A low-carbon oriented multi-category vehicle route guidance method under MaaS background

Abstract

This invention discloses a low-carbon-oriented multi-category vehicle routing induction method under the MaaS (MaaS) context, comprising the following steps: S1, calculating the marginal travel time of multiple vehicle categories; S2, determining a carbon emissions formula and then calculating marginal carbon emissions; S3, constructing routing pricing rules that meet the carbon reduction induction requirements; S4, determining factors influencing customer choices and then constructing a discrete choice model using a binomial logit model; S5, using simulation software to simulate an actual road network and then evaluating key indicators. This invention can effectively reflect the impact of changes in platform pricing strategies on transportation carbon emissions.

Description

Technical Field

[0001] The present invention relates to a vehicle route induction method, and in particular to a low-carbon oriented multi-category vehicle route induction method under the MaaS context. Background Art

[0002] In recent years, with the rapid development of emerging technologies such as autonomous driving, vehicle-road collaboration, intelligent connectivity, and shared transportation, the road travel environment has significantly improved. However, traffic congestion remains a prominent problem in some urban areas, seriously affecting people's travel efficiency. The concept of Mobility as a Service (MaaS) has provided a new approach to improving urban traffic congestion and is considered one of the development trends of future transportation systems. MaaS integrates travel services from various modes of transportation, allowing travelers to view travel as a service. Instead of purchasing transportation, travelers can purchase travel services provided by different operators based on their travel needs, effectively improving the user travel experience. Therefore, if MaaS platforms can rationally analyze travelers' travel behavior choices and formulate reasonable policies to guide travelers to choose the system's optimal routes, they can achieve social goals such as alleviating traffic congestion and reducing carbon emissions.

[0003] It's generally believed that there are three types of participants in a transportation system: User Equilibrium (UE) participants, or ordinary commuters, who consider only their own interests; Cournot-Nash (CN) participants, such as private transportation companies, who consider the overall interests of their companies; and System Optimal (SO) participants, such as government-controlled vehicle operating platforms, who consider the interests of all participants in the system. In the context of MaaS, CN and SO participants will become the mainstream participants, and vehicles owned by MaaS platforms (i.e., CN and SO platforms) will account for the majority of vehicles on the road.

[0004] In the context of MaaS, the structure of travel route decision-makers is shifting. Traditionally, the decision-makers for travel route selection are considered travelers themselves (i.e., UE travelers). However, in the MaaS context, a large portion of cars on the road are operated by MaaS platforms. Driven by their own interests, MaaS platforms can gain some control over route selection by formulating rules. However, existing MaaS platforms (such as Didi) often prioritize transporting passengers to their destinations, failing to fully consider the socioeconomic attributes of different individuals and their varying needs for travel services. Therefore, platforms need to account for the heterogeneity of passenger travel and optimize their pricing rules and operational strategies from the perspective of route decision-makers. Furthermore, because carbon emissions are a typical economic externality, platforms fail to factor them into their profit calculations. Consequently, while pursuing profit maximization, they neglect carbon emissions, which is inconsistent with carbon reduction goals. Summary of the Invention

[0005] Purpose of the invention: The purpose of the present invention is to provide a low-carbon oriented multi-category vehicle route induction method in the context of MaaS, which can induce passengers to change their route choices through reasonable pricing, thereby alleviating traffic congestion and reducing traffic carbon emissions.

[0006] Technical solution: The multi-category vehicle path guidance method of the present invention includes the following steps:

[0007] S1, calculates the marginal travel time of multiple categories of vehicles;

[0008] S2, determine the carbon emission formula and then calculate the marginal carbon emission;

[0009] S3, construct path pricing rules that meet the demand for carbon reduction induction;

[0010] S4, determine the factors that influence customer choice and then use the binomial logit model to construct a discrete choice model;

[0011] S5, uses simulation software to simulate the actual road network and then evaluates the key indicators.

[0012] Further, in step S1, for When the participants of type A are driving on road section a, The required travel time is calculated as follows:

[0013]

[0014] in, For a certain type of participant, For section a Traffic of class participants; For road section a, it is not The flow of class participants, t a () refers to the road resistance function of section a, t a ′() refers to the derivative of the road resistance function of section a with respect to the flow rate; UE is a general traveler who only considers his own interests; CN1 and CN2 are any two different private transportation companies, and the overall interests of the companies are considered; SO is a government-controlled vehicle operation platform, and the interests of all participants in the entire system are considered;

[0015] The marginal travel time is calculated as follows:

[0016]

[0017] Among them, N a is the number of vehicles on road section a, Δt a The time that the travel time of each CN vehicle on road section a increases after the unit flow is increased.

[0018] Furthermore, in step S2, the total marginal carbon emissions are calculated as follows:

[0019] MarV=∑ a ΔV a ·N a

[0020]

[0021] Among them, MarV is the total marginal carbon emission (kg), a is each road section on the vehicle’s travel path, ΔV a is the additional carbon emissions (kg) caused by the vehicle to other vehicles on road section a, N a is the number of vehicles on road section a; ω is the energy density of gasoline, EF is the emission factor, P e is the engine power, b e is the engine specific fuel consumption.

[0022] Furthermore, in step S3, the expression of the path pricing rule is as follows:

[0023]

[0024] Among them, A is the set of all road segments included in the path, N a is the number of vehicles belonging to the platform on road segment a, B is the starting price, w is the fee per kilometer, d0 is the mileage included in the starting price, d is the distance between the starting and ending points, k is the time compensation coefficient, l is the carbon emission coefficient, and t represents the travel time of a certain road under the current road network conditions;

[0025] V is carbon emissions, expressed as follows:

[0026] V=V perkm ·Distance·ω·EF

[0027] t 期望 It represents the expected time given by the departure to destination system, and the expression is as follows:

[0028]

[0029] Among them, s represents the shortest path length between two places, v 平均 Indicates the average speed of vehicles in the area during the period.

[0030] Furthermore, in step S4, the expression of the discrete choice model is:

[0031] V UE =β1+β2×income+β3×urgency+β4×Time UE +β5×W UE

[0032] V CN =β2×income+β3×urgency+β4×Time CN +β5×W CN

[0033] The probability that the customer chooses the UE path is:

[0034]

[0035] The probability of choosing CN path is:

[0036]

[0037] Among them, income represents the individual's income level, with a value of 0 corresponding to a low level and a value of 1 corresponding to a high level; urgency represents the urgency of an individual's travel, with a value of 0 corresponding to non-urgent and a value of 1 corresponding to urgent;

[0038] Time UE 、Time CN are the travel time required to travel in the current UE and CN paths respectively; W UE 、W CN are the prices given for the current UE and CN paths respectively; β1, β2, β3, β4, and β5 are weight coefficients.

[0039] Compared with the prior art, the present invention has the following significant effects:

[0040] 1. By adopting the marginal carbon emissions calculation method, carbon emissions can be effectively calculated. By implementing route guidance that takes carbon emissions into consideration, different travel routes can be provided to platform passengers, which can reduce total travel time, increase vehicle speed, reduce fuel consumption, and thus reduce carbon emissions.

[0041] 2. A pricing method that internalizes carbon emission costs is proposed. By adjusting the carbon emission coefficient in the CN platform pricing rules, the intensity of carbon tax collection is simulated, and the system operation under different scenarios is simulated. It has strong portability and can effectively reduce the total operating time and total carbon emissions of the system vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a flow chart of the present invention;

[0043] Figure 2 is a road network map of the study area in the present invention;

[0044] Figure 3(a) is a comparison of the congestion levels of the road sections in the area before the system was put into operation.

[0045] Figure 3(b) is a comparison of the congestion levels of the road sections in the area after the system was put into operation;

[0046] Figure 4 Comparison of platform revenue and total system emissions under different price increases;

[0047] Figure 5 Comparison of system average travel time indicators under different market structure environments;

[0048] Figure 6 Comparison of total system carbon emissions indicators under different market structures. DETAILED DESCRIPTION

[0049] The present invention will be described in further detail below with reference to the accompanying drawings and specific implementations.

[0050] This invention fully considers the heterogeneity among different passengers and proposes a method to induce passengers to change their route choices through reasonable pricing in the context of MaaS, thereby effectively alleviating traffic congestion and improving traffic efficiency. Figure 1 As shown, the specific implementation process includes the following steps:

[0051] Step 1: Calculate marginal travel time

[0052] In this embodiment, the system consists of UE travelers, an SO platform, and two CN platforms (denoted as CN1 and CN2). Their mode of travel is all cars: among them, UE (User Equilibrium) participants, that is, general travelers, only consider their own interests; CN (Cournot-Nash) participants, such as private transportation companies, consider the overall interests of the company; SO (System Optimal) participants, such as government-controlled vehicle operation platforms, consider the interests of all participants in the entire system.

[0053] belong When a car of type UE, SO, CN1, or CN2 travels on road section a, the required travel time is calculated as follows:

[0054]

[0055] in, For a certain type of participant, For section a The flow of class participants, For road section a, it is not The flow of class participants, t a () refers to the road resistance function of section a, t a ′() refers to the derivative of the road resistance function of section a with respect to the flow rate.

[0056] Therefore, for the CN platform, after the unit flow of road section a is increased, the travel time of each CN vehicle on the road section increases by:

[0057]

[0058] Therefore, the marginal travel time is:

[0059] MarT=∑ a Δt a ·N a (3)

[0060] Among them, N a is the number of vehicles on road segment a.

[0061] To calculate the marginal travel costs of different types of travelers, it is necessary to calculate the derivative of the segment travel time. Therefore, the present invention uses a quadratic function to fit the segment resistance function. This invention assumes that the time-varying nature of this derivative stems from changes in the segment travel time function. Since this function depends on the condition of the segment, and the traffic conditions on a segment do not change significantly over a short period of time, it is assumed that several consecutive observation points within a short time interval will fall on the same segment travel time function curve. In this embodiment, the segment flow rate is calculated once every 15 simulation steps. After obtaining six sets of (v, t) points, a quadratic function is used to fit this function, thereby calculating its derivative as the derivative of the segment travel time.

[0062] Step 2: Calculation of marginal carbon emissions

[0063] Carbon emissions are calculated using the following formula:

[0064] V=V perkm ·Distance·ω·EF (4)

[0065] Where V is the mass of carbon dioxide emissions, Distance is the driving distance, ω is the energy density of gasoline, and EF is the emission factor.

[0066] Since the average fuel consumption per kilometer is V perkm Speed-related:

[0067]

[0068] Among them, v a is the speed of the car, P e is the engine power, b e is the engine specific fuel consumption. For a given path, the distance is a constant value, so carbon emissions are mainly related to driving speed. Therefore, the marginal carbon emissions can be calculated by the extra travel time brought to other vehicles by choosing a path. For a car x on road section a, let the length of road section a be L a , then its carbon emission increment is:

[0069]

[0070] The calculation formula for total marginal carbon emissions is proposed as follows:

[0071] MarV=∑ a ΔV a ·N a (7)

[0072] Among them, MarV is the total marginal carbon emission, a is each road section on the vehicle’s travel path, ΔV aThe additional carbon emissions caused by the vehicle to other vehicles on road section a can be converted from the additional travel time caused by the vehicle to other vehicles. N a is the number of vehicles on road segment a.

[0073] Step 3: Determine pricing rules

[0074] This invention assumes that the routes of UE and SO travelers are fixed, while assuming that the CN platform provides its internal users with two optional routes tailored to their needs: one is the CN platform's optimal route, determined by the platform based on revenue maximization, and the other is the route with the shortest travel time, the UE route. However, travelers who choose the UE route must pay a higher fee than the platform's optimal route to compensate for the marginal impact of their choice on other users. To this end, a reasonable pricing strategy is needed that reflects the differences in route selection behavior caused by individual users while meeting these requirements.

[0075] The present invention divides the cost into three parts:

[0076] (f1) Basic Fees

[0077] It is set as: basic fee = starting price + mileage fee - time compensation fee.

[0078] The base price is the fee charged upon invoking the MaaS platform's services, regardless of the trip length or duration. The base price includes a certain mileage limit, and no additional mileage charges apply within the mileage limit. However, additional charges apply for mileage exceeding the limit. The time compensation fee refers to the system's calculation of the expected travel time for a particular route based on the average speed in the area. If the travel time exceeds the expected time, the total cost will be reduced as compensation; otherwise, the total cost will increase. This design allows for the calculation of the marginal costs associated with adding a vehicle.

[0079] (f2) Carbon emission fees

[0080] Carbon emission fees are charged based on the carbon emissions of each passenger's travel.

[0081] (f3) Marginal cost

[0082] As an induction system, the sum of the marginal costs brought by the vehicle must also be added to the cost.

[0083] In summary, the CN platform pricing model that simulates the intensity of carbon tax charges is as follows:

[0084]

[0085] Where A is the set of all road segments included in the path, B is the starting price, w is the fee per kilometer, d0 is the mileage included in the starting price, d is the distance between the starting and ending points, k is the time compensation coefficient, t represents the travel time of a certain road under the current road network conditions, MarT is the marginal travel time calculated by formula (3), l is the carbon emission coefficient, V is the carbon emission calculated by formula (4), and MarV is the marginal carbon emission calculated by formula (7).

[0086] t 期望 It represents the expected time given by the system from the departure point to the destination, which can be obtained by the following formula:

[0087]

[0088] Among them, s represents the shortest path length between two places, v 平均 It represents the average speed of vehicles in the area during the period, which can be obtained through other methods such as consulting data and big data analysis.

[0089] Step 4: Path selection model construction

[0090] The specific form and parameter values ​​of the discrete choice model can be determined by SP survey (Stated Preference Survey). This paper uses the binomial logit model for path selection as an example, and the final model is:

[0091] V UE =β1+β2×income+β3×urgency+β4×Time UE +β5×W UE (10)

[0092] V CN =β2×income+β3×urgency+β4×Time CN +β5×W CN (11)

[0093] Among them, income represents the individual's income level, with a value of 0 corresponding to a low level and 1 corresponding to a high level; urgency represents the urgency of an individual's travel, with a value of 0 corresponding to non-urgent (corresponding to leisure travel) and a value of 1 corresponding to urgent (corresponding to commuting travel); Time UE 、Time CN Represents the travel time required to travel in the current UE and CN paths respectively; W UE 、W CNRepresent the current UE and CN path prices calculated in step 3. The weight coefficients β1, β2, β3, β4, and β5 of each variable follow a lognormal distribution. To reduce the number of parameters, it is assumed that individual socioeconomic attributes other than income level have similar impacts on different path choices.

[0094] On this basis, the probability that the customer chooses the UE path is:

[0095]

[0096] The probability of choosing CN path is:

[0097]

[0098] Step 5: Multi-scenario simulation evaluation

[0099] In the context of low carbon, the MaaS system objective shifts from optimizing efficiency to weighted optimization of carbon emissions and travel efficiency. MaaS platforms now incorporate carbon emissions into their route planning. Using simulation software, we compared the system's total carbon emissions before and after joining the MaaS system to analyze the extent to which the MaaS platform reduces total carbon emissions.

[0100] This embodiment is further described by taking the area enclosed by Zhujiang Road, Longpan Middle Road, Daguang Road, Guanghua East Street, Zijin Road and Alfalfa Garden Street in Nanjing as the research location. The regional road network of this embodiment includes the area enclosed by Zhujiang Road, Longpan Middle Road, Daguang Road, Guanghua East Street, Zijin Road and Alfalfa Garden Street, ignoring the roads with too low a grade. The road network diagram is as follows: Figure 2 The embodiment uses the Nissan Sylphy 1.6XE CVT Comfort Edition as a standard sedan, assuming an average urban speed of 20 km / h, gasoline energy density ω = 12 kWh / kg, and emission factor EF = 69300 kg / TJ.

[0101] According to the CN and SO platform pricing formulas in formula (8), in this embodiment, B = 4, d0 = 1.75, w = 1.75, k = 2, and l = 0.05 are selected as known parameters. The simulation results show that after considering marginal travel time and marginal carbon emissions and inducing, the average travel time of the system decreased by 5.1%, and the overall operating efficiency was improved. Figure 3(a) 、 3(b) As shown, the total carbon emission of the system is changed from 2753kg to 2640kg, a reduction of 4.1%, which is in line with the induction target of the present invention.

[0102] The simulation results of SUMO software show that if the platform makes a decision to partially increase prices, its own revenue and the total carbon emissions of the system will decrease. However, if the price increase is large, the platform's revenue will increase, but the cost is that the overall carbon emissions of the system will also increase accordingly. Figure 4 shown.

[0103] This example also analyzes the system's operating conditions under different market structures, selecting a monopoly market (UE:SO:CN1:CN2=1:1:8:0), a duopoly market (UE:SO:CN1:CN2=1:1:6:2), and a market containing only CN but not SO (UE:SO:CN1:CN2=1:0:6.5:2.5). The results show that in the case of a single-platform monopoly market, the system's carbon emissions are 7.5% higher than those in a duopoly market. Figure 5 、 Figure 6 As shown, this proves that the existence of healthy competition helps to reduce system carbon emissions and achieve energy conservation and emission reduction goals.

[0104] The above description is merely a preferred embodiment of the present invention and does not constitute any other form of limitation to the present invention. Any modification or equivalent variation based on the technical essence of the present invention shall still fall within the scope of protection claimed by the present invention.

Claims

1. A low-carbon oriented multi-category vehicle route guidance method under the MaaS context, characterized by: The following steps are involved: S1, calculates the marginal travel time of multiple categories of vehicles; S2, determine the carbon emission formula and then calculate the marginal carbon emission; S3, construct path pricing rules that meet the demand for carbon reduction induction; S4, determine the factors that influence customer choice and then use the binomial logit model to construct a discrete choice model; S5, using simulation software to simulate the actual road network and then evaluate key indicators; In step S2, the total marginal carbon emissions are calculated as follows: MarV=∑ a ΔV a ·N a Among them, MarV is the total marginal carbon emission, in kg; a is each road section on the vehicle’s route; ΔV a is the additional carbon emissions caused by the vehicle to other vehicles on road section a, in kg; N a is the number of vehicles on road section a; ω is the energy density of gasoline, EF is the emission factor, P e is the engine power, b e is the engine specific fuel consumption; In step S3, the expression of the path pricing rule is as follows: Among them, A is the set of all road segments included in the path, N a is the number of vehicles belonging to the platform on road segment a, B is the starting price, w is the fee per kilometer, d0 is the mileage included in the starting price, d is the distance between the starting and ending points, k is the time compensation coefficient, l is the carbon emission coefficient, and t represents the travel time of a certain road under the current road network conditions; V is carbon emissions, expressed as follows: V=V perkm ·Distance·ω·EF t 期望 It represents the expected time given by the departure to destination system, and the expression is as follows: Among them, s represents the shortest path length between two places, v 平均 Indicates the average speed of vehicles in the area during the period; V perkm Indicates average fuel consumption per kilometer.

2. The low-carbon oriented multi-category vehicle route guidance method under the MaaS context according to claim 1 is characterized in that: In step S1, for When the participants of type A are driving on road section a, The required travel time is calculated as follows: in, For a certain type of participant, For section a Traffic of class participants; For road section a, it is not The flow of class participants, t a () refers to the road resistance function of section a, t a ′() refers to the derivative of the road resistance function of section a with respect to the flow rate; UE is a general traveler who only considers his own interests; CN1 and CN2 are any two different private transportation companies, and the overall interests of the companies are considered; SO is a government-controlled vehicle operation platform, and the interests of all participants in the entire system are considered; The marginal travel time is calculated as follows: Among them, N a is the number of vehicles on road section a, Δt a The time that the travel time of each CN vehicle on road section a increases after the unit flow is increased.

3. The low-carbon oriented multi-category vehicle route guidance method under the MaaS context according to claim 2 is characterized in that: In step S4, the expression of the discrete choice model is: V UE =β1+β2×income+β3×urgency+β4×Time UE +β5×W UE In CN =β2×income+β3×urgency+β4×Time CN +β5×W CN The probability that the customer chooses the UE path is: The probability of choosing CN path is: Among them, income represents the individual's income level, with a value of 0 corresponding to a low level and a value of 1 corresponding to a high level; urgency represents the urgency of an individual's travel, with a value of 0 corresponding to non-urgent and a value of 1 corresponding to urgent; Time UE 、Time CN are the travel time required to travel in the current UE and CN paths respectively; W UE 、W CN are the prices given for the current UE and CN paths respectively; β1, β2, β3, β4, and β5 are weight coefficients.

Images