A method for determining the minimum fare and maximum transport capacity of a network car to ensure the driver's income

Abstract

This invention discloses a method for determining the minimum fare and maximum capacity of ride-hailing services to guarantee driver income. Based on historical operating data and operating cost data of the target area, combined with the local minimum hourly wage, the method accurately calculates the minimum average monthly revenue and the corresponding minimum average price per ride to guarantee drivers' basic income. This leads to the construction of a structured minimum fare standard that includes a starting price, mileage fee, time fee, and long-distance fee. Using this minimum average price per ride as a fare constraint, a system dynamics model is used to simulate the changes in drivers' average monthly revenue under different numbers of drivers. Furthermore, by ensuring that drivers' average monthly revenue is not lower than the minimum average monthly revenue, the maximum capacity that the market can accommodate is determined. The advantage is that, based on a clearly defined minimum fare standard, the maximum capacity is managed collaboratively, ensuring a balance between supply and demand in the ride-hailing industry, preventing disorderly capacity growth that could dilute drivers' average monthly revenue, and thus guaranteeing ride-hailing drivers' income.

Description

Technical Field

[0001] This invention relates to a method for determining minimum fares and maximum capacity, and more particularly to a method for determining minimum fares and maximum capacity for ride-hailing services to guarantee drivers' income. Background Technology

[0002] With the rapid expansion of the ride-hailing market, the industry has gradually exposed prominent problems such as low fares, insufficient driver income, and an imbalance between supply and demand for transportation capacity. Policy has also clearly required that ride-hailing drivers be included in the minimum wage guarantee program. This necessitates balancing driver income rights, passenger convenience, and platform sustainability in the design of fare and capacity scale.

[0003] However, existing ride-hailing pricing mechanisms primarily focus on platform profits and passenger travel costs, failing to adequately protect the value of drivers' labor. Core pricing items such as starting fares, mileage fees, time-based fees, and long-distance fees lack clear minimum standards. Therefore, it is necessary to establish minimum fare standards for ride-hailing services to increase driver income. However, it is important to note that simply establishing and enforcing minimum fare standards without targeted control over capacity could easily lead to disorderly capacity growth, thereby diluting drivers' average monthly revenue.

[0004] Therefore, it is of great significance to coordinate and manage the maximum transport capacity while clearly defining the minimum freight rate standard, in order to solve the current prominent problems of supply and demand imbalance and insufficient driver income in the industry. Summary of the Invention

[0005] The technical problem this invention aims to solve is to provide a method for determining the minimum fare and maximum capacity of ride-hailing services to guarantee drivers' income. This method, based on a clearly defined minimum fare standard, collaboratively manages the maximum capacity, ensuring a balance between supply and demand in the ride-hailing industry, preventing disorderly capacity growth that could dilute drivers' average monthly revenue, and thus guaranteeing ride-hailing drivers' income.

[0006] The technical solution adopted by this invention to solve the above-mentioned technical problems is as follows: a method for determining the minimum fare and maximum capacity of ride-hailing services to guarantee drivers' income, comprising the following steps: S1. Establish a system dynamics model for the ride-hailing market, wherein the system dynamics model is at least related to the average monthly passenger demand, the number of drivers, and the average monthly driver revenue; S2. Obtain historical operating data and operating cost data of the ride-hailing industry in the target area, as well as industry benchmark data; S3. Based on the historical operating data and operating cost data, combined with the local minimum hourly wage, calculate the minimum monthly revenue and the corresponding minimum average order price to guarantee the driver's basic income; S4. Based on the lowest average order price, construct a minimum freight rate standard that includes a base price, mileage fee, time fee, and long-distance fee; S5. Using the minimum average price per ride as a fare constraint, the system dynamics model is used to simulate the changes in the average monthly revenue of drivers under different numbers of drivers, and the maximum capacity of the ride-hailing market in the target area is determined with the constraint that the average monthly revenue of drivers is not lower than the minimum average monthly revenue.

[0007] Compared with existing technologies, the advantages of this invention lie in its ability to accurately calculate the minimum monthly revenue and corresponding minimum average price per ride to guarantee drivers' basic income based on historical operational data and operational cost data of the target area, combined with the local minimum hourly wage. This allows for the construction of a structured minimum fare standard that includes a starting price, mileage fee, time fee, and long-distance fee, ensuring from the source that drivers' income can cover costs and reach the policy guarantee line. Furthermore, using this minimum average price per ride as a fare constraint, a system dynamics model is used to simulate the changes in drivers' average monthly revenue under different numbers of drivers. By ensuring that drivers' average monthly revenue is not lower than the minimum average monthly revenue, the maximum capacity that the market can accommodate is determined in reverse. This achieves the coordinated determination and dynamic balance between guaranteed fares and scientific capacity management, effectively avoiding the problem of driver income dilution or travel service shortages caused by a single policy. The parameters required by the method of this invention can all be obtained from publicly available or industry regulatory data. The calculation process is clear, and the results can be verified based on the input data, possessing both scientific validity and practical value for direct application in industry supervision and platform operation.

[0008] Furthermore, the historical operational data includes average monthly working days per vehicle, average daily operating hours per vehicle, average daily order volume per vehicle, average passenger-carrying mileage per order, average passenger-carrying time per order, average mileage for long-distance orders, average empty-running time for long-distance orders, and average empty-running time for regular orders. The operational cost data includes the contract fees for pure electric ride-hailing vehicles, the contract fees for hybrid ride-hailing vehicles, fuel costs for pure electric ride-hailing vehicles, fuel costs for hybrid ride-hailing vehicles, market share of pure electric ride-hailing vehicles, market share of hybrid ride-hailing vehicles, average order completion bonuses and average commission rates of ride-hailing platforms. The industry reference data includes minimum hourly wages, starting fares for taxis, starting mileage for taxis, starting time for taxis, and mileage fees for taxis. Furthermore, the minimum average monthly revenue is calculated based on the minimum hourly wage, the average monthly working days per vehicle, and the average daily operating hours per vehicle from the historical operational data, and further adjusted by combining the vehicle contract fees, fuel costs, platform bonuses, and commission rates from the operational cost data.

[0009] Furthermore, the minimum average price per order is determined by dividing the minimum average monthly revenue by the average monthly order volume per vehicle obtained from the historical operating data.

[0010] Furthermore, the mileage fee and time fee in the minimum fare standard satisfy the following constraint: the starting price of the taxi + mileage fee × K1 + time fee × K2 ≥ the minimum average price per order; where K1 is the difference between the average passenger mileage of the order obtained from historical operating data and the starting mileage of the taxi, and K2 is the difference between the average passenger time of the order obtained from historical operating data and the starting time of the taxi.

[0011] Furthermore, the determination of the long-haul fee for the minimum freight rate standard is related to the unit-time revenue converted from the minimum average monthly revenue, and is used to compensate for the additional empty-running costs of long-haul orders compared to regular orders.

[0012] Furthermore, the system dynamics model includes an evolution equation for average monthly passenger demand and an evolution equation for average monthly driver revenue. The evolution equation for average monthly passenger demand uses average monthly passenger demand (Q) as a state variable, and its evolution is affected by the average price of ride-hailing orders (P), the number of drivers (N), the growth rate of the resident population (β), and the external disturbance coefficient (K) of public safety events. The evolution equation for average monthly driver revenue is used to calculate the average monthly driver revenue (R), and its input variables include the average monthly passenger demand (Q), the average price of ride-hailing orders (P), the number of drivers (N), and the full-time driver coefficient (f).

[0013] Furthermore, the specific method for determining the maximum capacity of the ride-hailing market in the target area is as follows: using the minimum average price per ride as a fare constraint, the system dynamics model is used to simulate the average monthly revenue of drivers under different driver number scenarios; from all simulated scenarios where the average monthly revenue of drivers is not lower than the minimum average monthly revenue, the corresponding maximum number of drivers is selected as the maximum capacity. Attached Figure Description

[0014] Figure 1 A flowchart illustrating the method for determining the minimum fare and maximum capacity of ride-hailing services to guarantee driver income, as described in this invention. Figure 2 This is the core causal relationship diagram of the system dynamics model constructed in this invention; Figure 3 This is a historical verification fitting plot of the actual and simulated monthly average demand for ride-hailing services in Ningbo City from 2020 to 2024. Figure 4 This is a historical test fitting plot of the actual and simulated values ​​of the average monthly revenue of ride-hailing drivers in Ningbo City from 2020 to 2024. Detailed Implementation

[0015] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0016] Example 1: As Figure 1As shown, a method for determining the minimum fare and maximum capacity for ride-hailing services to guarantee drivers' income is characterized by the following steps: S1. The target area is set as Ningbo City. A system dynamics model of the ride-hailing market in Ningbo City is established. The system dynamics model should at least relate to the average monthly passenger demand, the number of drivers, and the average monthly driver revenue. The system dynamics model is used to determine the maximum capacity of the ride-hailing market to avoid the risk of secondary revenue dilution caused by the disorderly growth of the number of drivers after the implementation of the minimum fare standard, thus ensuring driver income and maintaining market supply and demand balance. The core causal relationships of the system dynamics model are as follows: Figure 2 As shown, Figure 2 In Chinese, "+" indicates that the variable changes in the same direction, and "-" indicates that the variable changes in the opposite direction. S2. Obtain historical operating data and operating cost data for the ride-hailing industry in Ningbo, as well as industry reference data; collect at least 6 months of historical operating data and operating cost data from Ningbo, as well as the most recent year of industry reference data; the historical operating data and operating cost data are all from the operating data of various ride-hailing platforms in the target area, obtained from the management departments, regulatory information systems, or legally authorized information channels responsible for transportation management, ride-hailing industry supervision, ride-hailing operation monitoring, or ride-hailing industry statistics in the target area. The industry reference data comes from the latest government documents and taxi-related reports in the target area. Among them, the historical operating data does not distinguish between ride-hailing platforms, and the operating cost data is the average of the operating cost data of each ride-hailing platform. The industry reference data has regional uniformity and stability. Only one set of standard data corresponds to one city administrative region and is directly used as the benchmark reference. Historical operational data: Average monthly working days per vehicle: 26 days; Average daily operating time per vehicle: 13.8 hours; Average daily order volume per vehicle: 25; Average passenger mileage per order: 7.2 kilometers; Average passenger mileage per order: 16.1 minutes; Average mileage per long-distance order: 24.12 kilometers; Average empty time per long-distance order: 18.48 minutes; Average empty time per regular order: 14.59 minutes. Operating cost data: Monthly contract fee for pure electric ride-hailing vehicles: 3200 yuan; Monthly contract fee for hybrid ride-hailing vehicles: 2900 yuan; Monthly fuel cost for pure electric ride-hailing vehicles: 1172 yuan; Monthly fuel cost for hybrid ride-hailing vehicles: 2598 yuan; Market share of pure electric ride-hailing vehicles: 92%; Market share of hybrid ride-hailing vehicles: 8%; Average order completion bonus for ride-hailing platforms: 1000 yuan; Average commission rate for ride-hailing platforms: 20%. Industry benchmark data: minimum hourly wage 22 yuan / hour, starting price for a taxi 11 yuan, starting distance for a taxi 3 kilometers, starting time for a taxi 3 minutes, and mileage fee for a taxi 2.4 yuan / kilometer; S3. Based on the minimum hourly wage and historical operating data, calculate the minimum average monthly net profit; and based on the minimum average monthly net profit and the average monthly total operating cost, calculate the minimum average monthly revenue. S4. Based on the minimum average monthly revenue, calculate the minimum average price per order, and based on the minimum average price per order, construct a minimum freight rate standard including the starting price, mileage fee, time fee, and long-distance fee; S5. Using the minimum average price per ride determined in step S4 as a price constraint, a system dynamics model is used to simulate the monthly revenue changes under different driver number scenarios. With the constraint that the monthly average revenue is not lower than the minimum monthly average revenue determined in step S3, the maximum capacity of the ride-hailing market is obtained.

[0017] In this embodiment, step S3 calculates the minimum average monthly net profit based on the minimum hourly wage and historical operating data; the specific process for calculating the minimum average monthly revenue based on the minimum average monthly net profit and the average monthly total operating cost is as follows: Step 3.1: The minimum average monthly net profit is calculated to be 7894 yuan using formula (1):

[0018] The minimum hourly wage is 22 yuan / hour; the average number of working days per month is 26 days; and the average daily operating time per day is 13.8 hours. Step 3.2: The average monthly total operating cost is calculated using formula (2), which is 3462 yuan.

[0019] Among them, the average order completion bonus of the ride-hailing platform is RMB 1,000 / month; the comprehensive contract fee is calculated using formula (3) as RMB 3,176 / month:

[0020] Among them, A represents the market share of pure electric ride-hailing vehicles at 92%; the contract fee for pure electric ride-hailing vehicles is 3200 yuan / month; B represents the market share of hybrid ride-hailing vehicles at 8%; the contract fee for hybrid ride-hailing vehicles is 2900 yuan / month. The comprehensive fuel cost is calculated similarly using formula (4), and is 1286 yuan / month:

[0021] Among them, A represents the market share of pure electric ride-hailing vehicles at 92%; the fuel cost for pure electric ride-hailing vehicles is 1172 yuan / month; B represents the market share of hybrid ride-hailing vehicles at 8%; the fuel cost for hybrid ride-hailing vehicles is 2598 yuan / month. Step 3.3: The minimum average monthly revenue is calculated using formula (5), which is 14,195 yuan.

[0022] The lowest average monthly net profit is 7,894 yuan; the average monthly total operating cost is 3,462 yuan; and the average commission rate of ride-hailing platforms is 20%.

[0023] In this embodiment, the specific process of calculating the minimum average price per order based on the minimum average monthly revenue in step S4, and constructing the minimum fare standard including the starting price, mileage fee, time fee, and long-distance fee based on the minimum average price per order, is as follows: Step 4.1: The average monthly order volume per vehicle is calculated using formula (6), which is 650 orders.

[0024] The average number of working days per month for a single vehicle is 26; the average number of orders received per day for a single vehicle is 25. The lowest average price per unit, calculated using formula (7), is 21.8 yuan.

[0025] Step 4.2: Set the minimum starting price for ride-hailing services to 11 yuan, based on the starting price of 11 yuan for taxis, which includes a starting distance of 3 kilometers and a starting time of 3 minutes. Step 4.3: Construct the interrelationship between mileage fees and time fees, as shown in formula (8):

[0026] The minimum starting price is 11 yuan; a is the mileage fee, in yuan / km; b is the time fee, in yuan / minute; the average passenger mileage per order is 7.2 km, and the average passenger time per order is 16.1 minutes; the starting mileage is 3 km, and the starting time is 3 minutes. Based on the interrelationship between mileage fees and time fees, five minimum combination standards are designed, each of which includes a minimum mileage fee and a minimum time fee. The first i The minimum mileage fee is denoted as a. i , No. i The minimum duration fee is recorded as b i For i = 1, 2, 3, 4, 5, the minimum mileage fee a for the i-th type is calculated using formula (9). i :

[0027] Where a1 is 70% of the taxi mileage fee, and the taxi mileage fee is 2.4 yuan / km, that is, a1 is 1.68 yuan / km; g is the interval between the lowest values ​​of the two adjacent mileage fees, in yuan / km, and is taken as 0.1; a i The specific divisions are shown in the second column of Table 1; The minimum duration fee for the i-th type is calculated using formula (10). b i :

[0028] The average passenger distance per order was 7.2 kilometers, and the average passenger time per order was 16.1 minutes. The lowest value of the category combination is taken as the mileage fee input, and the corresponding minimum duration fee is calculated by substituting it into formula (10). The calculation results are shown in the third column of Table 1: Table 1. Minimum Duration Fees for Various Mileage Fees

[0029] As long as the mileage fee and time fee of the ride-hailing vehicle meet any of the minimum combination standards above, it is judged to meet the minimum standard requirements. Step 4.4: The minimum hourly revenue is calculated to be 39.56 yuan using formula (11), which means converting the minimum monthly average revenue into an hourly revenue benchmark:

[0030] The average number of working days per month for each bicycle is 26, and the average daily operating time per bicycle is 13.8 hours. The minimum long-distance cost is calculated using formula (12) and is 0.28 yuan / km:

[0031] The extra empty time is the difference between the average empty time of long-distance orders and regular orders, which is 3.89 minutes; the constant 60 is the conversion factor from minutes to hours to keep the unit of extra empty time consistent with the minimum hourly revenue; the average mileage of more than 15 kilometers is the difference between the average mileage of long-distance orders and 15 kilometers, which is 9.12 kilometers.

[0032] In this embodiment, the system dynamics model includes the evolution equation of monthly passenger demand and the evolution equation of monthly driver revenue. The evolution equation of monthly passenger demand is used to reflect the updating process of monthly passenger demand under the combined effects of public safety events, population changes, changes in average price per unit, and changes in supply and demand structure, as shown in formula (13):

[0033] Where t represents the current year; t-1 represents the previous year; K t K represents the external interference coefficient of public safety incidents in year t; t-1The external interference coefficient for public safety incidents in year t-1 is 1 when there is no external shock (such as no epidemic or major public event). When there is an external shock, the external shock intensity is divided into three levels: severe shock, moderate shock, and mild shock or recovery stage, according to the shock intensity value, and each level corresponds to a different value range. In the specific implementation process, when there is a public safety incident, the year in which there is no public safety incident in the target area and the operation is relatively stable can be selected as the base year. The relative demand intensity index DI based on the average monthly passenger demand and the number of drivers is constructed to characterize the degree of change of passenger travel demand relative to the base year under the conditions of unit capacity and unit price, as shown in formula (14):

[0034] in, DI t Let K be the relative demand intensity index for year t; Q0 be the average monthly passenger demand in the baseline year; and N0 be the number of drivers in the baseline year. When the relative demand intensity index is less than 0.80, it is considered a year significantly below the baseline level, corresponding to a severe shock phase, with K ranging from 0.50 to 0.60. When the relative demand intensity index is between 0.80 and 0.90, it corresponds to a moderate shock phase, with K ranging from 0.60 to 0.85. When the relative demand intensity index is greater than 0.90, it corresponds to a mild shock or recovery phase, with K ranging from 0.85 to 1.00. β is the resident population growth rate, taken as the average of the resident population growth rates of the target area over the past five years, which is 4.34%. w E is the waiting time elasticity coefficient. p Here, E represents the price elasticity coefficient and the waiting time elasticity coefficient. w With price elasticity coefficient E p The elasticity parameter to be calibrated in the system dynamics model is used to characterize the sensitivity of passenger demand to changes in waiting time and the average price of ride-hailing orders. The waiting time elasticity coefficient E w With price elasticity coefficient E p The waiting time elasticity coefficient E is determined in this embodiment by historical fitting and parameter back-identification of historical operational data of the target area. w The price elasticity coefficient is -0.52, E p -1.30; Q t Q represents the average monthly passenger demand in year t; t-1 N represents the average monthly passenger demand in year t-1. t N is the number of drivers in year t; t-1 P represents the number of drivers in year t-1; t P represents the average price of ride-hailing orders in year t;t-1 This represents the average price of ride-hailing orders in year t-1. The monthly revenue evolution equation is used to reflect the relationship between the driver's monthly revenue and the average price of ride-hailing orders, the average monthly demand of passengers, the number of drivers, and the proportion of full-time drivers, as shown in formula (15):

[0035] Among them, R t t represents the average monthly revenue of drivers in year t; f represents the full-time driver coefficient, which is 0.55, based on the latest statistics on the proportion of full-time drivers in the target area.

[0036] In this embodiment, the specific process of obtaining the maximum capacity of the ride-hailing market in step S5 is as follows: Step 5.1: Construct two different control rules: The first control rule is to not control the number of drivers, and the number of drivers evolves according to the natural growth rate, with the natural growth rate γ ranging from 10% to 15%; The second control rule is to control the number of drivers, and the number of drivers does not grow naturally over time. Step 5.2: Set two control scenarios for the first control rule: The first control scenario is no fare adjustment, and the average price of ride-hailing orders in the following year is equal to the average price of ride-hailing orders in the previous year; The second control scenario is fare adjustment, and the average price of ride-hailing orders is fixed at the lowest average price per order obtained so far. Set two control scenarios for the second control rule: The first control scenario is that the number of drivers in the following year is fixed at the number of drivers in the previous year, and the average price per order in the following year is fixed at the lowest average price per order obtained so far; The second control scenario is that the average price per order in the following year is fixed at the lowest average price per order obtained so far, with the constraint that "the average monthly revenue is not lower than the currently determined minimum average monthly revenue", and the upper limit of the number of drivers is derived backward through the system dynamics model. Step 5.3: For the four regulation scenarios, based on the system dynamics model, perform evolution simulations for the three consecutive years following the current year (i.e., year t), namely, the evolution simulations for year t+1, year t+2, and year t+3, where t equals 2024, i.e., the 2024th year; (1) Simulation basic parameter settings: Select the average monthly passenger demand Q in Ningbo City in 2024 t The number of orders was 11.109 million, and the number of drivers was N. t For 35,600 people, the average price of ride-hailing orders is P t The average monthly revenue per driver is 20.6 yuan. t The simulation baseline data is set at 11,384 yuan. Minimum constraint parameter: The minimum average monthly revenue R for ride-hailing drivers in Ningbo City, calculated in step S3. min The lowest average unit price P, calculated in step S4, is 14195 yuan.min It costs 21.8 yuan; Parameter data: Resident population growth rate β is 4.34%, waiting time elasticity coefficient E w The price elasticity coefficient is -0.52, E p The coefficients are -1.30, the full-time driver coefficient f is 0.55, the driver natural growth rate γ is 10%, and the external interference coefficient K for public safety incidents is set to 1 for scenarios without external impact. (2) Evolutionary simulation: Taking the evolution simulation in year t+1 as an example, evolution simulations are carried out for four different scenarios: Scenario 1: No control over the number of drivers + no adjustment of fares ① At this point, the number of drivers N in year t+1 is... t+1(1) =N t ×(1+γ), average price of ride-hailing orders in year t+1 P t+1(1) =P t ; Given data for year t: Average monthly passenger demand Q t External interference coefficient K for public safety incidents t Number of drivers N t Average price of ride-hailing orders P t ; Given data from year t+1: External disturbance coefficient K for public safety incidents t+1(1) =1, resident population growth rate β, waiting time elasticity coefficient E w Price elasticity coefficient E p Number of drivers N t+1(1) Average price of ride-hailing orders P t+1(1) ; ② Substitute into formula (13) to calculate the simulated value Q of the average monthly passenger demand in year t+1. t+1(1) : Substituting the known data above into formula (13), we obtain the simulated value Q of the average monthly passenger demand in year t+1. t+1(1) ; ③ Substitute into formula (15) to calculate the simulated value R of the average monthly revenue of drivers in year t+1. t+1(1) : The simulated value Q of the average monthly passenger demand in year t+1 t+1(1) Average price of ride-hailing orders P t+1(1) Number of drivers N t+1(1) Substituting the full-time driver coefficient f into formula (15), the simulated value R of the driver's average monthly revenue in year t+1 is calculated. t+1(1) ; Scenario 2: No control over the number of drivers + regulated fares ① At this point, the number of drivers N in year t+1 is... t+1(2) =N t×(1+γ), average price of ride-hailing orders in year t+1 P t+1(2) =P min ; Given data for year t: Average monthly passenger demand Q t External interference coefficient K for public safety incidents t Number of drivers N t Average price of ride-hailing orders P t ; Given data from year t+1: External disturbance coefficient K for public safety incidents t+1(2) =1, resident population growth rate β, waiting time elasticity coefficient E w Price elasticity coefficient E p Number of drivers N t+1(2) Average price of ride-hailing orders P t+1(2) ; ② Substitute into formula (13) to calculate the simulated value Q of the average monthly passenger demand in year t+1. t+1 : Substituting the known data above into formula (13), we obtain the simulated value Q of the average monthly passenger demand in year t+1. t+1 ; ③ Substitute into formula (15) to calculate the simulated value R of the average monthly revenue of drivers in year t+1. t+1(2) : The simulated value Q of the average monthly passenger demand in year t+1 t+1(2) Average price of ride-hailing orders P t+1(2) Number of drivers N t+1(2) Substituting the full-time driver coefficient f into formula (15), the simulated value R of the driver's average monthly revenue in year t+1 is calculated. t+1(2) ; Scenario 3: Controlling the number of drivers + adjusting fares ① At this point, the number of drivers N in year t+1 is... t+1(3) =N t The average price of ride-hailing orders in t+1 years, P t+1(3) =P min ; Given data for year t: Average monthly passenger demand Q t External interference coefficient K for public safety incidents t Number of drivers N t Average price of ride-hailing orders P t ; Given data from year t+1: External disturbance coefficient K for public safety incidents t+1(3) =1, resident population growth rate β, waiting time elasticity coefficient E w Price elasticity coefficient E p Number of drivers N t+1(3) Average price of ride-hailing orders P t+1(3) ; ② Substitute into formula (13) to calculate the simulated value Q of the average monthly passenger demand in year t+1. t+1(3) : Substituting the known data above into formula (13), we obtain the simulated value Q of the average monthly passenger demand in year t+1. t+1(3) ; ③ Substitute into formula (15) to calculate the simulated value R of the average monthly revenue of drivers in year t+1. t+1(3) : The simulated value Q of the average monthly passenger demand in year t+1 t+1(3) Average price of ride-hailing orders P t+1(3) Number of drivers N t+1(3) Substituting the full-time driver coefficient f into formula (15), the simulated value R of the driver's average monthly revenue in year t+1 is calculated. t+1(3) ; Scenario 4: Dynamically controlling the number of drivers + adjusting fares ① At this point, the average price P of ride-hailing orders in year t+1 t+1(4) =P min With the constraint that "average monthly revenue is not lower than the currently determined minimum average monthly revenue", the simulated value R of the driver's average monthly revenue in year t+1 is... t+1(4) ≥R min ; Given data for year t: Average monthly passenger demand Q t External interference coefficient K for public safety incidents t Number of drivers N t Average price of ride-hailing orders P t ; Given data from year t+1: External disturbance coefficient K for public safety incidents t+1(4) =1, resident population growth rate β, waiting time elasticity coefficient E w Price elasticity coefficient E p Average price of ride-hailing orders P t+1(4) ; ②Derive the upper limit of the number of drivers by combining formulas (13) and (15): According to constraint R t+1(4) ≥R min Formula (16) is obtained by transforming formula (15):

[0037] ③ Substitute into formula (13) to calculate the simulated value Q of the average monthly passenger demand in year t+1. t+1(4) : Let N t+1(4) =x (number of drivers to be determined), substitute the known data and the unknown x into formula (13) to obtain Q. t+1(4) The expression relating to x:

[0038] ④ Substitute into inequality (16) to solve for the upper limit of the number of drivers: Q t+1(4) Substituting the relationship expression (17) with x into formula (16), we obtain the result that satisfies the constraint condition R. t+1(4) ≥R min The maximum value of x, i.e., the upper limit N of the number of drivers in year t+1. t+1(4) ; Similarly, simulation data for year t+2 and year t+3 can be obtained. A total of 12 sets of simulation data are generated for the four control scenarios (4 control scenarios × 3 years). The simulation data includes the simulated values ​​of average monthly passenger demand, number of drivers, and average monthly driver revenue. Evolutionary simulations were conducted for the four scenarios, and the specific data are shown in Table 2: Table 2 Simulation Results of Regulation Scenario

[0039] Step 5.4: Filter the 12 sets of simulation data according to the constraint that "average monthly revenue ≥ minimum average monthly revenue". In Scenario 4, when the number of drivers is controlled at 30,400, the average monthly revenue per driver will continue to grow from 2025 to 2027, reaching 14,217 yuan in 2027, thus meeting the constraints. Therefore, the maximum capacity threshold for ride-hailing services in Ningbo is determined to be 30,400 drivers.

[0040] Example 2: This example is basically the same as Example 1, except that: in this example, the system dynamics model has undergone a historical validity check before use. The specific process is as follows: Step A1: Construct the test dataset, which includes historical datasets and simulated datasets, specifically: Historical Dataset: Acquire historical data and parameter data. Historical data includes the average monthly passenger demand, average monthly driver revenue, number of drivers, and average price of ride-hailing orders for each year. The historical data collection period should be no less than five years. Historical data is obtained from management departments, regulatory information systems, or legally authorized information channels responsible for transportation management, ride-hailing industry supervision, ride-hailing operation monitoring, or ride-hailing industry statistics within the target area; as shown in Table 3. Table 3. Ride-hailing data in Ningbo City, 2019-2024

[0041] The parameter data includes the external interference coefficient K of public safety incidents, where 2020 to 2022 take 2019 as the base year, and 2023 takes 2024 as the base year. The relative demand intensity index DI for each year affected by external shocks can be calculated by formula (14): DI 2020 DI is 0.88. 2021 DI is 0.79. 2022 DI is 0.86. 2023 The value is 0.96. The impact stage is determined based on the relative demand intensity index, thus giving the K value in this embodiment; the resident population growth rate β is 4.34%, and the waiting time elasticity coefficient E... w The price elasticity coefficient is -0.52, E p The coefficient is -1.30, and the coefficient for a full-time driver is f = 0.55. Simulated Dataset: Taking the simulated value calculation for the current year (year t) as an example, the derivation process is explained in detail: ① Given the 2023 data: average monthly passenger demand Q t-1 The number of orders was 8.594 million, and the external interference coefficient K for public safety incidents was [missing information]. t-1 The value is 0.85, and the number of drivers is N. t-1 For 28,700 people, the average price of ride-hailing orders is P t-1 It costs 21.2 yuan; 2024 data: External interference coefficient K for public safety incidents t The resident population growth rate β is 4.34%, and the waiting time elasticity coefficient E is 1. w The price elasticity coefficient is -0.52, E p =-1.30, number of drivers N t For 35,600 people, the average price of ride-hailing orders is P t It costs 20.6 yuan; ② Substitute into formula (13) to calculate the simulated value of the average monthly passenger demand in 2024: Substituting the known data above into formula (13), we obtain the simulated value Q' of the average monthly passenger demand in 2024. t The total number of orders was 10.85 million. ③ Substitute into formula (15) to calculate the simulated value R' of the average monthly revenue of drivers in 2024. t : The simulated value Q' of the average monthly passenger demand in 2024 t The number of ride-hailing orders reached 10.85 million, with an average order price of P. t The price is 20.6 yuan, and the number of drivers is N. t Substituting 35,600 people and a full-time driver coefficient f of 0.55 into formula (15), the simulated value R' of the average monthly revenue of drivers in 2024 is calculated. t It is 11,416 yuan; Using the method described above, simulated values ​​of average monthly passenger demand and average monthly driver revenue for each historical year were calculated sequentially to form a complete simulation dataset, as shown in Table 4. Table 4 Dataset for Validating the System Dynamics Model of Ningbo's Online Ride-Hailing Market

[0042] Step A2: The effectiveness of the system dynamics model is evaluated using the coefficient of difference (U). Average monthly passenger demand and average monthly driver revenue are selected as test indicators. Historical indicator data are uniformly denoted as A_t = {A_Q}. t , A_R t}, A_Q t For the historical average monthly passenger demand over year t, A_R t The average monthly revenue of drivers over year t; the simulated indicator data is uniformly denoted as S_t = {S_Q} t , S_R t}, S_Q t Let S_R be the simulated value of the average monthly passenger demand in year t. t Let be the simulated value of the average monthly revenue of drivers in year t; the formula for calculating the coefficient of difference (U) is expressed as in (18): (18)

[0043] Where U∈[0,1], the smaller U is, the better the fit. When U<0.40, the system dynamics model is considered to be well fitted. To verify the average value of historical indicators, including the historical average monthly passenger demand and the historical average monthly driver revenue; To verify the average values ​​of the simulated indicators, including the average simulated monthly passenger demand and the average simulated monthly driver revenue, the verification results are shown in Table 5. Furthermore, based on the actual and simulated values ​​of the two sets of verification indicators, historical verification fit plots for average monthly passenger demand and average monthly driver revenue were plotted, as detailed below. Figure 3 and Figure 4 As shown, the fitting effect of the system dynamics model is presented intuitively.

[0044] Table 5. Historical Test Results of the System Dynamics Model of Ningbo's Online Ride-Hailing Market

[0045] The results show that the monthly passenger demand difference coefficient is 0.0182 and the monthly driver revenue difference coefficient is 0.1591, both lower than the industry standard of 0.40. Furthermore, the fluctuation characteristics of the simulated value line and the actual value line in the two fitting graphs are highly consistent, which fully demonstrates that the system dynamics model constructed in this invention has a good fit.

[0046] In the practical application of the system dynamics model of this invention, parameter calibration and historical validity verification are performed first, followed by price constraints and capacity scenario simulation, specifically including: Step B1: Parameter initialization and constraint setting, setting the waiting time flexibility coefficient E w With price elasticity coefficient E p The constraint condition is E w <0、E p <0, and give the waiting time elasticity coefficient E w With price elasticity coefficient E p Determine the initial search interval: E w ∈[-0.8,-0.2]、E p ∈[-2.0,-0.8]; Simultaneously set the coarse and fine calibration step sizes for the waiting time elasticity coefficient and the price elasticity coefficient, with the coarse calibration step size for the waiting time elasticity coefficient and the price elasticity coefficient being ΔE respectively. p =0.10,ΔE w =0.02, the fine calibration step sizes for the waiting time elasticity coefficient and the price elasticity coefficient are ΔE respectively. p =0.02,ΔE w =0.005; Step B2: Parameter calibration, iteratively updating E based on historical datasets. w E p To minimize the difference coefficient U between the simulated dataset and the historical dataset obtained by formulas (13) and (15), the specific method is as follows: ① Coarse calibration stage: During the waiting time, the elasticity coefficient E w With price elasticity coefficient E p Within the initial search interval, take the midpoint of the initial search interval as its initial point, and set the coarse calibration step size ΔE. p =0.10、ΔE w =0.02 for combined enumeration; wherein, enumeration is performed with positive step size increasing from the initial point to the upper bound of the initial search interval, and enumeration is performed with negative step size decreasing from the initial point to the lower bound of the initial search interval; for each group of parameters, the difference coefficients U(Q) and U(R) of the test index are calculated, and the parameter group that minimizes the comprehensive difference coefficient is selected as the optimal point for coarse calibration; ② When the optimal point of coarse calibration satisfies U(Q) < 0.40 and U(R) < 0.40, the optimal point of coarse calibration is determined to be the required calibration parameter set and output; otherwise, the fine calibration stage is entered. ③ Fine calibration stage: The optimal point of coarse calibration is used as the initial point of fine calibration, and the fine calibration is performed according to the fine calibration step size ΔE. p =0.02、ΔE w=0.005 for combined enumeration, with the enumeration direction consistent with the coarse calibration stage; calculate the difference coefficient for each set of parameters and update the optimal point; further, when the change in the optimal difference coefficient between two adjacent rounds... When the iteration stops and the calibration parameter set is output, the iteration is stopped. Step B3: Historical validity test. Substitute the calibration parameter set output from step B2 into the system dynamics model and calculate the difference coefficients U(Q) and U(R) for the test indicators, namely, the average monthly passenger demand and the average monthly driver revenue. When both test indicators satisfy U < 0.40, the system dynamics model is deemed to be fit and proceeds to the subsequent fare and capacity scenario simulation. Step B4: Parameter adjustment and re-verification when verification fails. When any inspection index does not meet the requirement of U < 0.40, keep the initial search interval of Step B1 unchanged, and adjust the waiting time elasticity coefficient E within the initial search interval. w With price elasticity coefficient E p The initial point is changed, and steps B2 to B3 are repeated. The method for changing the initial point is as follows: taking the midpoint of the initial search interval as the reference initial point, a set of candidate initial points is constructed by applying a preset offset to the initial point within the initial search interval. The offset is taken as an integer multiple of 0.01 (±0.01, ±0.02, ±0.03), that is, taking the offset of "midpoint ±0.01, ±0.02, ±0.03" respectively based on the midpoint. Each time, a candidate initial point is selected as a new initial point to enter step B2 for parameter calibration, and the historical validity is checked in step B3 until a calibration parameter set that simultaneously satisfies U(Q)<0.40 and U(R)<0.40 is obtained. After the historical validity check is qualified, the system dynamics model is used to determine the maximum capacity under the minimum freight rate constraint.

[0047] In summary, the method for determining the minimum fare and maximum capacity for ride-hailing services to guarantee driver income, based on historical operating data and operating cost data of the target area, combined with the local minimum hourly wage, accurately calculates the minimum average monthly revenue and corresponding minimum average price per ride to guarantee drivers' basic income. This leads to the construction of a structured minimum fare standard that includes a starting price, mileage fee, time fee, and long-distance fee. Using this minimum average price per ride as a fare constraint, a system dynamics model is used to simulate the changes in average monthly revenue per driver under different numbers of drivers. Furthermore, by ensuring that the average monthly revenue per driver is not lower than the minimum average monthly revenue, the maximum capacity that the market can accommodate is determined. Thus, based on a clear minimum fare standard, the maximum capacity is collaboratively managed to ensure a balance between supply and demand in the ride-hailing industry, preventing disorderly growth in capacity that could dilute drivers' average monthly revenue, and providing a guarantee for ride-hailing drivers' income.

Claims

1. A method for determining the minimum fare and maximum capacity of ride-hailing services to guarantee drivers' income, characterized in that, Includes the following steps: S1. Establish a system dynamics model for the ride-hailing market, wherein the system dynamics model is at least related to the average monthly passenger demand, the number of drivers, and the average monthly driver revenue; S2. Obtain historical operating data and operating cost data of the ride-hailing industry in the target area, as well as industry benchmark data; S3. Based on the historical operating data and operating cost data, combined with the local minimum hourly wage, calculate the minimum monthly revenue and the corresponding minimum average price per order to guarantee the driver's basic income, and form a minimum fare standard including the starting price, mileage fee, time fee and long-distance fee based on the minimum average price per order. S4. Using the minimum average price per ride as a fare constraint, the system dynamics model is used to simulate the change in the average monthly revenue of drivers under different numbers of drivers, and the maximum capacity of the ride-hailing market in the target area is determined with the constraint that the average monthly revenue of drivers is not lower than the minimum average monthly revenue.

2. The method for determining the minimum fare and maximum capacity of ride-hailing services to guarantee driver income as described in claim 1, characterized in that, The historical operational data includes average monthly working days per vehicle, average daily operating hours per vehicle, average daily order volume per vehicle, average passenger mileage per order, average passenger duration per order, average mileage for long-distance orders, average empty driving time for long-distance orders, and average empty driving time for regular orders. The operational cost data includes the contract fee for pure electric ride-hailing vehicles, the contract fee for hybrid ride-hailing vehicles, fuel costs for pure electric ride-hailing vehicles, fuel costs for hybrid ride-hailing vehicles, market share of pure electric ride-hailing vehicles, market share of hybrid ride-hailing vehicles, average order completion bonus and average commission rate of ride-hailing platforms. The industry reference data includes minimum hourly wage, starting price for taxis, starting mileage for taxis, starting time for taxis, and mileage fee for taxis.

3. The method for determining the minimum fare and maximum capacity of ride-hailing services to guarantee driver income as described in claim 2, characterized in that, The minimum average monthly revenue is calculated based on the minimum hourly wage, the average number of working days per vehicle per month and the average daily operating time per vehicle in the historical operating data, and is further adjusted by combining the vehicle contracting fee, fuel cost, platform rewards and commission rate in the operating cost data.

4. The method for determining the minimum fare and maximum capacity of ride-hailing services to guarantee driver income as described in claim 3, characterized in that, The minimum average price per unit is determined by dividing the minimum average monthly revenue by the average monthly order volume per vehicle obtained from the historical operating data.

5. The method for determining the minimum fare and maximum capacity of ride-hailing services to guarantee driver income as described in claim 1, characterized in that, The mileage fee and time fee in the minimum fare standard meet the following constraints: the starting price of the taxi + mileage fee × K1 + time fee × K2 ≥ the minimum average price per order; where K1 is the difference between the average passenger mileage of the order obtained from historical operating data and the starting mileage of the taxi, and K2 is the difference between the average passenger time of the order obtained from historical operating data and the starting time of the taxi.

6. The method for determining the minimum fare and maximum capacity of ride-hailing services to guarantee driver income as described in claim 5, characterized in that, The determination of the long-haul fee for the minimum freight rate standard is related to the unit-time revenue converted from the minimum average monthly revenue, and is used to compensate for the additional empty-running costs of long-haul orders compared to regular orders.

7. The method for determining the minimum fare and maximum capacity of ride-hailing services to guarantee driver income as described in claim 1, characterized in that, The system dynamics model includes an evolution equation for average monthly passenger demand and an evolution equation for average monthly driver revenue. The evolution equation for average monthly passenger demand uses average monthly passenger demand (Q) as a state variable, and its evolution is affected by the average price of ride-hailing orders (P), the number of drivers (N), the growth rate of the resident population (β), and the external disturbance coefficient (K) of public safety events. The evolution equation for average monthly driver revenue is used to calculate the average monthly driver revenue (R), and its input variables include the average monthly passenger demand (Q), the average price of ride-hailing orders (P), the number of drivers (N), and the full-time driver coefficient (f).

8. The method for determining the minimum fare and maximum capacity of ride-hailing services to guarantee driver income as described in claim 1, characterized in that, The specific method for determining the maximum capacity of the ride-hailing market in the target area is as follows: using the minimum average price per ride as a fare constraint, the system dynamics model is used to simulate the average monthly revenue of drivers under different driver number scenarios; from all simulated scenarios where the average monthly revenue of drivers is not lower than the minimum average monthly revenue, the maximum number of drivers corresponding to the maximum capacity is selected.

Images