Online car-hailing platform bilateral pricing optimization method and system based on user preference difference and multi-homing behavior

Abstract

The invention relates to the technical field of online car-hailing service platform pricing, and discloses an online car-hailing platform bilateral pricing optimization method and system based on user preference differences and multi-homing behaviors, and the method comprises the steps: obtaining the historical behavior data of passengers and drivers, quantifying the price sensitivity and quality sensitivity, and constructing user preference feature portraits; based on a significance theory and a Hotelling model, respectively constructing utility functions including heterogeneity preferences of passengers and drivers, and depicting selection behaviors of different user groups; the market structures of passenger single homing, multi-homing and driver single homing are distinguished, a double-homing platform competition model is constructed, and the pricing game process between platforms is simulated; solving the optimal pricing and profit function of the platform under the combination of the four user sensitive types through a reverse induction method; and analyzing the action mechanism of the significant thinking and the multi-homing behavior of the user on the pricing strategy. According to the method, the problems of pricing failure and profit loss caused by neglecting user preference heterogeneity and multi-platform use behaviors of a traditional online car-hailing platform are effectively solved, and the user matching efficiency, the resource allocation capability and the overall market competitiveness of the platform are remarkably improved through refined modeling and strategy optimization.

Description

Technical Field

[0001] This invention relates to the field of ride-hailing service provider management technology, and in particular to a method and system for optimizing two-sided pricing for ride-hailing platforms based on differences in user preferences and multi-attribution behavior. Background Technology

[0002] With the continuous penetration and cross-industry integration of mobile internet and information technology, especially in the context of deep integration with the traditional transportation industry, ride-hailing platforms have ushered in unprecedented development opportunities and are showing a booming trend. While the strong demand in the ride-hailing market is optimizing the travel experience, it has also given rise to a series of practical problems. The rise of "aggregation platforms" has reshaped the competitive landscape of the ride-hailing market, breaking the previous "one-superpower-many-strong" dominance represented by Didi. In a travel service market where multiple platforms coexist, passengers' decision-making behavior is not only influenced by objective factors such as price and service quality, but also exhibits obvious individual preferences. In fact, passengers do not make rational decisions entirely based on "cost-effectiveness," but rather show a significant preference for specific service attributes, giving them higher weight in the decision-making process, thus forming imperfectly irrational choice behavior based on attribute preferences. At the same time, since ride-hailing platforms and drivers usually do not have a traditional employment relationship, their participation often depends on the platform's salary structure and incentive measures, thus determining whether to join the platform to provide travel services. Therefore, drivers' preference behavior plays a crucial role in the design of platform operation strategies. Furthermore, as a medium connecting supply and demand, ride-hailing platforms exhibit an asymmetrical user affiliation structure. While passengers commonly use multiple platforms simultaneously, competition among platforms for transportation resources typically restricts drivers from accepting orders on multiple platforms concurrently, thus constraining their multi-affiliation behavior. Consequently, in this market, multi-affiliation is primarily observed on the passenger side, creating an asymmetrical affiliation pattern. Therefore, accurately understanding user preferences and developing differentiated pricing strategies for different user affiliations to achieve an effective balance between platform profits and market share has become a critical issue that ride-hailing platform operations urgently need to address. Summary of the Invention

[0003] This invention provides a two-sided pricing optimization method and system for ride-hailing platforms based on user preference differences and multi-attribution behavior. It provides a theoretical basis for ride-hailing platforms to optimize pricing mechanisms, identify core users, and address multi-attribution challenges. It also provides decision support for platforms to formulate differentiated pricing strategies, ensuring the maximization of ride-hailing platform profits.

[0004] To achieve the above objectives, the present invention employs the following technical solution:

[0005] In a first aspect, the present invention provides a two-sided pricing optimization method for ride-hailing platforms based on user preference differences and multi-attribution behavior, comprising:

[0006] (1) Obtain and analyze the historical behavior data of passengers and drivers, and construct a user preference profile by quantifying their sensitivity to service price and service quality.

[0007] (2) Based on the saliency theory in behavioral economics and combined with the Hotelling space competition model, a utility function that can simultaneously characterize the differentiated preferences of passengers and drivers in terms of service price and service quality is constructed, so as to more accurately simulate the user's choice behavior in real-world scenarios.

[0008] (3) Different market structure types are defined based on whether passengers adopt a single-attribution or multi-attribution strategy and whether drivers adopt a single-attribution strategy; on this basis, a duopoly platform competition model is constructed to simulate the dynamic pricing game process between platforms.

[0009] (4) For a market structure where both passengers and drivers are single-attributed, based on the principle of maximizing user utility, solve for the equilibrium number of passengers and drivers, equilibrium pricing and optimal profit of each platform under the four combinations of user preference types.

[0010] (5) For the market structure of multiple passenger affiliations and single driver affiliations, based on the principle of maximizing user utility, solve for the equilibrium number of passengers and drivers, equilibrium pricing and optimal profit of each platform under the four combinations of user preference types.

[0011] (6) Through numerical simulation, the impact of user preference differences and multi-attribution behavior on platform pricing strategy and profit level is analyzed, revealing the inherent law of platform optimal pricing and profit change with user sensitivity, and providing theoretical basis and decision support for platform to formulate differentiated pricing strategy.

[0012] Optionally, step (1) includes:

[0013] Use parameters Characterize the salience sensitivity of platform users (passengers and drivers) to service price and service quality. The smaller the value, the stronger the passenger's or driver's salient thinking, and the greater the difference in their preference for the two attributes of service price and service quality. For service price-sensitive passengers / drivers, the service price coefficient is... The service quality coefficient is .Depend on , can be obtained This indicates that, in this scenario, the marginal impact of service price on utility is dominant. Correspondingly, for service quality-sensitive passengers / drivers, the service price coefficient is... The service quality coefficient is .Depend on , can be obtained This indicates that service quality has a more significant impact on utility in this scenario. Based on historical behavioral data of passengers and drivers, users are categorized into four combinations: both sides are service price sensitive (PP), passengers are service price sensitive and drivers are service quality sensitive (PS), passengers are service quality sensitive and drivers are service price sensitive (SP), and both sides are service quality sensitive (SS).

[0014] Optionally, the utility functions for the passenger and driver in step (2) are as follows:

[0015] In the formula, and These respectively represent passengers and drivers on the platform. The effect, and These represent the basic utility of passengers and drivers joining a ride-hailing platform, respectively. This indicates that passengers have joined the platform. The service price charged This indicates that the driver has joined the platform. Commissions extracted, and These respectively represent passengers and drivers on the platform. Quantity, and Represent the cross-network externality strengths of passengers and drivers, respectively. and These represent the travel costs for passengers and drivers, respectively.

[0016] Optionally, the platform profit function in steps (4) and (5) is:

[0017] In the formula, Indicates platform Profits Indicates the platform to which the passenger belongs. The quantity.

[0018] Optionally, steps (3) and (4) include:

[0019] Based on the principle of maximizing user utility, the indifference point of users across different platforms is determined, and then the user scale is calculated. According to the first-order condition for maximizing platform profit, the user base is divided into pairs. and By calculating the first-order partial derivatives and solving the simultaneous equations, we finally obtain the equilibrium pricing and optimal profit for passengers and drivers on a duopoly platform corresponding to the four combinations of user preference types, under a market structure where both passengers and drivers are single-attributed: ; ; ; ; ; ; ; .

[0020] Optionally, steps (3) and (5) include:

[0021] Based on the principle of maximizing user utility, the indifference point of users across different platforms is determined, and then the user scale is calculated. According to the first-order condition for maximizing platform profit, the user base is divided into pairs. and By calculating the first-order partial derivatives and solving the simultaneous equations, we finally obtain the equilibrium pricing and optimal profit for passengers and drivers on a duopoly platform corresponding to the four user preference type combinations under a market structure with multiple passenger affiliations and single driver affiliations: ; ; ; .

[0022] Optionally, step (6) includes:

[0023] Set basic parameters , , Numerical simulation analysis was conducted to analyze the platform's optimal pricing and profit. In a single-homing structure, the platform prices quality-sensitive users higher than the perfectly rational benchmark, while pricing price-sensitive users lower. Platform profit is highest when both sides of the user base are quality-sensitive, followed by mixed types, and lowest when price-sensitive users are price-sensitive. Increased user salience improves pricing and profit, but may lead to decreased profit in asymmetric preference structures. In the case of multiple passenger affiliations and single driver affiliations, the platform consistently prices passengers lower than drivers, and passenger numbers decrease as their preference costs and network externalities increase. Platform profit is positively correlated with driver preference costs and negatively correlated with passenger preference costs and network externalities. Passengers shifting from single to multiple affiliations intensifies platform competition, leading to a general decrease in pricing for quality-sensitive passengers, while pricing for price-sensitive passengers exhibits a non-monotonic change. The platform profit change path depends on passenger demand type, exhibiting complex nonlinear characteristics.

[0024] Secondly, this invention proposes a two-sided pricing optimization system for ride-hailing platforms based on user preference differences and multi-attribution behavior, characterized in that it is applied to executing any of the pricing decision methods described above, and the system includes:

[0025] The data acquisition module is used to obtain historical behavioral data of passengers and drivers, and to construct user preference profiles by quantifying their sensitivity to service price and service quality.

[0026] The model building module is used to construct utility functions and platform profit functions that can simultaneously characterize the differentiated preferences of passengers and drivers in terms of service price and service quality, based on the salience theory in behavioral economics and combined with the Hotelling space competition model, and solve for the optimal pricing strategy.

[0027] The strategy formulation module is used to input real-time operational data of the ride-hailing platform, historical order behavior data of passengers and drivers, and their sensitivity to service price and service quality into the platform's pricing decision model. Based on the model's output, it determines whether the data to be updated needs to be updated. If so, it completes the data update in real time and generates the optimal pricing strategy combination and the optimal service strategy selection.

[0028] Beneficial effects.

[0029] The pricing optimization method for ride-hailing platforms provided by this invention effectively addresses the problems of existing ride-hailing platforms treating users as a homogeneous group during pricing strategy formulation. This approach fails to adequately consider the significant differences in price and service quality sensitivity between passengers and drivers, and also neglects the impact of user multi-affiliation behavior on the platform's competitive landscape. Consequently, pricing strategies struggle to accurately match market demand, leading to inefficient resource allocation, lost platform profits, and decreased market competitiveness. The proposed pricing optimization method significantly improves the platform's profit level in complex competitive environments. Furthermore, by guiding users with more precise price signals, it optimizes the matching efficiency between drivers and passengers, thereby enhancing the platform's overall resource allocation capabilities and market competitiveness. Attached Figure Description

[0030] Figure 1 This is a flowchart of a preferred embodiment of the ride-hailing platform bilateral pricing optimization method of the present invention.

[0031] Figure 2 The impact of user preference changes on passenger and driver pricing under four scenarios with passenger affiliation.

[0032] Figure 3 The impact of changes in user preferences on the platform's optimal profit under four scenarios with passenger single-ownership.

[0033] Figure 4 The impact of changing user preferences on passenger and driver pricing under four scenarios with multiple passenger affiliations.

[0034] Figure 5 The impact of user preference shifts on the platform's optimal profit under four scenarios with multiple passenger affiliations.

[0035] Figure 6 This study examines the impact of user preference shifts on passenger pricing under four scenarios of user multi-ownership behavior.

[0036] Figure 7 This study examines the impact of user preference shifts on driver pricing under four scenarios of user multi-ownership behavior.

[0037] Figure 8 This study examines the impact of user preference shifts on platform profits under four scenarios of user multi-ownership behavior. Detailed Implementation

[0038] The technical solution of the present invention will be described in detail below. It should be noted that the embodiments described herein are for illustrative purposes only and should not be construed as limiting the scope of protection of the present invention. All other implementations obtained by those skilled in the art based on the embodiments disclosed herein without inventive effort fall within the scope of protection of the claims of the present invention.

[0039] Unless otherwise expressly stated, the technical or scientific terms used in this invention shall have the meanings commonly understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," etc., used herein do not indicate any order, quantity, or importance, but are merely used to distinguish different components. The terms "a" or "an" do not constitute a limitation on quantity, but rather indicate the presence of at least one. Expressions such as "connected" or "linked" include not only physical or mechanical connections, but also direct or indirect electrical connections. Furthermore, directional terms such as "upper," "lower," "left," and "right" are used to describe relative positional relationships, which change accordingly when the absolute position of the described object changes.

[0040] Please see Figure 1 This application provides a two-sided pricing optimization method for ride-hailing platforms based on user preference differences and multi-attribution behavior, including:

[0041] (1) Obtain and analyze the historical behavior data of passengers and drivers, and construct a user preference profile by quantifying their sensitivity to service price and service quality.

[0042] (2) Based on the saliency theory in behavioral economics and combined with the Hotelling space competition model, a utility function that can simultaneously characterize the differentiated preferences of passengers and drivers in terms of service price and service quality is constructed, so as to more accurately simulate the user's choice behavior in real-world scenarios.

[0043] (3) Different market structure types are defined based on whether passengers adopt a single-attribution or multi-attribution strategy and whether drivers adopt a single-attribution strategy; on this basis, a duopoly platform competition model is constructed to simulate the dynamic pricing game process between platforms.

[0044] (4) For a market structure where both passengers and drivers are single-attributed, based on the principle of maximizing user utility, solve for the equilibrium number of passengers and drivers, equilibrium pricing and optimal profit of each platform under the four combinations of user preference types.

[0045] (5) For the market structure of multiple passenger affiliations and single driver affiliations, based on the principle of maximizing user utility, solve for the equilibrium number of passengers and drivers, equilibrium pricing and optimal profit of each platform under the four combinations of user preference types.

[0046] (6) Through numerical simulation, the impact of user preference differences and multi-attribution behavior on platform pricing strategy and profit level is analyzed, revealing the inherent law of platform optimal pricing and profit change with user sensitivity, and providing theoretical basis and decision support for platform to formulate differentiated pricing strategy.

[0047] In the above embodiments, by meticulously characterizing differences in user preferences and multi-affiliation behavior, the existing ride-hailing platforms effectively address the problem that, in the pricing strategy formulation process, they generally treat users as a homogeneous group, failing to fully consider the significant differences in price sensitivity and service quality sensitivity between passengers and drivers. They also neglect the impact of multi-affiliation behavior on the platform's competitive landscape, leading to pricing strategies that are difficult to accurately match market demand, resulting in inefficient resource allocation, lost platform profits, and decreased market competitiveness. The proposed pricing optimization method can significantly improve the platform's profit level in a complex competitive environment. Simultaneously, through more precise price signal guidance, it optimizes the matching efficiency between drivers and passengers, thereby enhancing the platform's overall resource allocation capabilities and market competitiveness.

[0048] The symbols and their explanations in the embodiments are shown in Table 1:

[0049] Table 1. Symbols and explanations used in this article.

[0050]

[0051] This example uses two ride-hailing platforms:

[0052] The two platforms are located in The services offered at both ends of a linear city are somewhat substitutable. Assume the user base on both sides of the platform is standardized to a size of 1, i.e. , And evenly distributed along a linear city of unit length, each represented by a letter. and This refers to users on both the passenger and driver sides. Except for users located at city endpoints, each user incurs certain travel costs when choosing the platform, including passenger travel costs. Travel costs with driver This type of cost can also be seen as a reflection of the degree of service differentiation between platforms. In practice, travel costs reflect the strength of a user's preference for a specific platform: the higher the value, the more likely the user is to choose their ideal platform, and therefore the higher the opportunity cost of switching to another platform. The fixed utility obtained by passengers and drivers from the platform is denoted as follows: and .in, This refers to the basic utility that passengers obtain from completing travel services through the platform. This refers to the basic utility that drivers receive from accepting orders and receiving corresponding compensation through the platform.

[0053] Use parameters Characterize the salience sensitivity of platform users (passengers and drivers) to service price and service quality. The smaller the value, the stronger the passenger's or driver's salient thinking, and the greater the difference in their preference for the two attributes of service price and service quality. For service price-sensitive passengers / drivers, the service price coefficient is... The service quality coefficient is .Depend on , can be obtained This indicates that, in this scenario, the marginal impact of service price on utility is dominant. Correspondingly, for service quality-sensitive passengers / drivers, the service price coefficient is... The service quality coefficient is .Depend on , can be obtained This indicates that service quality has a more significant impact on utility in this scenario. Based on historical behavioral data of passengers and drivers, users are categorized into four combinations: both sides are service price sensitive (PP), passengers are service price sensitive and drivers are service quality sensitive (PS), passengers are service quality sensitive and drivers are service price sensitive (SP), and both sides are service quality sensitive (SS).

[0054] Therefore, the utility functions for different types of users on both sides of the platform are as follows: .

[0055] When a passenger has a travel need, they submit a travel request to the platform via their mobile device. The platform then matches the passenger with a suitable vehicle and driver based on their travel information and personal preferences. To simplify the model calculations, this case assumes that the ride-hailing platform generates revenue by charging both users. This indicates the service fee charged by the platform to each passenger. This represents the commission the platform charges each driver. In reality, platforms typically take a percentage commission from drivers and charge passengers an information service fee; however, for ease of analysis, this case simplifies this by charging both passengers and drivers a fixed amount for information service and commission per order. This assumption means that for each completed order, the platform generates revenue from both users, and the platform's total profit increases with the number of completed orders—a mechanism consistent with the platform's actual operating model. Assuming the platform's marginal and fixed costs are zero, the platform's profit functions under single-ownership and multi-ownership structures are as follows:

[0056] Assuming a single service transaction occurs between the passenger and driver, the dynamic game between the platform, passenger, and driver can be divided into two phases: Phase 1, the platform... and platform Simultaneously, pricing strategies are developed for both bilateral users. and In the second phase, after observing the platform's pricing, passengers and drivers select a platform based on their own circumstances and the principle of maximizing utility.

[0057] Based on the principle of maximizing user utility, the indifference point of users across different platforms is determined, and then the user scale is calculated. According to the first-order condition for maximizing platform profit, the user base is divided into pairs. and By calculating the first-order partial derivatives and solving the simultaneous equations, we finally obtain the equilibrium pricing and optimal profit for passengers and drivers on a duopoly platform corresponding to the four combinations of user preference types, under a market structure where both passengers and drivers are single-attributed: ; ; ; ; ; ; ; .

[0058] To more intuitively demonstrate how the platform's optimal pricing and optimal profit change with user preferences under four scenarios, this case study sets the basic parameters as follows: , , Based on this, numerical simulation analysis is conducted to determine the platform's optimal pricing and optimal profit. Figure 2 It can be known that: , Based on this, we can derive the following propositions 1-2.

[0059] Proposition 1 states that in a market structure where both passengers and drivers are single-attributed, the pricing levels, user base, and profit levels of both platforms are identical in equilibrium. Specifically, regarding pricing, the platform's pricing for users on either side increases with the increase of the user's own preference costs, and decreases with the increase of the strength of the network externalities of the user on the other side. Regarding profit, the platform's profit is positively correlated with the preference costs of both sides of the user base, and negatively correlated with the strength of the cross-network externalities. In particular, the platform's user base remains constant at equilibrium and is independent of both user preference costs and the strength of the cross-network externalities.

[0060] Proposition 2: In a market structure where both passengers and drivers are single-attributed, if the market exhibits an asymmetric sensitivity structure, the platform will set prices higher than the market equilibrium level under the assumption of perfectly rational users for users with higher quality sensitivity, while pricing lower than that equilibrium level for users with higher price sensitivity. In a symmetric sensitivity market structure, the platform's prices fall between the two scenarios in the asymmetric structure; however, the platform still prices quality-sensitive users higher than price-sensitive users.

[0061] Figure 2 The study presents the specific impact of user sensitivity on the optimal profit of ride-hailing platforms under four different market scenarios, including: Based on this, we can derive the following proposition 3.

[0062] Proposition 3: In a market structure where both passengers and drivers are single-attributed, the platform's profit level exhibits the following characteristics: the profit is highest when both sides of the user base are quality-sensitive; the profit is second highest when both sides of the user base have mixed preferences; and the profit is lowest when both sides of the user base are price-sensitive, and this profit level is lower than that when both sides of the user base are perfectly rational.

[0063] according to Figure 1 and Figure 2 It can be known that: , , , , , , , , , , , Therefore, we can derive the following proposition 4.

[0064] Proposition 4: In a market structure where both passengers and drivers are single-attributed, as the salience of users on both sides increases, users' price sensitivity tends to decrease, while their sensitivity to platform preference increases, resulting in a decrease in price elasticity. When the externalities of the cross-network are symmetrical, the platform will increase the optimal price for both sides, and the platform's optimal profit will also increase; conversely, if salience weakens, both pricing and profit will decrease. In an asymmetric sensitive market, when the salience of users on one side increases, the optimal price for passengers rises, the optimal price for drivers falls, but the platform's overall optimal profit always tends to decrease.

[0065] Similarly, based on the principle of maximizing user utility, we determine the indifference point for users across different platforms, then solve for the user scale, and based on the first-order condition for maximizing platform profit, we divide the data into pairs. and By calculating the first-order partial derivatives and solving the simultaneous equations, we finally obtain the equilibrium pricing and optimal profit for passengers and drivers on a duopoly platform corresponding to the four user preference type combinations under a market structure with multiple passenger affiliations and single driver affiliations: ; ; ; .

[0066] according to Figure 3 The simulation results show that: , Based on this, we can derive the following propositions 5-6.

[0067] Proposition 5: In a market structure with multiple passenger affiliations and a single driver affiliation, the platform's pricing for passengers is consistently lower than its pricing for drivers in equilibrium. Specifically, on the passenger side, pricing increases with the strength of network externalities on that side, but decreases with the strength of network externalities on the driver side. On the driver side, pricing decreases with the strength of network externalities on either the passenger or driver side, increases with the increase of driver preference costs, but decreases with the increase of passenger preference costs. Regarding profits, platform profits increase with the increase of driver preference costs and decrease with the increase of passenger preference costs, and are negatively correlated with the strength of cross-network externalities between users on both sides. In particular, the number of drivers connected to both platforms remains constant, while the number of passengers decreases with the increase of passenger preference costs and the strength of cross-network externalities on both sides.

[0068] Proposition 6: In a market structure with multiple passenger affiliations and single driver affiliations, the platform's optimal pricing structure for both sides of the user base remains asymmetrical, meaning that quality-sensitive users are priced higher than price-sensitive users. Further analysis shows that the platform's pricing order for passengers remains consistent in both single-affiliation and multi-affiliation scenarios; however, on the driver side, the relative pricing order is reversed: in the case of single passenger affiliation, if the passenger is price-sensitive and the driver is quality-sensitive, the platform prices the driver higher; in the case of multiple passenger affiliations, when both the passenger and driver are quality-sensitive, the platform prices the driver higher.

[0069] Figure 4 The study presents the specific impact of user sensitivity on the optimal profit of ride-hailing platforms under four different market scenarios, including: Therefore, we can derive the following proposition 7.

[0070] Proposition 7: In a market structure with multiple passenger affiliations and single driver affiliations, the platform's profit level exhibits the following characteristics: When both passengers and drivers are quality-sensitive, the platform's optimal profit reaches its highest level; when both are price-sensitive, the optimal profit drops to its lowest level, falling below the benchmark level under the perfect rationality assumption; when the preferences of both parties are mixed, the platform's optimal profit lies between the above two—specifically, if passengers are price-sensitive and drivers are quality-sensitive, the platform's profit is initially lower than the perfect rationality benchmark, but will gradually surpass it as user sensitivity changes; if passengers are quality-sensitive and drivers are price-sensitive, the platform's profit will always be lower than the perfect rationality benchmark.

[0071] according to Figures 3-5 It can be known that: , , , , , , , , , , , .

[0072] Proposition 8: In a market structure with multiple passenger affiliations and a single driver affiliation, as the salience of users on both sides increases, users' price sensitivity tends to decrease, while their sensitivity to platform preference increases, manifesting as a decrease in user price elasticity. When the externalities of the cross-network are symmetrical, the platform will increase the optimal price for both sides of the user base, and the platform's optimal profit will also increase; conversely, if salience weakens, both pricing and profit will decrease. In an asymmetric sensitive market, when the salience of users on one side increases, the optimal price for passengers increases, the optimal price for drivers decreases, but the platform's overall optimal profit always shows an upward trend.

[0073] according to Figures 6-7 It can be known that: , Based on this, we can derive the following propositions 9-10.

[0074] Proposition 9: When passengers shift from single-affiliation to partially multi-affiliation, the platform's pricing for quality-sensitive passengers tends to decrease; however, the pricing for price-sensitive passengers exhibits a non-monotonic change, meaning that as their price sensitivity decreases, the price first increases and then decreases. Furthermore, the platform's price reduction for quality-sensitive passengers is greater than that for price-sensitive passengers. The platform's pricing adjustments for passengers differ significantly under different market structures: in a market structure where passengers are quality-sensitive and drivers are price-sensitive, the platform's price reduction for passengers is the largest; conversely, in a market structure where passengers are price-sensitive and drivers are quality-sensitive, the reduction is the smallest; when the passenger and driver preference types are symmetrical, the platform's price reduction for passengers falls between these two extreme cases.

[0075] Proposition 10: When passengers shift from single-affiliation to partially multi-affiliation, the platform's pricing strategy for drivers depends on the passenger's demand type. Specifically, if passengers are price-sensitive, the platform's pricing for drivers exhibits a non-monotonic trend of first decreasing and then increasing as their price sensitivity decreases, with this non-monotonicity being most pronounced when drivers are quality-sensitive; if passengers are quality-sensitive, the platform's pricing for drivers shows an overall upward trend, with the largest increase in pricing when drivers are price-sensitive.

[0076] according to Figure 8 The simulation results lead to Proposition 11.

[0077] Proposition 11: When passengers shift from single-affiliation to partially multi-affiliation, the direction and magnitude of platform profit changes primarily depend on the passenger's demand type. Specifically, if passengers are price-sensitive, platform profits exhibit a non-linear dynamic path of first decreasing, then increasing, and then decreasing again as their price sensitivity weakens; if passengers are quality-sensitive, platform profits generally tend to decrease. Further analysis shows that the magnitude of the profit decline varies significantly under different market structures. When passengers are price-sensitive and drivers are quality-sensitive, the decline in platform profits gradually converges from a maximum to a minimum as passengers' price sensitivity decreases; conversely, if passengers are quality-sensitive and drivers are price-sensitive, the profit decline reaches its maximum; when passenger and driver preference types have a symmetrical structure, the profit decline lies between these two extreme cases.

[0078] Based on the same inventive concept, this embodiment also proposes a two-sided pricing optimization system for ride-hailing platforms based on user preference differences and multi-attribution behavior, applied to the above-mentioned two-sided pricing optimization method for ride-hailing platforms. The system includes:

[0079] The data acquisition module is used to obtain historical behavioral data of passengers and drivers, and to construct user preference profiles by quantifying their sensitivity to service price and service quality.

[0080] The model building module is used to construct utility functions and platform profit functions that can simultaneously characterize the differentiated preferences of passengers and drivers in terms of service price and service quality, based on the salience theory in behavioral economics and combined with the Hotelling space competition model, and solve for the optimal pricing strategy.

[0081] The strategy formulation module is used to input real-time operational data of the ride-hailing platform, historical order behavior data of passengers and drivers, and their sensitivity to service price and service quality into the platform's pricing decision model. Based on the model's output, it determines whether the data to be updated needs to be updated. If so, it completes the data update in real time and generates the optimal pricing strategy combination and the optimal service strategy selection.

[0082] The aforementioned ride-hailing platform bilateral pricing optimization system can implement all the above-mentioned embodiments of ride-hailing platform bilateral pricing optimization and achieve the same beneficial effects, which will not be elaborated here.

[0083] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A two-sided pricing optimization method for ride-hailing platforms based on user preference differences and multi-attribution behavior, characterized in that, The method includes: (1) Obtain and analyze the historical behavior data of passengers and drivers, and construct a user preference profile by quantifying their sensitivity to service price and service quality. (2) Based on the saliency theory in behavioral economics and combined with the Hotelling space competition model, a utility function that can simultaneously characterize the differentiated preferences of passengers and drivers in terms of service price and service quality is constructed, so as to more accurately simulate the user's choice behavior in real-world scenarios. (3) Different market structure types are defined based on whether passengers adopt a single-attribution or multi-attribution strategy and whether drivers adopt a single-attribution strategy; on this basis, a duopoly platform competition model is constructed to simulate the dynamic pricing game process between platforms. (4) For a market structure where both passengers and drivers are single-attributed, based on the principle of maximizing user utility, solve for the equilibrium number of passengers and drivers, equilibrium pricing and optimal profit of each platform under four combinations of user preference types. (5) For the market structure of multiple passenger affiliations and single driver affiliations, based on the principle of maximizing user utility, solve for the equilibrium number of passengers and drivers, equilibrium pricing and optimal profit of each platform under the four combinations of user preference types. (6) Through numerical simulation, the impact of user preference differences and multi-attribution behavior on platform pricing strategy and profit level is analyzed, revealing the inherent law of platform optimal pricing and profit change with user sensitivity, and providing theoretical basis and decision support for platform to formulate differentiated pricing strategy.

2. The method according to claim 1, characterized in that, Step (1) includes: Use parameters Characterize the salience sensitivity of platform users (passengers and drivers) to service price and service quality. The smaller the value, the stronger the passenger's or driver's salient thinking, and the greater the difference in their preference for the two attributes of service price and service quality. For service price-sensitive passengers / drivers, the service price coefficient is... The service quality coefficient is .Depend on , can be obtained This indicates that, in this scenario, the marginal impact of service price on utility is dominant. Correspondingly, for service quality-sensitive passengers / drivers, the service price coefficient is... The service quality coefficient is .Depend on , can be obtained This indicates that service quality has a more significant impact on utility in this scenario. Based on historical behavioral data of passengers and drivers, users are categorized into four combinations: both sides are service price sensitive (PP), passengers are service price sensitive and drivers are service quality sensitive (PS), passengers are service quality sensitive and drivers are service price sensitive (SP), and both sides are service quality sensitive (SS).

3. The method according to claim 1, characterized in that, The utility functions for the passenger and driver in step (2) are as follows: ； In the formula, and These respectively represent passengers and drivers on the platform. The effect, and These represent the basic utility of passengers and drivers joining a ride-hailing platform, respectively. This indicates that passengers have joined the platform. The service price charged This indicates that the driver has joined the platform. Commissions extracted, and These respectively represent passengers and drivers on the platform. Quantity, and Represent the cross-network externality strengths of passengers and drivers, respectively. and These represent the travel costs for passengers and drivers, respectively.

4. The method according to claim 1, characterized in that, The platform profit function in steps (4) and (5) is: ； In the formula, Indicates platform Profits Indicates the platform to which the passenger belongs. The quantity.

5. The method according to claim 1, characterized in that, Steps (3) and (4) include: Based on the principle of maximizing user utility, the indifference point of users across different platforms is determined, and then the user scale is calculated. According to the first-order condition for maximizing platform profit, the user base is divided into pairs. and By calculating the first-order partial derivatives and solving the simultaneous equations, we finally obtain the equilibrium pricing and optimal profit for passengers and drivers on a duopoly platform corresponding to the four combinations of user preference types, under a market structure where both passengers and drivers are single-attributed: ； ； ； ； ； ； ； 。 6. The method according to claim 1, characterized in that, Steps (3) and (5) include: Based on the principle of maximizing user utility, the indifference point of users across different platforms is determined, and then the user scale is calculated. According to the first-order condition for maximizing platform profit, the user base is divided into pairs. and By calculating the first-order partial derivatives and solving the simultaneous equations, we finally obtain the equilibrium pricing and optimal profit for passengers and drivers on a duopoly platform corresponding to the four user preference type combinations under a market structure with multiple passenger affiliations and single driver affiliations: ； ； ； 。 7. The method according to claim 1, characterized in that, In step (6), basic parameters are set. , , Numerical simulation analysis was conducted to analyze the platform's optimal pricing and profit. In a single-homing structure, the platform prices quality-sensitive users higher than the perfectly rational benchmark, while pricing price-sensitive users lower. Platform profit is highest when both sides of the user base are quality-sensitive, followed by mixed types, and lowest when price-sensitive users are price-sensitive. Increased user salience improves pricing and profit, but may lead to decreased profit in asymmetric preference structures. In the case of multiple passenger affiliations and single driver affiliations, the platform consistently prices passengers lower than drivers, and passenger numbers decrease as their preference costs and network externalities increase. Platform profit is positively correlated with driver preference costs and negatively correlated with passenger preference costs and network externalities. Passengers shifting from single to multiple affiliations intensifies platform competition, leading to a general decrease in pricing for quality-sensitive passengers, while pricing for price-sensitive passengers exhibits a non-monotonic change. The platform profit change path depends on passenger demand type, exhibiting complex nonlinear characteristics.

8. A two-sided pricing optimization system for ride-hailing platforms based on user preference differences and multi-attribution behavior, characterized in that, The system comprising the steps of any one of claims 1-7, wherein the system implements the method steps of claims 1-7. The data acquisition module is used to obtain historical behavioral data of passengers and drivers, and to construct user preference profiles by quantifying their sensitivity to service price and service quality. The model building module is used to construct utility functions and platform profit functions that can simultaneously characterize the differentiated preferences of passengers and drivers in terms of service price and service quality, based on the salience theory in behavioral economics and combined with the Hotelling space competition model, and solve for the optimal pricing strategy. The strategy formulation module is used to input real-time acquired ride-hailing platform operation data, passenger and driver historical order behavior data, and the sensitivity of both in terms of service price and service quality into the platform pricing decision model, and determine whether the data to be updated needs to be updated based on the model output. If necessary, data updates will be completed in real time, and the optimal pricing strategy combination and optimal service strategy selection will be generated.

Images