Car-pooling optimization method considering demand adaptation and balanced pricing

Abstract

The invention belongs to the technical field of intelligent traffic and shared travel, and particularly relates to a car-pooling optimization method considering demand adaptation and balanced pricing, and the method comprises the steps: obtaining passenger information; a demand adaptation graph is constructed based on passenger information, the passenger information comprises a binary feature demand, a double-binary feature demand and a quantitative feature demand, and the binary feature demand represents the binarization tendency of the passenger to the first type of behaviors; the dual-binary feature demand represents the binarization tendency of the passenger to the second type of behaviors and the acceptance degree of the passenger to the behaviors of other people; the quantitative characteristic demand represents an acceptable range of the passenger for the third type of behaviors; determining adaptation results of individual demands of different passengers based on the adaptation function, and forming connection edges based on the adaptation results; forming a demand adaptation graph based on the passenger information and the connecting edges; forming an alliance structure based on the demand adaptation graph, and searching in the alliance structure to obtain an optimal alliance structure meeting constraints; and based on the optimal alliance structure, calculating a payment result meeting an equilibrium condition through surplus calculation.

Description

Technical Field

[0001] The present application belongs to the field of intelligent transportation and shared travel technology, and specifically relates to a carpooling optimization method that considers demand adaptation and balanced pricing. Background Art

[0002] With the widespread use of shared travel platforms, carpooling has gradually become an important means to improve the utilization of transportation resources and alleviate urban congestion. Existing carpooling matching methods mainly match the shortest path or minimum cost based on spatial and temporal constraints, often ignoring the service conflicts that may arise from the personalized needs of passengers, thus affecting the stability and acceptance of carpooling matching. In addition, the current common cost-sharing mechanisms are mostly simple rules based on path or time, lacking systematic consideration of the fairness of passenger payments. This may cause individual dissatisfaction or even withdrawal from carpooling, making the matching results unimplementable and affecting the overall platform revenue and user experience. Therefore, how to balance the adaptation of personalized needs and balanced cost sharing, and design a stable carpooling matching and pricing mechanism, is a key issue that shared travel platforms urgently need to solve. Summary of the Invention

[0003] This application is proposed based on the above-mentioned needs of the prior art. The technical problem to be solved by this application is to provide a carpooling optimization method that takes demand adaptation and balanced pricing into consideration so as to perform carpooling matching in a targeted manner.

[0004] In order to solve the above problems, the technical solutions provided in this application include:

[0005] A carpooling optimization method considering demand adaptation and equilibrium pricing is provided, comprising: constructing a demand adaptation graph based on passenger information, the passenger information including binary feature demands, dual binary feature demands, and quantitative feature demands, wherein the binary feature demands represent the passenger's tendency to binaryize a first type of behavior; the dual binary feature demands represent the passenger's tendency to binaryize a second type of behavior and the passenger's acceptance of the behavior of others; and the quantitative feature demands represent the passenger's acceptable range for a third type of behavior; determining adaptation results of different passengers' personalized demands based on a demand adaptation function, and forming edges based on the adaptation results; forming a demand adaptation graph based on the passenger information and the edges; forming an alliance structure based on the demand adaptation graph, searching within the alliance structure to obtain an optimal alliance structure that satisfies constraints, the optimal alliance structure being one in which the personalized demands of the passengers are mutually adapted and the total cost of the alliance structure is minimized; and calculating a payment result that satisfies equilibrium conditions based on the optimal alliance structure through surplus, including calculating a payment transfer amount based on a maximum surplus difference if the current payment scheme has not reached equilibrium, recalculating the maximum surplus difference based on the payment transfer amount, and repeating the iteration until the current scheme reaches equilibrium.

[0006] Preferably, the demand adaptation function is Among them, f k (·) is the adaptability judgment operator of the kth demand feature, F(p i ,p j )=1 means passenger p i With passengers j The personalized needs of each other are adapted; F(p i ,p j )=0 means passenger p i With passengers j There are conflicts in individual needs.

[0007] Preferably, the adaptability determination operator has different values ​​for different requirements, including: for binary feature requirements, the adaptability determination operator satisfies: f k (α i ,α j )=1-|α i -α j |,α i ∈{0,1},α i =1 means passenger p i With this tendency, α i =0 means not available; for dual binary feature requirements, the adaptability judgment operator satisfies: f k (β i ,β j )=sgn(β i2 -β j1 )·sgn(β j2 -β i1 ), β i =(β i1 ,β i2 ), β i1 ∈{0,1} represents passenger p i Do you have this tendency? i2 ∈{0,1} represents passenger p i Whether to accept that others have this tendency; for quantitative feature requirements, the adaptability judgment operator satisfies: in in,

[0008] Preferably, the demand adaptation function satisfies: F(p i ,p i )=1,F(p i ,p j )=F(p j ,p i ),

[0009] Preferably, the total cost of the alliance structure includes the sum of all alliance costs, and the alliance cost is expressed as u(C)=d(R * (C))·(c f +c d +c v ), where u(C) is the alliance cost, d(R * (C)) is the mileage of the alliance’s optimal path, c f 、c d with c v They represent fuel cost per unit mileage, driver cost and vehicle depreciation cost respectively.

[0010] Preferably, the road network is a graph M = (N, L), where N is the set of boarding points and alighting points. is the shortest path set, which means the shortest distance between all the places corresponding to the pick-up and drop-off points. The places corresponding to the pick-up and drop-off points are represented by R = (n1, n2, ..., n m ), n i ∈N, the vehicle will travel along the above point sequence as a path, and the valid path set is satisfy: Among them, R[k] represents the kth element in R, that is, the corresponding n k .

[0011] Preferably, the alliance optimal path R * (C) is the shortest path required for the vehicle to serve passengers in alliance C, expressed as: where d R[k],R[k+1] Represents the distance between corresponding locations of adjacent point sequences.

[0012] Preferably, the searching in the alliance structure to obtain the optimal alliance structure that satisfies the constraints includes: adapting the demand to each node p of the graph G∈(P,E) i Convert to an alliance formed by individual passengers to form the initial alliance structure All edges e ij ∈E is marked as shrinkable, and the search frontier stack is initialized and the initial cost As the current optimal value; where each node p in the adaptation graph i ∈P represents a passenger if and only if passenger p i With p j When the personalized needs of ij ∈E, represents passenger p i With p jThere is an opportunity to share a ride. The alliance is a set of passengers arranged to ride in the same vehicle, which satisfies the size constraint 1≤C≤λ, where λ is the maximum number of passengers that the vehicle can carry; pop the top element of the current search front stack Select a contractible edge e from ij , check the new alliance structure generated by performing edge contraction on this edge Whether the scale constraint and demand adaptation constraint are met; if not, mark the edge as non-shrinkable; otherwise, perform edge shrinkage and mark the edge as non-shrinkable to avoid repeated exploration; calculate If it is less than the current optimal value, update the current optimal value to Will Push The edge contraction includes converting the edge e ij The two connected alliances C i with C j Merged into the new Alliance C ij , generating a new alliance structure For the newly generated alliance structure Arrange the alliances in ascending order of alliance cost, i.e. Where u(C1)≤u(C2≤...≤u(C k );calculate The lower bound like If the value is greater than the current optimal value, it is considered that the node and its child nodes cannot produce a better solution and are no longer pushed into the search frontier stack.

[0013] Preferably, the payment result that satisfies the equilibrium condition through surplus calculation based on the optimal alliance structure includes: the payment result is represented by the payment vector Indicates that x i For passengers i ∈C, the payment vector satisfies The equilibrium condition is where s ij is called surplus, which means passenger p i Leave alliance C and join alliance C which does not include passenger p j Other alliances The maximum additional cost savings that can be obtained when

[0014] Preferably, the payment result that satisfies the equilibrium condition through surplus calculation based on the optimal alliance structure further includes: calculating equilibrium pricing based on the optimal alliance structure, including: evenly allocating alliance costs: Calculate the maximum surplus difference: The condition δ≤ε·u(C) is used to determine whether the current payment scheme has reached equilibrium, where ε is the preset convergence threshold; if the current payment scheme has not reached equilibrium, the maximum surplus difference is set by the passenger (p i ,p j ) is generated, and payment transfer amount is calculated Update the payment vector according to the transfer amount d, let x i ←x i +d,x j ←x j -d, recalculate the maximum surplus difference and perform iterative updates of the payment vector in subsequent steps.

[0015] Compared with existing technologies, this application matches passengers based on their personalized needs, thereby improving the stability of carpooling results and passenger acceptance. The stability of carpooling results is generally measured by two factors: one is preference stability, that is, whether the personalized needs of passengers are mutually compatible; the other is payment stability, that is, whether the payment reflects the bargaining power and fairness of passengers. This application takes into account both stability factors under a unified framework. Specifically, this application:

[0016] 1) Carpooling optimization that balances personalization and system efficiency. By constructing a passenger demand adaptation graph that includes binary, double-binary, and quantitative features, it achieves comprehensive modeling of personalized needs. This fundamentally solves the service conflict problem caused by traditional methods that ignore passenger behavioral preferences, and effectively improves the stability and acceptance of carpooling services.

[0017] 2) Achieve higher-quality alliance matching results: By introducing a demand adaptation graph and an optimal alliance structure search algorithm, the present invention can identify the lowest-cost carpooling combination while satisfying personalized constraints, ensuring system efficiency while enhancing behavioral compatibility among passengers and improving matching quality.

[0018] 3) Model expansion to support heterogeneous demands and diverse behaviors: Thanks to the highly modular design of the adaptation function and coalition search algorithm, this method can flexibly embed new demand feature types or adaptation rules without changing the overall solution process, and has good scalability and the ability to adapt to different operating scenarios.

[0019] 4) The optimization algorithm has good computational efficiency and deployability: The depth-first search combined with the upper bound pruning strategy adopted by this invention effectively reduces the computational complexity of alliance structure generation, enabling real-time deployment on small and medium-sized platforms. At the same time, the cache mechanism improves the query efficiency of repeated alliance costs, meeting the technical requirements for practical implementation. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this specification. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0021] Figure 1 The figure is a flowchart of the steps of a carpooling optimization method considering demand adaptation and balanced pricing in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0023] In the description of the embodiments of this application, it should be noted that, unless otherwise expressly specified or limited, the term "connected" should be understood in a broad sense. For example, it can mean a fixed connection, a detachable connection, or an integral connection. It can be a mechanical connection, an electrical connection, a direct connection, or an indirect connection through an intermediate medium. Those skilled in the art will understand the specific meanings of the above terms in this application based on specific circumstances.

[0024] The terms "top," "bottom," "above," "below," and "on" used throughout the description refer to relative positions of components of a device, such as the relative positions of top and bottom substrates within a device. It will be understood that devices are multifunctional regardless of their orientation in space.

[0025] To facilitate understanding of the embodiments of the present application, further explanation will be given below with reference to specific embodiments in conjunction with the accompanying drawings. The embodiments do not constitute a limitation on the embodiments of the present application.

[0026] This embodiment provides a carpooling optimization method that considers demand adaptation and balanced pricing. Figure 1 shown.

[0027] The carpooling optimization method considering demand adaptation and equilibrium pricing includes:

[0028] Get passenger information.

[0029] The passenger information includes the passenger's personalized needs.

[0030] Build a demand adaptation graph based on passenger information.

[0031] The demand adaptation graph is represented as G = (P, E), which is used to describe the demand adaptation relationship between passengers. In the demand adaptation graph, each node p i ∈P represents a passenger if and only if passenger p i With p j When the personalized needs of ij ∈E, represents passenger p i With p j There is an opportunity to rideshare.

[0032] The personalized demand is represented by a vector v=(v1, v2, ..., v n ) description, where each element v j (j=1,2,...,n) is called demand characteristics, n is the number of demand characteristics. i The corresponding personalized needs are determined by v i =(v i1 ,v i2 ,...,v in )describe.

[0033] The demand characteristics are divided into the following categories:

[0034] (1) Binary feature: α i ∈{0,1}, used to describe the passenger's binary tendency towards a certain behavior, such as communicating, listening to music, opening the window, and passing through toll roads. i =1 means passenger p i With this tendency, α i =0 means not available.

[0035] (2) Double binary features: β i =(β i1 ,β i2 ), used to describe the passenger's binary tendency towards a certain behavior and the degree of acceptance of the behavioral tendencies of others, such as smoking, carrying luggage, carrying pets, and gender matching. i1 ∈{0,1} represents passenger p i Do you have this tendency? i2 ∈{0,1} represents passenger p i Do you accept that others have this tendency?

[0036] (3) Quantitative characteristics: Used to describe the passenger's acceptable range for a certain behavior, such as departure time, air conditioning temperature, etc.

[0037] Passenger p i With pj Whether the personalized needs of the users are compatible with each other is determined by the need adaptation function.

[0038] The demand adaptation function is: Among them, f k (·) is the adaptability judgment operator of the kth demand feature. F(p i ,p j )=1 means passenger p i With passengers j The personalized needs of each other are adapted; F(p i ,p j )=0 means passenger p i With passengers j There are conflicts in individual needs.

[0039] The adaptability determination operator satisfies:

[0040] (1) For binary features: f k (α i ,α j )=1-|α i -α j |;

[0041] (2) For dual binary features: f k (β i ,β j )=sgn(β i2 -β j1 )·sgn(β j2 -β i1 );

[0042] (3) For quantitative characteristics:

[0043] in,

[0044] The demand adaptation function satisfies the following properties:

[0045] (1) Reflexivity: F(p i ,p i )=1

[0046] (2) Symmetry: F(p i ,p j )=F(p j ,p i )

[0047] (3) Non-transitive:

[0048] An alliance structure is formed based on the demand adaptation graph, and a search is performed in the alliance structure to obtain the optimal alliance structure that meets the constraints.

[0049] The alliance structure is a passenger matching solution, namely Each subset in represents a coalition and satisfies In addition, the formation of an alliance also needs to meet the following requirements: the subgraph of the adaptation graph G induced by the alliance C is a complete graph, that is, That is, the corresponding passengers who form an edge can form an alliance.

[0050] The coalition C∈2 P is the set of passengers arranged to take the same vehicle, which satisfies the size constraint 1≤|C|≤λ, where λ is the maximum number of passengers that the vehicle can carry.

[0051] The alliance has a corresponding alliance cost, which represents the total payment that the passengers in the alliance need to share, and its expression is:

[0052] u(C)=d(R * (C))·(c f +c d +c v )

[0053] Among them, u(C) is the alliance cost, d(R * (C)) is the mileage of the alliance’s optimal path, c f 、c d with c v They represent fuel cost per unit mileage, driver cost and vehicle depreciation cost respectively.

[0054] The alliance optimal path R * (C) is the shortest path required by the vehicle to serve passengers in alliance C, which is defined as follows:

[0055] Define the road network as a graph M = (N, L), where N = {p σ ,p ω |p∈P} is the set of boarding and alighting points, p σ ,p ω They are the boarding and alighting points for all passengers. is the shortest path set, which means the shortest distance between all the places corresponding to the pick-up and drop-off points. The places corresponding to the pick-up and drop-off points are represented as R = (n1, n2, ..., n m ), n i ∈N, the vehicle will travel along the above point sequence as a path.

[0056] The effective path set of the alliance is The following constraints are met:

[0057]

[0058] R[k] represents the kth element in R, that is, the corresponding n k .

[0059] The above conditions constrain the vehicle's driving path to pass through the passenger's boarding and alighting points, and not pass through other points except the passenger's boarding and alighting points. In addition, it is also guaranteed to pass through the passenger's boarding point first and then the passenger's alighting point.

[0060] The optimal path of the alliance is the path that minimizes the total distance, that is, in where d R[k],R[k+1] Represents the distance between corresponding locations of adjacent point sequences.

[0061] The optimal alliance structure is the one that ensures that the personalized needs of passengers are mutually adapted and the total cost is minimized. The alliance structure with the minimum total cost is Among them, ∏(G) is the set of all alliance structures that meet the demand adaptation constraints; Alliance structure The cost of

[0062] The optimal alliance structure is obtained by combining a depth-first search algorithm with a branch-and-bound strategy. The process includes:

[0063] (1) Adapt the demand to each node p of the graph G = (P, E) i Convert to an alliance formed by individual passengers to form the initial alliance structure All edges e ij ∈E is marked as shrinkable. Initialize the search frontier stack and the initial cost as the current optimal value.

[0064] (2) Pop up the top element of the current search front stack Select a contractible edge e from ij , check the new alliance structure generated by performing edge contraction on this edge Whether the scale constraint and demand adaptation constraint are met. If not, mark the edge as non-shrinkable. Otherwise, perform edge shrinkage and mark the edge as non-shrinkable to avoid repeated exploration; calculate If it is less than the current optimal value, update the current optimal value to Will Push

[0065] The edge contraction is to: ij The two connected alliances C i with C j Merged into the new Alliance C ij , generating a new alliance structure

[0066] (3) For the newly generated alliance structure Arrange the alliances in ascending order of alliance cost, i.e. Where u(C1)≤u(C2≤...≤u(C k ).

[0067] calculate The lower bound like If the value is greater than the current optimal value, it is considered that the node and its child nodes cannot produce a better solution and are no longer pushed into the search frontier stack.

[0068] The payment results that meet the equilibrium conditions are calculated through surplus based on the optimal alliance structure.

[0069] The payment result is represented by the payment vector Indicates that x i For passengers i ∈C’s payment amount.

[0070] The payment vector satisfies

[0071] The equilibrium condition is where s ij is called surplus, which means passenger p i Leave alliance C and join alliance C which does not include passenger p j Other alliances The maximum additional cost savings that can be obtained when Surplus ij Reflects the passenger p i Compared to passenger p j Bargaining power, condition s ij ≥s ji This means that no two members of the alliance have enough power to ask the other party to pay more.

[0072] The equilibrium pricing algorithm has the following steps:

[0073] (1) Initialization. Averagely share the alliance cost, i.e.

[0074] (2) Calculate the maximum surplus difference. The maximum surplus difference represents the maximum value of the difference in surplus between any pair of passengers, that is,

[0075] (3) Convergence judgment: Determine whether the current payment scheme has reached equilibrium. The judgment condition is δ≤ε·u(C), where ε is the preset convergence threshold.

[0076] (4) Payment transfer. If the current payment scheme does not reach equilibrium, perform iterative update of the payment vector. Let the current maximum surplus difference be the passenger's (p i ,p j ) is generated, and payment transfer amount is calculated Update the payment vector according to the transfer amount d, let x i ←x i +d,x j ←x j -d, return to step (2).

[0077] The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of this application. It should be understood that the above description is only the specific implementation methods of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of this application should be included in the scope of protection of this application.

Claims

1. A carpooling optimization method considering demand adaptation and equilibrium pricing, characterized by: include: Obtain passenger information; Constructing a demand adaptation graph based on passenger information, wherein the passenger information includes binary feature demands, double binary feature demands, and quantitative feature demands. The binary feature demands represent the passenger's binary tendency for the first type of behavior; the double binary feature demands represent the passenger's binary tendency for the second type of behavior and the degree of acceptance of others' behavior; and the quantitative feature demands represent the passenger's acceptable range for the third type of behavior. Determine the adaptation results of different passengers' personalized needs based on the demand adaptation function, and form edges based on the adaptation results; form a demand adaptation graph based on passenger information and edges; forming an alliance structure based on the demand adaptation graph, and searching within the alliance structure to obtain an optimal alliance structure that satisfies the constraints, wherein the optimal alliance structure is an alliance structure in which the personalized needs of passengers are mutually adapted and the total cost of the alliance structure is minimized; The payment result that satisfies the equilibrium condition is calculated through surplus calculation based on the optimal alliance structure, including, if the current payment scheme has not reached the equilibrium state, calculating the payment transfer amount through the maximum surplus difference, recalculating the maximum surplus difference based on the payment transfer amount, and repeating the iteration until the current scheme reaches the equilibrium state.

2. The carpooling optimization method considering demand adaptation and balanced pricing according to claim 1 is characterized in that: The demand adaptation function is: Among them, f k (·) is the adaptability judgment operator of the kth demand feature, F(p i ,p j )=1 means passenger p i With passengers j The personalized needs of each other are adapted; F(p i ,p j )=0 means passenger p i With passengers j There are conflicts in individual needs.

3. The carpooling optimization method considering demand adaptation and balanced pricing according to claim 2 is characterized in that: The adaptability determination operator has different values ​​for different requirements, including: For binary feature requirements, the adaptability judgment operator satisfies: f k (α i ,α j )=1-|α i -α j |,α i ∈{0,1},α i =1 means passenger p i With this tendency, α i =0 means not available; For dual binary feature requirements, the adaptability judgment operator satisfies: f k (β i ,β j )=sgn(β i2 -β j1 )·sgn(β j2 -β i1 ), β i =(β i1 ,β i2 ), β i1 ∈{0,1} represents passenger p i Do you have this tendency? i2 ∈{0,1} represents passenger p i Whether to accept that others have this tendency; For quantitative feature requirements, the adaptability judgment operator satisfies: in in, 4. The carpooling optimization method considering demand adaptation and balanced pricing according to claim 2 is characterized in that: The requirement adaptation function satisfies: F(p i ,p i )=1 F(p i ,p j )=F(p j ,p i ) 5. The carpooling optimization method considering demand adaptation and balanced pricing according to claim 1 is characterized in that: The total cost of the alliance structure includes the sum of all alliance costs, and the alliance cost is expressed as u(C)=d(R * (C))·(c f +c d +c v ), where u(C) is the alliance cost, d(R * (C)) is the mileage of the alliance’s optimal path, c f 、c d with c v They represent fuel cost per unit mileage, driver cost and vehicle depreciation cost respectively.

6. The carpooling optimization method considering demand adaptation and balanced pricing according to claim 5 is characterized in that: The road network is a graph M = (N, L), where N is the set of boarding points and alighting points. is the shortest path set, which means the shortest distance between all the places corresponding to the pick-up and drop-off points. The places corresponding to the pick-up and drop-off points are represented by R = (n1, n2, ..., n m ), n i ∈N, the vehicle will travel along the above sequence of locations as a path, and the valid path set is satisfy: Among them, R[k] represents the kth element in R, that is, the corresponding n k .

7. The carpooling optimization method considering demand adaptation and balanced pricing according to claim 6, characterized in that: The alliance optimal path R * (C) is the shortest path required for the vehicle to serve passengers in alliance C, expressed as: where d R[k] ,R [k+1] Represents the distance between corresponding locations of adjacent point sequences.

8. The carpooling optimization method considering demand adaptation and balanced pricing according to claim 1, characterized in that: The searching in the alliance structure to obtain the optimal alliance structure that satisfies the constraints includes: Adapt the demand to each node p of the graph G = (P, E) i Convert to an alliance formed by individual passengers to form the initial alliance structure All edges e ij ∈E is marked as shrinkable, and the search frontier stack is initialized and the initial cost As the current optimal value; where each node p in the adaptation graph i ∈P represents a passenger if and only if passenger p i With p j When the personalized needs of ij ∈E, represents passenger p i With p j There is an opportunity to share a ride, and the alliance is a set of passengers arranged to ride in the same vehicle, which satisfies the size constraint 1≤C≤λ, where λ is the maximum number of passengers that the vehicle can carry; Pop the top element of the current search stack Select a contractible edge e from ij , check the new alliance structure generated by performing edge contraction on this edge Whether the scale constraint and demand adaptation constraint are met; if not, mark the edge as non-shrinkable; otherwise, perform edge shrinkage and mark the edge as non-shrinkable to avoid repeated exploration; calculate If it is less than the current optimal value, update the current optimal value to Will Push The edge contraction includes converting the edge e ij The two connected alliances C i with C j Merged into the new Alliance C ij , generating a new alliance structure For the newly generated alliance structure Arrange the alliances in ascending order of alliance cost, i.e. Where u(C1)≤u(C2≤...≤u(C k );calculate The lower bound like If the value is greater than the current optimal value, it is considered that the node and its child nodes cannot produce a better solution and are no longer pushed into the search frontier stack.

9. The carpooling optimization method considering demand adaptation and balanced pricing according to claim 1, characterized in that: The payment results that meet the equilibrium conditions through surplus calculation based on the optimal alliance structure include: The payment result is determined by the payment vector Indicates that x i For passengers i ∈C, the payment vector satisfies The equilibrium condition is where s ij is called surplus, which means passenger p i Leave alliance C and join alliance C which does not include passenger p j Other alliances The maximum additional cost savings that can be obtained when 10. The carpooling optimization method considering demand adaptation and balanced pricing according to claim 1, characterized in that: The payment result that satisfies the equilibrium condition through surplus calculation based on the optimal alliance structure also includes: Calculate equilibrium pricing based on the optimal alliance structure, including: Average allocation of alliance costs: Calculate the maximum surplus difference: The condition δ≤ε·u(C) is used to determine whether the current payment scheme has reached equilibrium, where ε is the preset convergence threshold; if the current payment scheme has not reached equilibrium, the current maximum surplus difference is determined by the passenger's (p i ,p j ) is generated, and payment transfer amount is calculated Update the payment vector according to the transfer amount d, let x i ←x i +d,x j ←x j -d, recalculate the maximum surplus difference and perform iterative updates of the payment vector in subsequent steps.

Images