3.1 Research problem description

By examining current cross-industry cooperative R&D models for intelligent vehicles globally, this paper proposes four cooperation models based on the core enterprises in the supply chain (as shown in Table 2). These include the technology-dominant, supplier-cooperation, manufacturer-dominant, and joint decision-making models, each reflecting different power dynamics and strategic choices in real-world cooperation between core component suppliers and vehicle manufacturers. The technology-dominant model arises when core suppliers like Huawei or Waymo lead innovation and control both R&D and pricing, typically in markets driven by rapid technological advancements. In the supplier-cooperation model, suppliers such as CATL (with BMW) provide key technologies, but vehicle manufacturers retain pricing control to adjust for market demand. The manufacturer-dominant model occurs when vehicle manufacturers like Tesla or XPeng, with strong in-house R&D capabilities and market influence, oversee both innovation and pricing. Finally, the joint decision-making model is seen in highly integrated partnerships (e.g., Honda-GM), where suppliers and manufacturers share R&D and pricing responsibilities to address complex market needs.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Classification of cooperation modes.

https://doi.org/10.1371/journal.pone.0321903.t002

This paper first derives the optimal decisions and profit functions for each supply chain participant under the four cooperation models. It then compares the optimal profit, R&D effort, vehicle pricing decisions, and overall supply chain profits across different models, analyzing the influence of key parameters on the equilibrium results. The notations and descriptions used in this paper are shown in Table 3.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Notations and descriptions.

https://doi.org/10.1371/journal.pone.0321903.t003

3.2 Fundamental assumptions

Assume there is a risk-neutral technology firm serving as the core component supplier, primarily responsible for providing intelligent vehicle technologies and components such as autonomous driving systems, intelligent connectivity solutions, and battery management systems. Meanwhile, the vehicle manufacturer is responsible for vehicle manufacturing, including body design, assembly, and quality control. To keep pace with the rapid evolution of vehicle intelligence, the core component supplier and the vehicle manufacturer engage in cooperative R&D to enhance their competitiveness. In this cooperation, the vehicle manufacturer not only procures intelligent components but also receives technical R&D support from the core component supplier. Furthermore, both parties share R&D costs to foster continuous innovation and accelerate the application of advanced technologies [46]. This cooperative R&D not only leads to product or service quality improvements but also reduces production costs. For example, in the cooperation between Huawei and Seres, they not only jointly established the Seres Super Factory but also cooperated in developing the AITO series, enhancing both production efficiency and product intelligence. The above is shown in Fig 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Supply chain cooperative research and development model.

https://doi.org/10.1371/journal.pone.0321903.g001

Based on the above description and to simplify the game decision-making model for cooperative R&D, we propose the following basic assumptions:

1. (1) Unit production cost of a vehicle manufacturing enterprise C is a function of the R&D effort e, specifically . Where is the initial production cost, θ is the sensitivity factor of R&D cost reduction. A higher θ indicates a higher degree of cost reduction due to R&D efforts. However, there is a lower bound to the cost reduction, that is, there exists a minimum material consumption cost denoted as . In summary, the production cost satisfies the following condition, , .
2. (2) The automobile demand function is . Where β represents the sensitivity coefficient of consumer product quality, and the higher the β, the higher the degree of increase of R&D efforts on demand [6].
3. (3) According to the research of Zheng YL et al. [6] and the real situation, the technical service fee charged by the core component supplier T is a function of the R&D effort e, specifically . Where k is the unit research and development efforts technical service fee, is a fixed service fee.
4. (4) With reference to the studies of Atasu and Subramanian [47] and Ranjan and Jha [48], it is assumed that the cost of R&D investment is a quadratic function of R&D investment level, i.e., . Where, e represents the level of R&D effort of the enterprise, h represents the cost coefficient of R&D effort of the enterprise, and the smaller h is, the higher the R&D efficiency.

4.1 Technology-dominant model

In this model, the core component supplier determines the R&D effort and sets the vehicle price , while the vehicle manufacturer will only agree to cooperate if its profit is non-negative. Consequently, it is assumed that the vehicle manufacturer will offer the core component supplier a minimum price . In this scenario, the core component supplier becomes the dominant entity in the supply chain, relegating the vehicle manufacturer to a lower position, effectively serving as a contract manufacturer. For example, in the ‘Huawei Smart Selection’ cooperative R&D model between Huawei and Seres, Huawei leads the design, sales, production, and services for the AITO intelligent car brand. Both the R&D effort level and vehicle pricing for AITO are determined by Huawei. The assumptions in the subsequent sections follow the same logic as those in this section. Similar to previous literature [6,7], the profit function of the core component supplier can be expressed as:

(1)

The profit function of the vehicle manufacturer is:

(2)

The objective function of the core component supplier is .

Taking the first derivative of equation (1) with respect to gives: , which indicates that the profit function of the core component supplier decreases with increasing price . Consequently, the core component supplier will set the price at the base price , as provided by the vehicle manufacturer, so:

(3)

Substitute equation (3) into equation (1), we can get:

(4)

Then, we can find the first derivatives of equation (4) with respect to :

(5)

And the second derivatives of equation (4) with respect to is:

(6)

According to equation (6), reaches a maximum, which occurs at . Setting equation (5) equals to 0, the optimal R&D effort level under the technology-dominant model can be obtained as:

(7)

By substituting equations (3) and (7) into (1), we can obtain the optimal profit of the core component supplier under the technology-dominant model:

(8)

By substituting equations (3) and (7) into (2), we can obtain the optimal profit of the vehicle manufacturer under the technology-dominant model:

(9)

Where .

Theorem 1: Under the technology-dominant model.

The optimal vehicle selling price is: .

The optimal R&D effort level is: .

The profit of core component supplier is:

The profit of the vehicle manufacturer is:

4.2 Supplier-cooperation model

In the supplier-cooperation model, the core component supplier first decides the R&D effort , and the vehicle manufacturer subsequently determines the vehicle pricing . In this model, the core component supplier and the vehicle manufacturer hold relatively equal positions in the supply chain. For instance, in the cooperative R&D between Huawei and Changan, Huawei leads the R&D efforts, while Changan retains control over vehicle pricing for the Avatr cooperative model. The profit function of the core component supplier is:

(10)

The profit function of the vehicle manufacturer is:

(11)

The objective function of the core component supplier is and the objective function of the vehicle manufacturer is .

Backward induction is used to solve, and we can find the first derivatives of equation (11) with respect to :

(12)

And the second derivatives of equation (11) with respect to is:

(13)

According to equation (13), reaches a maximum, which occurs at . By setting equation (12) equals to 0, we obtain:

(14)

By substituting equation (14) into the profit function (10), we obtain:

(15)

Then, we can find the first derivatives of equation (15) with respect to :

(16)

And the second derivatives of equation (15) with respect to is:

(17)

According to equation (17), reaches a maximum, which occurs at . By setting equation (16) equals to 0, we obtain:

(18)

By substituting equation (18) into (14), we obtain:

(19)

By substituting equation (18) and (19) into equation (10), we can obtain the optimal profit of the core component suppliers under the supplier-cooperation model:

(20)

By substituting equation (18) and (19) into equation (11), we can obtain the optimal profit of the vehicle manufacturers under the supplier-cooperation model:

(21)

Where .

Theorem 2: Under the supplier-cooperation model.

The optimal vehicle selling price is:

The optimal R&D effort level is:

The profit of core component supplier is:

The profit of the vehicle manufacturer is:

4.3 Manufacturer-dominant model

The vehicle manufacturer decides both the R&D effort and the vehicle pricing . In this model, the core component supplier functions solely as a provider, with the vehicle manufacturer holding the dominant position in the supply chain. For instance, in the cooperation between Huawei and Audi, Huawei supplies intelligent vehicle components and basic R&D services, while Audi retains full control over the R&D effort and vehicle pricing for its cars. The profit function of the core component supplier is:

(22)

The profit function of the vehicle manufacturer is:

(23)

The objective function of the vehicle manufacturer is .

When the level of R&D effort is fixed, we can find the first derivatives of equation (23) with respect to :

(24)

And the second derivatives of equation (23) with respect to is:

(25)

By setting equation (24) equals to 0, we obtain:

(26)

By substituting equation (26) into (23), we obtain:

(27)

We can find the first derivatives of equation (27) with respect to :

(28)

And the second derivatives of equation (27) with respect to is:

(29)

By setting equation (28) equals to 0, we obtain the optimal R&D efforts under the manufacturer-dominant model:

(30)

Assume that . Here, represents the demand positively influenced by process and product innovation driven by R&D, while represents the demand negatively influenced by the technical service fees paid by the vehicle manufacturer. Only when the positive demand impact exceeds the negative impact will the core component supplier and the vehicle manufacturer engage in cooperative R&D.

Additionally, when , . Due to the significant R&D costs of intelligent vehicles, h is sufficiently large to satisfy , thus ensuring .

By substituting equation (30) into (26), we obtain the optimal vehicle price under the manufacturer-dominant model:

(31)

By substituting equation (30) and (31) into (22), we obtain the optimal profit of the core component supplier under the manufacturer-dominant model:

(32)

By substituting equation (30) and (31) into (23), we obtain the optimal profit of the vehicle manufacturer under the manufacturer-dominant model:

(33)

Where .

Theorem 3: Under the manufacturer-dominant model.

The optimal vehicle selling price is:

The optimal R&D effort level is:

The profit of core component supplier is:

The profit of the vehicle manufacturer is:

4.4 Joint decision-making model

The core component supplier and the vehicle manufacturer jointly decide the R&D effort and the vehicle price . Honda, for example, has partnered with GM to develop vehicle platforms, jointly develop technology, and jointly set prices. The profit function of the supply chain is:

(34)

The objective function of the supply chain is:

When the level of R&D effort is fixed, we can find the first derivatives of equation (34) with respect to :

(35)

And the second derivatives of equation (34) with respect to is:

(36)

By setting equation (35) equals to 0, we obtain:

(37)

By substituting equation (37) into (34), we obtain:

(38)

We can find the first derivatives of equation (38) with respect to :

(39)

And the second derivatives of equation (38) with respect to is:

(40)

By setting equation (39) equals to 0, we obtain the optimal R&D efforts under the joint decision-making model:

(41)

By substituting (41) into (37), we obtain the optimal vehicle price under the joint decision-making model:

(42)

Then we can obtain the optimal profit of the core component supplier under the joint decision-making model:

(43)

And the optimal profit of the vehicle manufacturer under the joint decision-making model is:

(44)

Where .

Theorem 4: Under the joint decision-making model.

The optimal vehicle selling price is:

The optimal R&D effort level is: .

The profit of core component supplier is:

The profit of the vehicle manufacturer is:

5.1 Optimization decision analysis of cooperative R&D model

Proposition 2.

An increase in consumers’ sensitivity to product quality, coupled with a decrease in sensitivity to cost-reducing R&D, will lead to a rise in the vehicle’s price [21]. Additionally, in the technology-dominated model, vehicle prices remain consistently low. In the supplier cooperation model, a greater share of R&D costs borne by the vehicle manufacturer leads to higher prices. In the manufacturer-dominated model, a higher share of R&D costs borne by the vehicle manufacturer results in lower prices. Specifically,

Proposition 2 reveals the impact of consumers’ quality sensitivity, R&D cost-reduction sensitivity, and R&D cost-sharing on the vehicle’s price under different cooperation models. An increase in consumer sensitivity to product quality drives up vehicle prices, reflecting the heightened R&D investments required to meet this demand. For instance, companies like NIO and Tesla justify the premium pricing of models such as the ES8 and Model S by offering advanced features like autonomous driving and intelligent systems, which are the direct result of increased R&D efforts. Moreover, in the technology-dominated model, the optimal vehicle price is not influenced by either the consumer’s sensitivity to quality or the sensitivity of R&D to cost reduction. Combined with Proposition 1, this indicates that core component suppliers set lower prices to quickly capture market share, often without focusing on R&D-driven innovations. While this strategy boosts sales volume, it limits vehicle manufacturers’ profit margins, a dynamic observed in Waymo’s partnerships with vehicle manufacturers for autonomous driving technology. In the supplier-cooperation model, as the proportion of R&D costs borne by the manufacturer increases, the vehicle price also rises, and according to Proposition 1, the R&D effort level increases as well. For example, in the CATL-BMW partnership, CATL covers most battery R&D costs while BMW sets vehicle pricing. As BMW’s R&D share rises, the vehicle price increases, and the company invests more in intelligent features. Conversely, in the manufacturer-dominant model, a higher share of R&D costs borne by the manufacturer leads to a lower vehicle price, along with a decrease in R&D effort. For example, as BYD increased its proportion of in-house R&D, it reduced the level of intelligent features in certain models and set extremely low prices to remain competitive.

5.2 Comparative analysis of cooperative research and development models

5.3 Analysis of influencing factors on the optimal R&D effort and vehicle price

Proposition 7.

Under the supplier cooperation model, the relative contribution of the consumer quality sensitivity coefficient and R&D cost sensitivity coefficient to the optimal product price is affected by consumer preference and R&D cost. Specifically, by setting , we obtain:

Proposition 7 reveals that as consumer price sensitivity increases or R&D costs decrease, the contribution of cost reduction to optimal pricing surpasses that of quality sensitivity. Conversely, when price sensitivity decreases or R&D costs rise, quality sensitivity plays a larger role in determining prices. When price sensitivity is high, R&D efforts to reduce component costs allow vehicle manufacturers to lower vehicle prices, making cost reduction a key driver of pricing. Conversely, when consumers prioritize quality over price, vehicle manufacturers focus on product enhancements and set higher prices to meet demand.

6.1 Comparison of R&D effort and analysis of influencing factors

As shown in Figs 2 and 3, across all four cooperation models, the optimal R&D effort increases with both the quality sensitivity coefficient and the product R&D cost reduction sensitivity coefficient, indicating a positive correlation between these coefficients and R&D investment. Whether improving product quality or reducing R&D costs, both drive firms to boost R&D efforts to stay competitive, supporting Proposition 1. Moreover, the simulation results in Figs 2 and 3 show that the impact of consumers’ quality sensitivity coefficient and the R&D cost-reduction sensitivity coefficient on the optimal R&D effort is consistent [21]. In the manufacturer-dominant model, the two coefficients have the greatest impact on the optimal R&D effort level, and the optimal R&D effort level is the highest at this point. Additionally, Figs 2 and 3 reveal that the technology-dominated model surpasses the supplier cooperation model in terms of optimal R&D effort when consumer quality sensitivity and R&D cost reduction sensitivity are high. This suggests that in the technology-dominated model, core component suppliers are more likely to increase R&D efforts compared to the supplier cooperation model when these factors are critical. According to the simulation results, if the cooperative R&D in the supply chain aims for a high level of intelligence, the manufacturer-dominant model is a more suitable choice.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. The influence of β on e^*.

https://doi.org/10.1371/journal.pone.0321903.g002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. The influence of θ on e^*.

https://doi.org/10.1371/journal.pone.0321903.g003

As shown in Fig 4, the optimal R&D effort in the technology-dominated and supplier cooperation models decreases as core component suppliers bear a larger share of R&D costs. In contrast, the joint decision-making model remains unaffected. In the manufacturer-dominated model, the lower the proportion of R&D costs borne by the manufacturer, the greater the R&D effort. In the joint decision-making model, R&D effort remains stable regardless of cost-sharing, ensuring balanced cooperation. This highlights how supply chain participants optimize resource allocation for efficiency and long-term cooperation, supporting Proposition 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. The influence of τ on e^*.

https://doi.org/10.1371/journal.pone.0321903.g004

6.2 Comparison of vehicle selling price and analysis of influencing factors

As shown in Figs 5 and 6, due to the minimal impact of the quality sensitivity coefficient and the R&D cost reduction sensitivity coefficient on vehicle pricing, the base price is set to zero in the simulation in Section 6.2 for clearer observation of these correlations. In all models except the technology-dominated model, vehicle prices rise as these coefficients increase, supporting Propositions 2 and 5. Additionally, the results show a negative correlation between the R&D cost reduction sensitivity and vehicle pricing: as R&D reduces production costs, prices decrease to remain competitive, aligning with Proposition 2. However, the overall effect of both coefficients on pricing remains small.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. The influence of β on p^*.

https://doi.org/10.1371/journal.pone.0321903.g005

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. The influence of θ on p^*.

https://doi.org/10.1371/journal.pone.0321903.g006

As shown in Fig 7, vehicle pricing is independent of the manufacturer’s R&D cost share in the technology-dominated and joint decision-making models. However, in the supplier cooperation model, there is a negative correlation between vehicle pricing and the core component suppliers’ R&D share, while in the manufacturer-dominated model, this correlation is positive. This reflects that in the manufacturer-dominated model, where the manufacturer controls R&D decisions, prices rise as core component suppliers take on more R&D costs, consistent with Proposition 2. In addition, while the manufacturer’s R&D share impacts pricing in some models, the overall effect is limited. This suggests that factors like market competition, consumer demand, and economies of scale are more influential in determining vehicle pricing than the R&D cost share. For instance, Toyota’s Prius pricing strategy remained stable despite significant R&D investment, as economies of scale played a larger role in cost management.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 7. The influence of τ on p^*.

https://doi.org/10.1371/journal.pone.0321903.g007

6.3 Profit comparison and analysis of influencing factors

As shown in Fig 8, the impact of consumer quality sensitivity and R&D cost reduction sensitivity on core component supplier profits varies by cooperation model. For core component suppliers, profits are highest in the technology-dominant model. Therefore, when selecting a cooperative model, core component suppliers should aim to establish themselves as the industry’s key players and adopt a technology-dominant model; if not a core player, the joint decision-making model is preferable. Additionally, in the technology-dominant model, the core component supplier’s profit increases with the consumer quality sensitivity coefficient. Acting as the dominant player, the supplier may use promotional strategies to elevate this coefficient to maximize profits, as it positively correlates with R&D effort. Consequently, the supply chain adopts a quality-focused strategy, consistent with Proposition 3.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 8. The influence of β and θ on Π^s.

https://doi.org/10.1371/journal.pone.0321903.g008

As shown in Fig 9, vehicle manufacturer profits increase in the supplier-cooperation, manufacturer-dominant, and joint decision-making models as consumer quality sensitivity and R&D cost reduction sensitivity rise, aligning with Proposition 3. Additionally, vehicle manufacturer profits are highest in the manufacturer-dominant model [22–24]. Therefore, when choosing a cooperative model, vehicle manufacturers should aim to establish themselves as industry leaders and select the manufacturer-dominant model; if they are not key players, they should consider the joint decision-making model in highly competitive markets or the supplier-cooperation model when competition is less intense. With an increase in consumer quality sensitivity and R&D cost reduction sensitivity, profits under the joint decision model will exceed those in the supplier-cooperation model. It is also worth noting that, under this simulation, profits in the technology-dominant model turned negative, which is why it is not shown in Fig 9, suggesting that the technology-dominant model may be unfavorable for vehicle manufacturers.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 9. The influence of β and θ on Π^z.

https://doi.org/10.1371/journal.pone.0321903.g009

Additionally, based on Proposition 3, Figs 8 and 9, we observe that in the supplier-cooperation model, the profits of both core component suppliers and vehicle manufacturers are positively correlated with consumer quality sensitivity and R&D cost reduction sensitivity. Consequently, the supply chain strives to increase these two coefficients, which also positively correlate with R&D effort levels, leading the supply chain to adopt a quality-focused strategy in the supplier-cooperation model. However, these two coefficients have opposing effects on vehicle pricing: one positive and one negative. According to Proposition 9, when consumer price sensitivity is high, the impact of R&D cost reduction sensitivity surpasses that of quality sensitivity. In summary, in the supplier-cooperation model, the supply chain adopts a quality-focused strategy and, as consumer price sensitivity rises, simultaneously pursues a low-price strategy.

As shown in Fig 10, supply chain profits in the joint decision-making model increase with higher consumer quality and R&D cost reduction sensitivity. Combining Figs 8 and 9, these coefficients have opposing effects on the core component supplier and vehicle manufacturer: they positively correlate with manufacturer profits but negatively with supplier profits. From a supply chain power structure perspective, the manufacturer typically holds a dominant position in the joint decision-making model, allowing it to shift the pressure of quality sensitivity and cost control onto the supplier, who then bears more costs to enhance product quality. This indicates that the resulting sales growth and market expansion from these coefficients more than offset the supplier’s reduced profits, thus boosting overall supply chain profitability.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 10. The influence of β and θ on Π^sc.

https://doi.org/10.1371/journal.pone.0321903.g010

As shown in Figs 11 and 12, with the increase in the consumer price sensitivity coefficient, moving from Fig 11 to Fig 12, it can be observed that the influence of the consumer quality sensitivity coefficient on the optimal R&D effort decreases, while the influence of the R&D cost reduction sensitivity coefficient increases. This is in line with Proposition 6.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 11. The influence of β and θ on e^*, b = 0.5.

https://doi.org/10.1371/journal.pone.0321903.g011

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 12. The influence of β and θ on e^*, b = 1.5.

https://doi.org/10.1371/journal.pone.0321903.g012

As shown in Figs 13 and 14, as the consumer price sensitivity coefficient increases, moving from Fig 13 to Fig 14, the influence of the consumer quality sensitivity coefficient on the optimal vehicle price diminishes, while the influence of the R&D cost reduction sensitivity coefficient grows. This observation aligns with Proposition 7. Additionally, based on Figs 8, 9, and 11, in the manufacturer-dominant model, manufacturer profits positively correlate with both coefficients, even though supplier profits are negatively affected. Since the manufacturer is the dominant player, the supply chain adopts a quality-focused strategy and, when consumer price sensitivity is high, also implements a low-price strategy. In the joint decision-making model, supply chain profits positively correlate with both parameters; thus, similarly, the supply chain adopts a quality-focused strategy and adjusts the low-price strategy according to the level of consumer price sensitivity.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 13. The influence of β and θ on p^*, b = 0.5.

https://doi.org/10.1371/journal.pone.0321903.g013

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 14. The influence of β and θ on p^*, b = 1.2.

https://doi.org/10.1371/journal.pone.0321903.g014

7.1 Discussion

In response to the current R&D and price competition under cross-organizational cooperative models in the intelligent vehicle supply chain, this paper proposes four cooperative R&D models. The aim is to analyze the applicability conditions of different models, the shifts in supply chain power across models [6–9], and the strategic choices under each model [46], with a focus on the impact of product innovation and process innovation on R&D effort and pricing decisions. By analyzing and comparing the equilibrium outcomes of core component suppliers and vehicle manufacturers across the four models and evaluating the impact of various parameters on these equilibrium results, the following key conclusions are drawn.

First, we analyze the conditions under which different cross-organizational R&D cooperation models are applicable and the shifts in power structures associated with each model. We find that the profits of both core component suppliers and vehicle manufacturers are higher in models where they hold a dominant position compared to non-dominant positions [6, 22–24]. Therefore, when selecting a cooperative model, core component suppliers should choose the technology-dominant model, while vehicle manufacturers should opt for the manufacturer-dominant model. If not in a leading position, the joint decision-making model is more appropriate for both core component suppliers and vehicle manufacturers. Additionally, if the supply chain aims to maximize vehicle intelligence rather than short-term profits, the manufacturer-dominant model is preferable. This paper also reveals that when product innovation from R&D exceeds process innovation, pricing power shifts from vehicle manufacturers to core component suppliers, transitioning the supply chain from the supplier-cooperation model to the technology-dominant model. Conversely, when process innovation surpasses product innovation, pricing power reverts to vehicle manufacturers, shifting the supply chain from the technology-dominant model to the supplier-cooperation model.

Second, we analyze the strategic choices across different supply chain power structures. In the technology-dominant model, the supply chain prioritizes a quality-focused strategy. Although this may introduce a risk of price wars, it is not an intentional outcome. In contrast, in the supplier-cooperation model, the manufacturer-dominant model, and the joint decision-making model, the supply chain consistently adopts a quality strategy. However, the decision to implement a low-price strategy alongside the quality strategy depends on consumer price sensitivity. As consumer price sensitivity increases, the impact of the R&D cost reduction coefficient on vehicle pricing grows stronger than that of the product innovation coefficient, thereby making the cost-reduction effect more significant than the price increase driven by product innovation. Thus, as consumer price sensitivity increases, the likelihood of the supply chain simultaneously adopting both quality and low-price strategies rises. Conversely, when consumer price sensitivity is lower, the supply chain tends to rely solely on the quality strategy.

Finally, we examine the impact of key parameters on the equilibrium outcomes. We find consistent positive effects of process and product innovation coefficients on the optimal R&D effort level across various supply chain power structures [21]. However, while the process innovation coefficient is negatively correlated with the optimal vehicle price, the product innovation coefficient shows a positive correlation. A notable distinction appears in the manufacturer-dominated model, where both the optimal R&D effort and vehicle price are negatively correlated with the manufacturer’s R&D cost-sharing ratio. In contrast, in other power structures, these metrics are positively correlated with the manufacturer’s R&D cost-sharing ratio. The technology-dominant model consistently maintains the lowest optimal vehicle price, which remains unaffected by the product or process innovation coefficients or cost-sharing ratios. Simulation results further indicate that the manufacturer-dominant model achieves the highest level of product intelligence [5]. Additionally, as consumer price sensitivity rises, the influence of the R&D cost reduction coefficient on both R&D effort and vehicle price surpasses that of the quality sensitivity coefficient.

7.2 Management implications

Against the backdrop of high R&D costs in vehicle intelligence, this paper aims to explore the applicability of different supply chain cooperation models, power shifts, and strategic choices, providing valuable insights for technology companies and vehicle manufacturers on cooperation model selection, strategic choices, and R&D effort and pricing decisions within the context of cross-industry joint R&D in the intelligent vehicle supply chain. Specifically, it offers the following insights for participants in the intelligent vehicle supply chain.

First, companies should select cooperation models flexibly based on supply chain positioning and competitive advantages [8,15]. Since profits in a dominant position are consistently higher than those in a non-dominant role [6,22–24]. For instance, core component suppliers like Huawei or Google may benefit from a technology-dominant model to control R&D and innovation, while core vehicle manufacturers such as Tesla and BYD are better suited to a manufacturer-dominant model, enabling control over vehicle intelligence and cost management [16]. If neither party holds a core position, as with startups in emerging markets, a joint decision-making model allows resource sharing for mutual benefit. When cross-industry R&D has strategic long-term importance, such as developing advanced intelligence to establish competitive barriers, the manufacturer-dominant model is preferable. Additionally, the high technological demands of the intelligent vehicle industry make cooperation essential, with partnerships among vehicle manufacturers, tech firms, and data companies helping to complement resources and drive innovation. Such partnerships enable vehicle manufacturers to accelerate intelligent technology advancements, while component suppliers can leverage alliances with AI and autonomous driving firms to maintain their technological edge.

Second, intelligent vehicle companies should first leverage AI and data mining to deeply analyze consumer preferences, market dynamics, and technological trends, providing foundational support for strategic choices [23,42]. In different cooperation models, companies should adapt their strategies based on consumer quality and price sensitivity. In the technology-dominant model, a quality-focused strategy is recommended for the supply chain, although the potential risk of price wars should be anticipated and mitigated through differentiation and brand premium strategies. In the supplier-cooperation, manufacturer-dominant, and joint decision-making models, quality remains the core strategy, but a low-price strategy should be integrated if consumer price sensitivity is high to enhance market competitiveness. In these models, companies need to balance product quality enhancement with cost optimization to maximize profitability [44]. Furthermore, as consumer price sensitivity rises, companies should emphasize cost optimization through cross-organizational cooperation and process innovation, effectively sharing R&D cost pressures. This approach helps achieve an optimal balance between quality and cost, ensuring the sustainability and adaptability of strategic choices in a competitive market.

Finally, based on the research findings, companies in the intelligent vehicle supply chain should align their R&D and pricing strategies with their innovation priorities and supply chain power structures [23–25]. Specifically, intelligent vehicle supply chain companies should optimize R&D resource allocation by balancing process and product innovation to enhance product quality while controlling costs. In the manufacturer-dominant model, manufacturers can reduce their share of R&D costs to drive product intelligence and maintain a pricing advantage. Conversely, in supplier-cooperation and joint decision-making models, moderately increasing the cost-sharing ratio can help sustain innovation investment. In the technology-dominant model, the supply chain may adopt a low-price strategy to capture market share, particularly in price-sensitive markets. As consumer price sensitivity rises, companies should focus on process innovation and cost control, employing data analysis to dynamically monitor consumer price sensitivity and ensure that R&D direction and resource allocation align with market demands [24,42]. This data-driven management approach not only supports competitive pricing but also enables a balanced focus on product intelligence and cost optimization, thereby enhancing adaptability and long-term competitiveness in a rapidly changing market.

7.3 Limitations and future research

This paper can be further expanded and enriched in several ways. First, this paper treated the wholesale price of components as an exogenous variable, but future research could explore endogenous wholesale prices by examining the differences between joint R&D and independent R&D, especially linking wholesale prices to order volumes. Particularly for the procurement of high-cost intelligent components in intelligent vehicles, varying demand levels and different R&D cooperation models may lead to different wholesale pricing strategies. Second, while this paper focused on wholesale pricing contracts, real-world cooperation between core component suppliers and vehicle manufacturers often involves joint design, R&D, sales, and services, where revenue-sharing contracts are more common [46]. Investigating these contracts could provide valuable insights into alternative pricing mechanisms. Third, this paper only examined strategic choices in a single supply chain, but competition between different supply chain cooperation models also exists. Future research could incorporate these competitive dynamics to provide a more comprehensive understanding of the cross-industry R&D landscape. Lastly, previous studies by Yan et al. [11] and Wu et al. [12] suggest that privileged royalty licensing can effectively mitigate price competition and enable higher profit margins. Future research could explore how similar mechanisms might be applied to cooperative R&D to reduce price competition and enhance profitability across the supply chain.