Speed and effectiveness of cross-departmental decision-making in response to risks. |
| Information Sharing Depth | Real-time sharing and interaction of key information among supply chain partners. | |
| Regional Production Capacity Substitution Capability | Ability to quickly mobilize alternative production capacities in different regions. | |
| Emergency Plan Completeness | Comprehensiveness and feasibility of contingency plans for various risk scenarios. | |
| Government-Enterprise Resource Coordination Efficiency | Effectiveness of collaboration with government entities in resource allocation and policy support during crises. | |
| Learning and Transformation Capability | Disruptive Technology Transformation Capability | Ability to convert cutting-edge technologies, such as new battery systems, into practical applications. |
| Ecological Synergy Innovation Depth | Depth of collaboration with partners in technological and model innovation. | |
| Organizational Experience Iteration Speed | Speed at which lessons from past crises are integrated into process improvements. | |
| Supply Chain Training System Maturity | Comprehensiveness of training programs for supply chain personnel. | |
| Global Technology Collaboration Closeness | Intensity of international partnerships in technology development and resource integration. | |
| Sustainable Development Capability | Carbon Neutrality Practice Leadership | Effectiveness in achieving carbon neutrality across the supply chain and industry influence. |
| Community Integration Development Contribution | Contributions to local communities through employment, infrastructure, and social development. | |
| Ethical Governance Mechanism Maturity | Maturity of governance systems addressing labor rights and environmental protection. | |
| Globalization Capability | Transnational Policy Adaptation Ability | Ability to quickly adapt to and comply with diverse international policies and standards. |
| Global Resource Strategic Control | Strategic layout and control over global critical resources like lithium and semiconductors. | |
| Crisis Diplomatic Mediation Capability | Effectiveness in using diplomatic means to resolve transnational supply chain crises. |
Methodology for Weight Determination and Resilience Evaluation
The evaluation of electric car supply chain resilience involves two key steps: determining indicator weights using game theory combination weighting and applying the extension cloud model for resilience level assessment. The weight determination combines subjective weights from trapezoidal fuzzy AHP and objective weights from an improved Critic method, optimized through game theory to achieve a balanced combination. This approach minimizes biases and enhances the scientific rigor of the evaluation. The extension cloud model then translates qualitative resilience levels into quantitative measures, handling the模糊性 and randomness inherent in the indicators. Below, the mathematical formulations and procedures are detailed.
Game Theory Combination Weighting
The subjective weights are derived using trapezoidal fuzzy AHP, which addresses uncertainties in expert judgments. Experts provide trapezoidal fuzzy numbers for pairwise comparisons, which are aggregated into a comprehensive fuzzy judgment matrix. Let $$r_{ij} = (a_{ij}, b_{ij}, c_{ij}, d_{ij})$$ represent the trapezoidal fuzzy number for the comparison between indicators i and j, aggregated from l experts as:
$$r_{ij} = \frac{1}{l} \left( r_{ij}^{(1)} \oplus r_{ij}^{(2)} \oplus \cdots \oplus r_{ij}^{(l)} \right) = \left[ \frac{1}{l} \sum_{k=1}^{l} a_{ij}^{(k)}, \frac{1}{l} \sum_{k=1}^{l} b_{ij}^{(k)}, \frac{1}{l} \sum_{k=1}^{l} c_{ij}^{(k)}, \frac{1}{l} \sum_{k=1}^{l} d_{ij}^{(k)} \right]$$
The matrix is then converted to a crisp value for consistency check, and the fuzzy evaluation expectation $$I(v_j)$$ is calculated. The subjective weight $$w_{1j}$$ for indicator j is obtained by normalizing $$I(v_j)$$:
$$w_{1j} = \frac{I(v_j)}{\sum_{i=1}^{n} I(v_j)}, \quad j \in N$$
For objective weights, the improved Critic method incorporates the coefficient of variation to account for differences in data scales. The original evaluation matrix $$X = (x_{ij})_{n \times m}$$ is standardized using Z-score method. The variation coefficient $$v_j$$ and independence coefficient $$\eta_j$$ are computed, leading to the comprehensive coefficient $$C_j$$. The objective weight $$w_{2j}$$ is given by:
$$w_{2j} = \frac{C_j}{\sum_{j=1}^{m} C_j}$$
Game theory is applied to combine the subjective and objective weights by minimizing the relative information entropy. The combined weight $$W_j$$ is derived as:
$$W_j = \frac{\sqrt{w_{1j} w_{2j}}}{\sum_{j=1}^{n} \sqrt{w_{1j} w_{2j}}}$$
This combination ensures that the weights reflect both expert insights and data-driven analysis, enhancing the reliability of the electric car supply chain resilience evaluation.
Extension Cloud Model for Resilience Evaluation
The extension cloud model integrates matter-element theory with cloud modeling to handle qualitative and quantitative transformations. The matter-element for evaluation is defined as $$R = [N, C, V]$$, where N is the object (e.g., electric car supply chain), C is the characteristic (indicator), and V is the value. Resilience levels are divided into five grades: low resilience, relatively low resilience, medium resilience, relatively high resilience, and high resilience. The standard cloud parameters for each grade are calculated based on the interval boundaries. For a resilience grade interval [I_min, I_max], the expected value Ex, entropy En, and hyper-entropy He are computed as:
$$Ex = \frac{I_{\text{max}} + I_{\text{min}}}{2}, \quad En = \frac{I_{\text{max}} – I_{\text{min}}}{6}, \quad He = k$$
where k is a constant, typically set to 0.005 for this evaluation. The standard cloud parameters for the five resilience grades are summarized in the following table.
| Resilience Grade | Interval | Standard Cloud Parameters (Ex, En, He) |
|---|---|---|
| Low Resilience | (0.0, 0.2] | (0.000, 0.017, 0.005) |
| Relatively Low Resilience | (0.2, 0.4] | (0.300, 0.033, 0.005) |
| Medium Resilience | (0.4, 0.6] | (0.500, 0.033, 0.005) |
| Relatively High Resilience | (0.6, 0.8] | (0.700, 0.033, 0.005) |
| High Resilience | (0.8, 1.0] | (1.000, 0.017, 0.005) |
For each secondary indicator, expert scores are collected and transformed into cloud parameters using the inverse cloud generator. The cloud parameters for an indicator i are calculated as:
$$Ex_i = \frac{1}{n} \sum_{i=1}^{n} x_i, \quad En_i = \sqrt{\frac{\pi}{2}} \frac{1}{n} \sum_{i=1}^{n} |x_i – Ex_i|, \quad He_i = \sqrt{ \left| \frac{1}{n} \sum_{i=1}^{n} (x_i – Ex_i)^2 – En_i^2 \right| }$$
where $$x_i$$ are the sample scores. The comprehensive cloud parameters for primary indicators and the overall resilience are computed using virtual cloud algorithms. For primary indicators, the floating cloud algorithm is applied:
$$Ex = \frac{Ex_1 w_1 + Ex_2 w_2 + \cdots + Ex_m w_m}{w_1 + w_2 + \cdots + w_m}, \quad En = \frac{En_1 w_1^2 + En_2 w_2^2 + \cdots + En_m w_m^2}{w_1^2 + w_2^2 + \cdots + w_m^2}, \quad He = \frac{He_1 w_1^2 + He_2 w_2^2 + \cdots + He_m w_m^2}{w_1^2 + w_2^2 + \cdots + w_m^2}$$
For the overall resilience, the comprehensive cloud algorithm is used:
$$Ex = \frac{Ex_1 En_1 w_1 + Ex_2 En_2 w_2 + \cdots + Ex_m En_m w_m}{En_1 w_1 + En_2 w_2 + \cdots + En_m w_m}, \quad En = En_1 w_1 + En_2 w_2 + \cdots + En_m w_m, \quad He = \frac{He_1 En_1 w_1 + He_2 En_2 w_2 + \cdots + He_m En_m w_m}{En_1 w_1 + En_2 w_2 + \cdots + En_m w_m}$$
The resilience grade is determined by comparing the comprehensive cloud with standard clouds, using cloud maps and membership degrees. The membership degree $$\mu(x)$$ for a value x in a cloud is calculated as:
$$\mu(x) = e^{-\frac{(x – Ex)^2}{2 y^2}}$$
where y is a normal random number with mean En and variance He². By running the forward cloud generator multiple times (e.g., 3000 iterations), the average membership degrees for each resilience grade are computed, and the grade with the highest membership is assigned. This process ensures a robust evaluation of the electric car supply chain resilience, accounting for uncertainties in the China EV context.
Evaluation Results and Analysis
The evaluation of electric car supply chain resilience was conducted based on data from experts in the field of China EV supply chain management. The weights for indicators were determined using the game theory combination approach, and the extension cloud model was applied to assess resilience levels. The results indicate that the overall resilience of the electric car supply chain is “relatively high,” with a comprehensive cloud parameter of Ex = 0.7099, En = 0.0249, and He = 0.0050. This aligns with the practical performance of China EV supply chains, which have demonstrated strong capabilities in adapting to market fluctuations and technological changes. However, variations exist across different dimensions, highlighting areas for improvement.
The following table presents the cloud parameters and resilience grades for primary and secondary indicators. The primary indicators show that adaptation and recovery capability, resistance and defense capability, learning and transformation capability, and sustainable development capability are all at “relatively high resilience,” while globalization capability is at “medium resilience.” Among secondary indicators, supply chain digitalization level and regional production capacity substitution capability achieve “high resilience,” whereas crisis diplomatic mediation capability is at “relatively low resilience.” These findings provide valuable insights for prioritizing actions to enhance the resilience of electric car supply chains.
| Primary Indicator | Cloud Parameters (Ex, En, He) | Resilience Grade | Secondary Indicator | Cloud Parameters (Ex, En, He) | Resilience Grade |
|---|---|---|---|---|---|
| Resistance and Defense Capability | (0.7387, 0.0203, 0.0054) | Relatively High Resilience | Risk Management Culture Penetration | (0.7270, 0.0198, 0.0052) | Relatively High Resilience |
| Supply Chain Digitalization Level | (0.8790, 0.0165, 0.0050) | High Resilience | |||
| Smart Logistics Network Redundancy | (0.6754, 0.0213, 0.0048) | Relatively High Resilience | |||
| Geopolitical Sensitivity | (0.5688, 0.0479, 0.0145) | Medium Resilience | |||
| Adaptation and Recovery Capability | (0.7522, 0.0283, 0.0047) | Relatively High Resilience | Agile Decision-Making Level | (0.7820, 0.0256, 0.0061) | Relatively High Resilience |
| Information Sharing Depth | (0.6860, 0.0271, 0.0039) | Relatively High Resilience | |||
| Regional Production Capacity Substitution Capability | (0.8090, 0.0316, 0.0025) | High Resilience | |||
| Emergency Plan Completeness | (0.7124, 0.0243, 0.0042) | Relatively High Resilience | |||
| Government-Enterprise Resource Coordination Efficiency | (0.7540, 0.0877, 0.0158) | Relatively High Resilience | |||
| Learning and Transformation Capability | (0.7316, 0.0282, 0.0052) | Relatively High Resilience | Disruptive Technology Transformation Capability | (0.7761, 0.0163, 0.0034) | Relatively High Resilience |
| Ecological Synergy Innovation Depth | (0.7520, 0.0256, 0.0072) | Relatively High Resilience | |||
| Organizational Experience Iteration Speed | (0.6850, 0.0301, 0.0087) | Relatively High Resilience | |||
| Supply Chain Training System Maturity | (0.6190, 0.0288, 0.0065) | Relatively High Resilience | |||
| Global Technology Collaboration Closeness | (0.7432, 0.0431, 0.0107) | Relatively High Resilience | |||
| Sustainable Development Capability | (0.6953, 0.0231, 0.0059) | Relatively High Resilience | Carbon Neutrality Practice Leadership | (0.7967, 0.0195, 0.0051) | Relatively High Resilience |
| Community Integration Development Contribution | (0.6273, 0.0178, 0.0047) | Relatively High Resilience | |||
| Ethical Governance Mechanism Maturity | (0.5723, 0.0331, 0.0081) | Medium Resilience | |||
| Globalization Capability | (0.5991, 0.0254, 0.0036) | Medium Resilience | Transnational Policy Adaptation Ability | (0.6547, 0.0251, 0.0018) | Relatively High Resilience |
| Global Resource Strategic Control | (0.5980, 0.0213, 0.0061) | Medium Resilience | |||
| Crisis Diplomatic Mediation Capability | (0.3810, 0.0501, 0.0193) | Relatively Low Resilience |
The cloud maps for specific secondary indicators, such as supply chain digitalization level and crisis diplomatic mediation capability, visually confirm the resilience grades. For instance, the cloud drops for supply chain digitalization level are concentrated in the “high resilience” region, while those for crisis diplomatic mediation capability align with “relatively low resilience.” The overall resilience cloud map shows a strong overlap with the “relatively high resilience” standard cloud, validating the evaluation result. The membership degrees from 3000 simulations further support this, with the highest average membership of 0.7790 for “relatively high resilience.” This comprehensive analysis underscores the strengths and weaknesses in the electric car supply chain, particularly for China EV, and guides targeted improvements.
Conclusions and Recommendations
This study presents a novel approach for evaluating the resilience of electric car supply chains using game theory combination weighting and the extension cloud model. The results demonstrate that the overall resilience is “relatively high,” with strong performance in adaptation, defense, learning, and sustainability capabilities, but moderate performance in globalization. Key secondary indicators like supply chain digitalization and regional capacity substitution exhibit high resilience, while geopolitical sensitivity, ethical governance, and crisis diplomacy require enhancement. These findings are consistent with the operational realities of China EV supply chains, where digital transformation and regionalization strategies have bolstered resilience, but global uncertainties pose ongoing challenges.
To further strengthen the electric car supply chain resilience, several recommendations are proposed. First, address weaknesses in globalization capability by increasing investments in global resource strategic control, such as through overseas partnerships and diversified sourcing for critical materials like lithium. Enhance crisis diplomatic mediation by fostering multilateral cooperation mechanisms to resolve trade disputes and geopolitical tensions. Second, consolidate advantages in digitalization and regional capacity by expanding the application of blockchain and IoT technologies across the supply chain and establishing cross-regional产能 backup systems. Third, promote synergy across the supply chain through information-sharing platforms and public-private partnerships, improving coordination and resource allocation during disruptions. Finally, advance research on the relationship between resilience and performance metrics, such as China EV export volumes, to quantify the value of resilience investments and inform policy decisions.
In conclusion, the integration of game theory combination weighting and extension cloud model provides a robust framework for assessing and improving electric car supply chain resilience. This approach not only addresses the模糊性 and randomness in evaluation but also offers practical insights for industry stakeholders. As the electric car sector continues to evolve, particularly in China EV markets, prioritizing resilience will be essential for sustaining growth and navigating complex global risks. Future work could explore dynamic resilience modeling and the impact of emerging technologies on supply chain robustness, further enriching the understanding of this critical domain.