A matching optimization method for online carpooling and subway intermodal transport

Abstract

The present invention provides a method for optimizing the matching of online carpooling and subway intermodal transport. The method includes: forming a set of effective intermodal transport paths based on passenger travel information; calculating the generalized travel cost of carpooling intermodal transport, searching for carpooling intermodal transport transfer stations with the lowest carpooling intermodal transport cost; dividing carpooling intermodal transport passengers into three intermodal subtypes of passengers: WSR, RSW and RSR; eliminating passengers of various types who do not meet the travel similarity constraints and travel cost saving constraints, respectively, to obtain a feasible matching set of carpooling passengers; matching carpooling passengers with the maximum motor vehicle road mileage saving as the optimization goal, establishing a carpooling passenger matching model, using an optimization solver to solve the carpooling passenger matching model, and obtaining the optimal matching result information of online carpooling and subway intermodal transport. The method of the present invention strengthens the synergy between the two modes of transportation, reduces the motor vehicle road mileage and increases the utilization rate of public transportation while meeting their travel needs.

Description

Technical Field

[0001] The present invention relates to the technical field of urban traffic operation management, and in particular to a method for optimizing the matching of online carpooling and subway combined transport. Background Art

[0002] In some areas of megacities, taxis account for a much higher proportion of traffic than subway systems, and subways are relatively less competitive. It can be seen that within megacities, the relationship between taxi services, including online ride-hailing services, and subways is no longer dominated by complementary cooperation. On the contrary, the competition between the two is significant.

[0003] In order to prevent this phenomenon from getting worse, improve the efficiency of multimodal transport of public transportation, maintain the current and future passenger flow of the subway network, and encourage taxi passengers to take the subway more often, it is particularly necessary. Some passengers' starting and ending points are both near subway stations, but these passengers still do not choose to travel by subway, indicating that these users have a certain dependence on cars. Based on the current subway service level, it is difficult to attract passengers to take the subway for the entire journey. Summary of the invention

[0004] An embodiment of the present invention provides a method for optimizing the matching of online carpooling and subway intermodal transport, so as to reduce the road mileage of motor vehicles, increase the utilization rate of public transportation and reduce environmental pollution.

[0005] In order to achieve the above object, the present invention adopts the following technical scheme.

[0006] A method for optimizing the matching of online carpooling and subway intermodal transport, comprising:

[0007] Based on the travel information of passengers using carpooling services, the shortest paths between all subway stations in the subway station set are searched according to the spatial proximity principle to form a set of effective intermodal paths.

[0008] Calculating the generalized travel cost of passenger carpooling and intermodal transport considering the time value, and searching for a carpooling and intermodal transport transfer station with the minimum carpooling and intermodal transport cost according to the generalized travel cost of passenger carpooling and intermodal transport and the set of effective intermodal transport paths;

[0009] According to the passengers' carpooling transfer stations, carpooling passengers are divided into three sub-types: walking-subway-online carpooling WSR, online carpooling-subway-walking RSW, and online carpooling-subway-online carpooling RSR.

[0010] Eliminate the passengers who do not meet the itinerary similarity constraints and travel cost saving constraints among the WSR, RSW and RSR type passengers, and obtain the feasible matching sets of WSR, RSW and RSR type carpooling passengers respectively;

[0011] Based on the feasible matching set of carpooling passengers, carpooling passengers are matched with the maximum motor vehicle road mileage saving as the optimization goal, a carpooling passenger matching model is established, and the carpooling passenger matching model is solved using an optimization solver to obtain the optimal matching result information of online carpooling and subway combined use.

[0012] Preferably, based on the travel information of passengers using the carpooling service, searching for the shortest paths between all subway stations in the subway station set according to the spatial proximity principle to form a valid intermodal path set P includes:

[0013] For passengers who use carpooling services, they publish their itinerary information on the carpooling platform in advance. The itinerary information includes the latitude and longitude information of the starting and ending points of the trip, the planned departure time of the passenger, and the planned arrival time of the passenger.

[0014] Based on the obtained passenger travel information and the principle of spatial proximity, a set of subway stations within a certain range of the starting and ending points of the travel is found, and the shortest paths between all subway stations in the subway station set are searched to form a set of effective intermodal transport paths.

[0015] Preferably, the calculation of the generalized travel cost of passenger carpooling taking into account the time value includes:

[0016] Calculate the cost from the starting point to the transfer station C(o,so), that is, the front-end connection cost. When the passenger belongs to the WSR type, the front-end connection cost C(o,so)=C w (o,so)=γT w (o,so), where o represents the starting point, so represents the entry station of the intermodal subway, and γ is the unit time value of walking; T w is the walking time. When the passenger is of RSW or RSR type, the front connection cost of the first passenger is:

[0017] C(o,so)=C rs (o1,so1)=μ(T rs (o1,o2)+T rs (o2,so1))+β(D rs (o1,o2)+D rs (o2,so1))

[0018] The forward connection cost of the second passenger is C(o,so)=C rs (o2,so2)=μT rs (o2,so2)+βD rs(o2, so2), where o1 and o2 represent the starting points of the first and second passengers respectively; so1 and so2 represent the entry points of the two passengers to the subway; T represents the time in the car; D represents the travel distance; μ represents the unit time value of the online carpooling; β represents the unit mileage cost;

[0019] Calculate the cost from the transfer station to the terminal, C(d, sd), which is the back-end connection cost. When the passenger belongs to the RSW type, the back-end connection cost C(d, sd) = C w (sd,d)=γT w (sd,d), where d is the end point, sd is the exit station of the intermodal subway, and γ is the unit time value of walking; T w is the walking time. When the passenger belongs to WSR or RSR type, the rear-end connection cost for the first passenger getting off is:

[0020] C(d,sd)=C rs (sd1,d1)=αT rs (sd1,d1)+βD rs (sd1,d1),

[0021] The cost of the second passenger getting off the bus is:

[0022] C(d,sd)=C rs (sd2,d2)=α(T rs (sd2,d1)+T rs (d1,d2))+β(D rs (sd2,d1)+D rs (d1,d2));

[0023] Where α and β represent the unit time value parameter and the unit mileage cost respectively; d1 and d2 represent the destinations of the first and second passengers who get off the bus respectively; sd1 and sd2 represent the exit stations where the two passengers connect to the subway respectively;

[0024] The travel cost of the subway section is calculated, which consists of three parts: travel time cost, subway fare and carriage congestion cost. The in-train time and subway fare considering congestion penalty are summed up to obtain the generalized travel cost C of the subway section. s (so, sd), by the travel time cost θT on the subway s (so, sd), the time cost of transferring in the subway εT change (so, sd) and subway fare F s (so, sd), converted into the expression:

[0025]

[0026] In the formula, so, sd represent the intermodal transfer entry and exit stations, T run,e represents the train running time on the subway network edge e; T dwell,v represents the train's stop time at subway station v; T change,e represents the time required for the virtual transfer edge e at the subway station; E so,sd represents the set of interval network edges on the path from so station to sd station; V so,sd represents the set of subway stations on the path from so station to sd station; represents the set of virtual transfer edges on the path from so station to sd station, f c (ω e ) represents the penalty coefficient for carriage congestion, θ represents the unit time cost of the subway in-train time; ε represents the time cost of passengers transferring in the subway; F s It represents the subway fare between so and sd stations;

[0027] Passengers walk from the ride-hailing drop-off point to the station or from the station to the ride-on point. The cost function C tr (o,d)=λT tr , where λ is the unit time value of walking, T tr is the walking time, take the walking time as a constant;

[0028] Calculate the time deviation cost based on the advance and delay of departure time, as well as the advance and delay of arrival time:

[0029] C td (o,d)=ηΔT o +ζΔT d , where ηΔT o =η + max{(t o -t' o ),0}+η - max{(t' o -t o ),0}

[0030] ζΔT d =ζ + max{(t d -t' d ),0}+ζ - max{(t' d -t d ),0}

[0031] Where ΔT o and ΔT d They represent departure time deviation and arrival time deviation respectively; t o and t dRespectively represent the planned departure and arrival time; t' o and t' d represents the actual departure and arrival time; parameter η + , η - , + and - The unit time costs of early departure, delayed departure, early arrival and delayed arrival are respectively. When the trip belongs to the WSR type, t' o =t o , t' d =t o +T w (o,so)+T s (so,sd)+T rs (sd,d), when the trip belongs to RSW or RSR, assuming that passengers i and j share a ride, pick up i first and meet his on-time departure, t' oi =t oi , t' di =t' oi +T rs (o i ,o j )+T rs (o j ,so i )+T s (so i ,sd i )+T w (sd i ,d i ), at this time t' oj =t' oi +T rs (o i ,o j ), t' dj =t' dj +T rs (o j ,so j )+T s (so j ,sd j )+T w (sd j ,d j );

[0032] The generalized travel cost C of carpooling considering time driving rss (o,d) is the sum of the above five parts, that is, C rss (o,d)=C(o,so)+C(d,sd)+C s (so,sd)+C tr (o,d)+C td(o, d), where so represents the entrance station of the intermodal subway; sd represents the exit station of the intermodal subway; the subscripts “rss”, “s”, “tr”, and “td” represent the intermodal transfer between online ride-hailing and subway, subway, intermodal transfer, and time deviation, respectively.

[0033] Preferably, the step of searching for a carpooling transfer station with the minimum carpooling cost based on the passenger carpooling generalized travel cost and the effective carpooling path set includes:

[0034] Step 4.1: For a given trip of a certain passenger, the shortest network paths between all subway stations within a certain range of the starting and ending points are taken as the effective intermodal paths for the trip. Assuming that the given trip is from the starting point o to the end point d, the process of calculating the effective intermodal paths for the given trip is as follows:

[0035] Step 1: Search for all subway stations in the neighborhood of the starting point o and the end point d to form the starting station set Meet at the final station

[0036] Step 2: Arrange the station set obtained in the first step to obtain a total of m×n station pairs.

[0037] Step 3: For each station pair in the previous step, use the Dijkstra algorithm to solve the shortest path on the subway network, taking into account the on-board time and transfer perception time, and obtain the effective intermodal path set P = {p1, p2...p m×n}, where the path P is saved as a sequence of sites passed through in sequence;

[0038] Step 4.2: Search for intermodal transfer stations. The basic principle of transfer station search is to minimize the generalized travel cost of passenger carpooling. Select a valid intermodal route, replace any two stations on the route with carpooling travel in sequence, update the generalized travel cost of passenger carpooling, compare it with the original taxi cost, and obtain the transfer station with the minimum carpooling cost. The original taxi cost is the cost required to travel by taxi with the same itinerary.

[0039] Preferably, the intermodal passengers are divided into three intermodal sub-types of passengers according to their intermodal transfer stations: walking-subway-online carpooling WSR, online carpooling-subway-walking RSW and online carpooling-subway-online carpooling RSR, including:

[0040] For any passenger trip, the carpooling and intermodal transfer search algorithm is used to determine the transfer stations one by one. According to the distance between the transfer station and the starting and ending points of each passenger, the carpooling and intermodal passengers are divided into three intermodal sub-type passengers: WSR, RSW and RSR, and the passenger set is obtained:

[0041] Ω RSW ={i|D(o i ,so i )>D threshold ,D(d i ,sd i )≤D threshold}

[0042] Ω WSR ={i|D(o i ,so i )≤D threshold ,D(d i ,sd i )>D threshold}

[0043] Ω RSR ={i|D(o i ,so i )>D threshold ,D(d i ,sd i )>D threshold}

[0044] In the formula, Ω RSW ,Ω WSR ,Ω RSR is a user subset of the three carpooling subtypes RSW, WSR and RSR. o, d, so and sd are the starting point, end point, front-end transfer station and back-end transfer station of the trip respectively. threshold is the threshold of the walkable distance;

[0045] For the WSR type, two passengers walk to the same subway station separately, take the subway to the transfer station, and after exiting the station, take the same online car-hailing vehicle to their respective destinations; for the RSW type, passengers first use online car-hailing vehicles to go to the same connecting subway station, and then take the subway to their respective destinations; for the RSR type, passengers first use online car-hailing vehicles to reach the connecting front subway station, take the subway to the connecting back subway station, and after exiting the station, they share a ride again to their respective trip destinations. Only passengers belonging to the same subcategory can be matched together for intermodal transport.

[0046] Preferably, the passengers of WSR, RSW and RSR types that do not meet the itinerary similarity constraint and the travel cost saving constraint are eliminated respectively, and feasible matching sets of WSR, RSW and RSR type carpooling passengers are obtained respectively, including:

[0047] For RSW and RSR passengers, calculate the absolute value of the difference between the planned departure times of shared passengers i and j. If the absolute value is greater than the threshold tthreshold , then the shared passengers i and j are judged as unconstrained passenger matching pairs and are eliminated; for WSR type passengers, the absolute value of the time difference between shared passengers i and shared passengers j arriving at the front station is calculated If the absolute value is greater than the threshold t threshold , then the shared passengers i and j are judged as a passenger matching pair that does not meet the constraint and are eliminated. After the above operation of eliminating passengers that do not meet the constraint, a feasible matching set of shared passengers that meets the itinerary similarity is obtained;

[0048] Based on the feasible matching set of carpool passengers that meet the itinerary similarity, the cost from the starting point to the transfer station, the cost from the transfer station to the destination, the travel cost of the subway section, the conversion cost between online carpooling and subway, and the deviation cost of departure / arrival time of carpooling passenger i and carpooling passenger j are calculated respectively, and the travel cost of the online carpooling and subway combined transport mode of carpooling passenger i is obtained. Get the travel cost of the online carpooling and subway combined mode for the shared passenger j If the travel cost Greater than the travel cost of a shared passenger i using a cruising taxi or online ride-hailing car alone or travel costs Greater than the travel cost of carpooling passenger j using a cruising taxi or online ride-hailing car alone Then, carpooling passengers i and carpooling passengers j are judged as a passenger matching pair that does not meet the travel cost saving constraint and are eliminated; after the above operation of eliminating unmatched carpooling passenger combinations, feasible matching sets of carpooling passengers of types WSR, RSW and RSR are obtained respectively.

[0049] Preferably, the matching of carpool passengers is performed based on the feasible matching set of carpool passengers, with the maximum motor vehicle road mileage saving as the optimization goal, a carpool passenger matching model is established, and the carpool passenger matching model is solved using an optimization solver to obtain the optimal matching result information of online carpooling and subway combined use, including:

[0050] Based on the feasible matching sets of carpool passengers of the WSR, RSW and RSR types, carpool passengers are matched with the maximum motor vehicle road mileage saving as the optimization goal, and a carpool passenger matching model is established. The construction method of the carpool passenger matching model is as follows:

[0051] Step 8.1: Shared passenger A is passenger i, shared passenger B is passenger j, subject to the acceptance order requirement, (i, j) ≠ (j, i), the feasible matching set Ψ of shared passengers is composed of the feasible sets Ψ of its three subtypes RSW, WSR, and RSR RSW , WSR , RSR Composition, that is, Ψ=ΨRSW ∪Ψ WSR ∪Ψ RSR ;

[0052] The feasible matching set of RSW type carpool passengers is:

[0053] The front-end transfer station so of passenger i and passenger j is the same, and the absolute value of the difference in planned departure time Less than the threshold t threshold , the travel cost of using online carpooling and subway combined transport mode C rss Should be less than the travel cost of using a cruising taxi or online car-hailing service alone C t / r , passenger i and passenger j are two different passengers, and for any passenger i and passenger j, they belong to the RSW subtype carpooling passenger set;

[0054] The feasible matching set of WSR type carpooling passengers is:

[0055] The rear transfer station sd of passenger i and passenger j is the same, and the absolute value of the time difference between arriving at the front station Less than the threshold t threshold , the remaining constraints are the same as those of the RSW type;

[0056] The feasible matching set of RSR type carpool passengers is:

[0057]

[0058] The front connecting transfer station so and the back connecting transfer station sd of passenger i and passenger j are the same, and the other constraints are the same as the RSW type;

[0059] Step 8.2: Taking the maximum motor vehicle road mileage saving as the optimization goal, establish the optimization equation, that is,

[0060] Max∑ (i,j)∈Ψ c ij x ij .

[0061] The constraints of the optimization equation include:

[0062] (1) Carpooling passenger matching constraint∑ i∈R x ij ≤1, ∑ j∈R' x ij ≤1, ∑ i∈R (x ij +x ji )≤1,

[0063] (2) Decision variable constraints x ij ={0,1}

[0064] In the above equation, Ψ is the feasible matching set, x ij is the decision-side variable constraint, the purpose of which is to ensure that each passenger in the feasible set can only be matched to one carpool combination. The decision variable constraint is a 0-1 Boolean value constraint on the matching decision variable, 0 represents a failed match, and 1 represents a successful match. R is the first passenger set to get on the bus in the feasible matching combination, and R' is the second passenger set to get on the bus in the feasible matching combination. The coefficient c ij The road mileage that can be reduced by the passenger combination is calculated as follows:

[0065] c ij =D t / r (o i ,d i )+D t / r (o j ,d j )-[D rs (o i ,o j )+min(D rs (o i ,so i ),D rs (o j ,so j )+min(D rs (d i ,sd i ),D rs (d j ,sd j ))+D rs (d i ,d j )]

[0066] Step 8.3: Model the online carpooling matching problem from passengers to drivers as a mixed integer programming problem, and use the optimization solver to solve the carpooling passenger matching model to ensure that each passenger in the feasible set can only be matched to one carpooling combination, and obtain the optimal online carpooling and subway combined matching result information, which includes the matching results of carpooling passengers, the nearest boarding and alighting points to the passengers, the online carpooling and subway transfer points, and the passengers' planned departure / arrival time information. The service platform sends the above matching result information to the driver and carpooling passengers.

[0067] It can be seen from the technical solutions provided by the above-mentioned embodiments of the present invention that the method for optimizing the matching of online carpooling and subway intermodal transport of the present invention improves the coordination and cooperation between the two modes of transportation while meeting the travel needs of taxi passengers, reduces the competition between the two, and can ensure the continuous stability of the current and future subway network passenger flow. It aims to encourage taxi users to take the subway more often, reduce taxi or online car-hailing trips, and reduce motor vehicle road mileage, increase the utilization rate of public transportation and reduce environmental pollution while meeting their travel needs.

[0068] Additional aspects and advantages of the present invention will be given in part in the following description, which will become obvious from the following description, or may be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0070] Figure 1 A flow chart of an optimization method for matching online carpooling with subway combined transport provided in an embodiment of the present invention.

[0071] Figure 2 A schematic diagram of a complete travel process for different carpooling itinerary types provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0072] The embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention.

[0073] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or coupling. The term "and / or" used herein includes any unit and all combinations of one or more associated listed items.

[0074] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless defined as herein.

[0075] To facilitate understanding of the embodiments of the present invention, several specific embodiments will be further explained below with reference to the accompanying drawings, and each embodiment does not constitute a limitation on the embodiments of the present invention.

[0076] The processing flow of a method for optimizing the matching of online carpooling and subway intermodal transport provided by an embodiment of the present invention is as follows: Figure 1 As shown, the method comprises the following steps:

[0077] Step S1: For passengers who use the carpooling service, the passengers publish their itinerary information in advance on the carpooling platform. The platform obtains the passenger's itinerary information, which includes the latitude and longitude information of the starting and ending points of the trip, the passenger's planned departure time, and the passenger's planned arrival time.

[0078] Step S2: Based on the acquired passenger travel information and according to the spatial proximity principle, a set of subway stations within a certain range of the starting and ending points of the travel is found, and the shortest paths between all subway stations in the subway station set are searched to form a set of effective intermodal transport paths.

[0079] Step S3: Calculate the generalized travel cost of passenger carpooling considering the time value, including the cost of carpooling passengers from the starting point to the transfer station, the cost from the transfer station to the destination, the travel cost of the subway section, the transfer cost between online carpooling and the subway, and the deviation cost of departure time and arrival time.

[0080] The method for calculating the generalized travel cost of passenger carpooling considering the time value is as follows:

[0081] Step 3.1 Calculate the cost from the starting point to the transfer station C(o,so), that is, the front-end connection cost, which varies according to the connection method. When the passenger belongs to the WSR type, the front-end connection cost C(o,so)=C w (o,so)=γT w (o,so), where o represents the starting point, so represents the entry station of the intermodal subway, and γ is the unit time value of walking; T w is the time required for walking. When the passenger is of RSW or RSR type, the front-end connection cost of the first passenger on the bus is

[0082] C(o,so)=C rs (o1,so1)=μ(T rs (o1,o2)+T rs (o2,so1))+β(D rs (o1,o2)+D rs (o2,so1)), the front-end connection cost of the second passenger is C(o,so)=C rs (o2,so2)=μT rs (o2,so2)+βD rs (o2,so2), where o1 and o2 represent the starting points of the first and second passengers respectively; so1 and so2 represent the entry points of the two passengers to the subway respectively; T represents the time on the bus; D represents the distance of the trip; μ represents the unit time value of the online carpooling; β represents the unit mileage cost.

[0083] Step 3.2 calculates the cost C(d, sd) from the transfer station to the terminal, i.e., the back-end connection cost, which varies according to the connection method. When the passenger belongs to the RSW type, the back-end connection cost C(d, sd) = C w (sd,d)=γT w (sd,d), where d is the end point, sd is the exit station of the intermodal subway, and γ is the unit time value of walking; T w The time required for walking. When the passenger belongs to WSR or RSR type, the rear-end connection cost of the first passenger getting off

[0084] C(d,sd)=C rs (sd1,d1)=αT rs (sd1,d1)+βD rs (sd1, d1), the rear-end connection cost of the second passenger getting off is C(d, sd) = C rs (sd2,d2)=α(T rs(sd2,d1)+T rs (d1,d2))+β(D rs (sd2,d1)+D rs (d1, d2)), where α and β represent the unit time value parameter and the unit mileage cost respectively, d1 and d2 represent the destinations of the first and second passengers getting off the bus respectively; sd1 and sd2 represent the exit stations where the two passengers connect to the subway respectively.

[0085] Step 3.3 calculates the travel cost of the subway section, which consists of three parts: travel time cost, subway fare and carriage congestion cost. Sum up the time on the train and subway fare considering congestion penalty to obtain the generalized travel cost C of the subway section. s (so, sd), by the travel time cost θT on the subway s (so, sd), the time cost of transferring in the subway εT change (so, sd) and subway fare F s (so, sd). It can be converted into the expression

[0086] In the formula, so, sd represent the intermodal transfer entry and exit stations, T run,e represents the train running time on the subway network edge e; T dwell,v represents the train's stop time at subway station v; T change,e represents the time required for the virtual transfer edge e at the subway station; E so,sd represents the set of interval network edges on the path from so station to sd station; V so,sd represents the set of subway stations on the path from so station to sd station; represents the set of virtual transfer edges on the path from so station to sd station, f c (ω e ) represents the penalty coefficient for carriage congestion, θ represents the unit time cost of the subway in-train time; ε represents the time cost of passengers transferring in the subway; F s Indicates the subway fare between so and sd stations.

[0087] Step 3.4 Passengers walk from the ride-hailing drop-off point to the station or from the station to the ride-on point. The cost function C tr (o,d)=λT tr , where λ is the unit time value of walking, T tr is the walking time. Due to the flexibility of online ride-hailing vehicles, they will try to stop at the nearest entrance to the subway station, so they will not walk too far. Here, the walking time is taken as a constant, T tr =3 minutes.

[0088] Step 3.5 Calculate the time deviation cost C based on the advance and delay of departure time and the advance and delay of arrival time. td (o,d)=ηΔT o +ζΔT d , where ηΔT o =η + max{(t o -t' o ),0}+η - max{(t' o -t o ),0},ζΔT d =ζ + max{(t d -t' d ),0}+ζ - max{(t' d -t d ),0}. Where ΔT o and ΔT d They represent departure time deviation and arrival time deviation respectively; t o and t d Respectively represent the planned departure and arrival time; t' o and t' d represents the actual departure and arrival time; parameter η + , η - , + and - They correspond to the unit time costs of early departure, delayed departure, early arrival and delayed arrival respectively. When the trip belongs to the WSR type, t' o =t o , t' d =t o +T w (o,so)+T s (so,sd)+T rs (sd,d). When the trip belongs to RSW or RSR, assuming that passengers i and j share a ride, pick up i first and meet his on-time departure requirement, t' oi =t oi , t' di =t' oi +T rs (o i ,o j )+T rs (o j ,so i )+T s (so i ,sd i )+T w (sd i ,d i). At this time t' oj =t' oi +T rs (o i ,o j ), t' dj =t' dj +T rs (o j ,so j )+T s (so j ,sd j )+T w (sd j ,d j ).

[0089] Step 3.6 Considering the generalized travel cost C of carpooling with time driving rss (o,d) is the sum of the above five parts, that is, C rss (o,d)=C(o,so)+C(d,sd)+C s (so,sd)+C tr (o,d)+C td (o, d). In the formula, so represents the entrance station of the intermodal subway; sd represents the exit station of the intermodal subway; the subscripts "rss", "s", "tr", and "td" represent the intermodal transfer between online ride-hailing and subway, subway, intermodal transfer, and time deviation, respectively.

[0090] Step S4: searching for a transfer station with the minimum carpooling and intermodal transport cost through a carpooling and intermodal transport search algorithm according to the passenger carpooling and intermodal transport generalized travel cost and the effective intermodal transport path set, and determining the carpooling and intermodal transport transfer station.

[0091] The method for searching for a carpooling transfer station with the lowest carpooling cost in an embodiment of the present invention is as follows:

[0092] Step 4.1 First, according to the principle of spatial proximity, find the set of subway stations within a certain range of the starting and ending points of the trip, search for the shortest path between all subway stations in the set, and form a set of effective intermodal paths. Define the effective intermodal path as: for a given trip, the shortest path of all networks between all subway stations within a certain range of the starting and ending points is the effective intermodal path of the trip. Given a trip from starting point o to end point d, the process of calculating the effective intermodal path of the trip is as follows:

[0093] Step 1: Search for all subway stations in the neighborhood of the starting point o and the end point d to form the starting station set Meet at the final station

[0094] Step 2: Arrange the station set obtained in the first step to obtain a total of m×n station pairs.

[0095] Step 3: For each station pair in the previous step, use the Dijkstra algorithm to solve the shortest path on the subway network, taking into account the time on the train and the transfer perception time. Finally, the effective intermodal path set P = {p1, p2...p m×n}, where the path P is saved as a sequence of sites passed through in sequence.

[0096] Step 4.2 searches for intermodal transfer stations. The basic principle of searching for transfer stations is to minimize the generalized cost of carpooling. The core idea is to replace the path segment by segment. The basic idea is: select a valid intermodal path, replace any two stations on the path with carpooling in turn, update the generalized travel cost, compare it with the original taxi cost, and finally get the transfer station with the minimum carpooling cost.

[0097] The search algorithm for the above-mentioned carpooling transfer station is shown in the following table:

[0098]

[0099]

[0100] Step S5: According to the passengers' transfer stations, the intermodal passengers are divided into three intermodal sub-types of passengers: Walk-to-Subway-to-Ridesplitting (WSR), Ridesplitting-to-Subway-to-Walk (RSW), and Ridesplitting-to-Subway-to-Ridesplitting (RSR).

[0101] For any passenger trip, the carpooling and intermodal transfer search algorithm is used to determine the transfer stations one by one. After the transfer stations are determined, the carpooling and intermodal passengers are divided into three intermodal sub-types of passengers: WSR, RSW and RSR according to the distance between the transfer stations and the starting and ending points, and the passenger set is obtained:

[0102] Ω RSW ={i|D(o i ,so i )>D threshold ,D(d i ,sd i )≤D threshold}

[0103] Ω WSR ={i|D(o i ,so i )≤Dthreshold ,D(d i ,sd i )>D threshold}

[0104] Ω RSR ={i|D(o i ,so i )>D threshold ,D(d i ,sd i )>D threshold}.

[0105] In the formula, Ω RSW ,Ω WSR ,Ω RSR is a user subset of the three carpooling subtypes (RSW, WSR, and RSR), o, d, so, and sd are the starting point, end point, front-end transfer station, and back-end transfer station of the trip, respectively. threshold is the walking distance threshold.

[0106] Only passengers belonging to the same sub-category can be matched together for interline travel.

[0107] The embodiment of the present invention provides a complete travel process of different types of carpooling itineraries. Figure 2 As shown. For the WSR sub-mode, two passengers walk to the same subway station separately, take the subway to the transfer station, and after exiting the station, take the same online car-hailing vehicle to their respective destinations. For the RSW sub-mode, its travel process is the opposite of WSR. Passengers first use online car-hailing vehicles to go to the same connecting subway station, and then take the subway to their respective destinations. For the RSR sub-mode, passengers first use online car-hailing vehicles to reach the connecting front-end subway station, take the subway to the connecting back-end subway station, and after exiting the station, share a ride again to their respective trip destinations. In the RSR sub-mode, the online car-hailing vehicles for the connecting front-end and connecting back-end services are not required to be the same vehicle.

[0108] Step S6: Based on the travel characteristics of online carpooling and subway intermodal transportation, the carpooling platform defines trips with departure time differences within a certain threshold range and with the same intermodal transfer stations as similar trips. Through trip similarity pruning, all passengers who meet the conditions are fully arranged to obtain a feasible matching set of carpooling passengers that meet the trip similarity.

[0109] It is determined whether WSR, RSW and RSR type passengers meet the itinerary similarity constraints respectively, and the passengers who do not meet the constraint passenger matching pairs are eliminated to obtain a feasible matching set of carpooling passengers that meet the itinerary similarity.

[0110] For RSW and RSR passengers, calculate the absolute value of the difference between the planned departure times of shared passengers i and j. If the absolute value is greater than the threshold t threshold , it is not listed as a matching object and is removed.

[0111] For WSR type passengers, calculate the absolute value of the time difference between the shared passengers i and j arriving at the front station If the absolute value is greater than the threshold t threshold , it is not listed as a matching object and is removed.

[0112] After the above operation of eliminating passengers that do not meet the constraint passenger matching pairs, a feasible matching set of carpooling passengers that meet the itinerary similarity is obtained.

[0113] Step S7: The carpooling platform aims to make the travel cost of carpooling lower than the original taxi or online car-hailing cost, and screens passenger combinations that meet the cost requirements.

[0114] It is determined whether the carpooling passengers in the feasible matching set of carpooling passengers that meet itinerary similarity meet the travel cost saving constraint, and the carpooling passenger matching pairs that do not meet the constraint are eliminated to further obtain a feasible matching set of carpooling passengers that meet the travel cost saving.

[0115] Based on the feasible matching set of carpool passengers that meet the trip similarity, the cost from the starting point to the transfer station, the cost from the transfer station to the destination, the travel cost of the subway section, the transfer cost between online carpooling and subway, and the deviation cost of departure / arrival time are calculated for carpooling passenger i and carpooling passenger j respectively. The travel cost of the online carpooling and subway combined mode of carpooling passenger i is obtained. Get the travel cost of the online carpooling and subway combined mode for the shared passenger j If the travel cost Greater than the travel cost of a shared passenger i using a cruising taxi or online ride-hailing car alone or travel costs Greater than the travel cost of carpooling passenger j using a cruising taxi or online ride-hailing car alone Then, the carpooling passengers i and j are judged as a passenger matching pair that does not meet the travel cost saving constraint and are eliminated.

[0116] After the above operation of eliminating unmatched carpooling passenger combinations, a feasible matching set of carpooling passengers that satisfies travel cost savings is obtained.

[0117] Step S8: Based on the feasible matching set of carpool passengers that meet the travel cost saving requirement, carpool passengers are matched with the maximum motor vehicle road mileage saving as the optimization goal, and the system role constraints in the matching process are considered to ensure that each passenger can only be matched to one carpool combination.

[0118] Based on the above feasible matching set of carpool passengers that can save travel costs, carpool passengers are matched with the maximum motor vehicle road mileage saving as the optimization goal, and a carpool passenger matching model is established.

[0119] The construction method of the above-mentioned carpooling passenger matching model is as follows:

[0120] Step 8.1: Establish a feasible matching set of carpool passengers. Carpool passenger A is passenger i, carpool passenger B is passenger j, and subject to the acceptance order requirement, (i, j) ≠ (j, i). The final feasible matching set Ψ is composed of the feasible sets Ψ of its three subtypes RSW, WSR, and RSR RSW , WSR , RSR Composition, that is, Ψ=Ψ RSW ∪Ψ WSR ∪Ψ RSR .

[0121] The feasible set of RSW type is:

[0122] The front-end transfer station so of passenger i and passenger j is the same, and the absolute value of the difference in planned departure time Less than the threshold t threshold , the travel cost of using online carpooling and subway combined transport mode C rss Should be less than the travel cost of using a cruising taxi or online car-hailing service alone C t / r , passenger i and passenger j are two different passengers, and for any passenger i and passenger j, they belong to the RSW subtype carpooling passenger set.

[0123] The feasible set of WSR type is:

[0124] The rear transfer station sd of passenger i and passenger j is the same, and the absolute value of the time difference between arriving at the front station Less than the threshold t threshold , and the remaining constraints are the same as those of the RSW type.

[0125] The feasible set of RSR type is:

[0126]

[0127] The front connecting transfer station so and the rear connecting transfer station sd of passenger i and passenger j are the same, and the other constraints are the same as the RSW type.

[0128] Step 8.2: Construction of optimization equation. Taking the maximum motor vehicle road mileage saving as the optimization goal, establish the optimization equation, namely Max∑ (i,j)∈Ψ c ij x ij.

[0129] The constraints include (1) passenger matching constraint ∑ i∈R x ij ≤1, ∑ j∈R' x ij ≤1, ∑ i∈R (x ij +x ji )≤1,

[0130] (2) Decision variable constraints x ij ={0,1}

[0131] In the above equation, Ψ is the feasible matching set, x ij is the decision-side variable constraint, the purpose of which is to ensure that each passenger in the feasible set can only be matched to one carpool combination; the decision variable constraint is a 0-1 Boolean value constraint on the matching decision variable, 0 represents a failed match, and 1 represents a successful match. R is the first passenger set to board the bus in the feasible matching combination, and R' is the second passenger set to board the bus in the feasible matching combination. Coefficient c ij The road mileage that can be reduced by the passenger combination is calculated as follows:

[0132] c ij =D t / r (o i ,d i )+D t / r (o j ,d j )-[D rs (o i ,o j )+min(D rs (o i ,so i ),D rs (o j ,so j )+min(D rs (d i ,sd i ),D rs (d j ,sd j ))+D rs (d i ,d j )]

[0133] Step 8.3: Solve the above-mentioned carpooling passenger matching model using an optimization solver. Through step S8.2, the online carpooling matching problem from passengers to drivers is modeled as a mixed integer programming problem. The carpooling passenger matching model is solved using the branch-and-bound method with the help of optimization solver software (such as IBM ILOG CPLEX), ensuring that each passenger in the feasible set can only be matched to one carpooling combination, and obtaining the optimal online carpooling and subway combined operation matching result information, which includes the matching results of carpooling passengers, the nearest boarding and alighting points to the passengers, the online carpooling and subway transfer points, and the passengers' planned departure / arrival time information. The service platform sends the above-mentioned matching result information to the driver and the carpooling passenger.

[0134] According to the principles of intermodal travel services, carpooling services pick up passengers from the vehicle to the passenger's starting point, realizing door-to-door service. Therefore, the driver needs to arrive at the boarding point at the time specified by the passenger and wait for the passenger to board the vehicle. The "from far to near" principle is adopted when picking up passengers. When the passenger arrives at the drop-off point, he pays the fare online through the carpooling platform to complete a carpooling trip. The "from near to far" principle is adopted when sending passengers. Depending on the type of passenger, the drop-off point is the end point of the intermodal trip (WSR, RSR), the online carpooling and subway transfer point (RSW, RSR).

[0135] In summary, the embodiment of the present invention considers the user characteristics of taxi passengers and proposes a carpooling and subway travel service model and matching model for such passengers. Based on the passenger mode selection, taking into account factors such as the travel time value of taxi passengers, subway congestion penalty, and carpooling order, a carpooling and intermodal transfer station search algorithm with the goal of minimizing the generalized travel cost is proposed; on this basis, a carpooling and intermodal matching optimization model under the door-to-door service mode is constructed; the matching model is applied to actual cases, and the social benefits of the carpooling and intermodal subway service model are quantified from multiple perspectives.

[0136] The method for optimizing the matching of online carpooling and subway intermodal transport provided in the embodiment of the present invention is based on the principle of public transportation priority, and proposes a matching mode of online carpooling and subway intermodal transport, which can prevent the competition between online carpooling and subways, while improving the efficiency of multi-modal transport of public transportation, maintaining the continuous stability of current and future subway network passenger flow, and appropriately increasing subway trips while improving the utilization rate of empty seats in cars, so that passengers can gradually transfer to public transportation.

[0137] The present invention proposes a matching optimization method for online-booked carpooling and intermodal subway transportation, which improves the utilization rate of empty seats in cars and appropriately increases subway trips, so that passengers can gradually transfer to public transportation.

[0138] Those skilled in the art can understand that the accompanying drawings are only schematic diagrams of an embodiment, and the modules or processes in the accompanying drawings are not necessarily required to implement the present invention.

[0139] It can be known from the description of the above implementation methods that those skilled in the art can clearly understand that the present invention can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present invention or certain parts of the embodiments.

[0140] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The device and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0141] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for optimizing the matching of online carpooling and subway intermodal transport, characterized in that: include: Based on the travel information of passengers using carpooling services, the shortest paths between all subway stations in the subway station set are searched according to the spatial proximity principle to form a set of effective intermodal paths. Calculating the generalized travel cost of passenger carpooling and intermodal transport considering the time value, and searching for a carpooling and intermodal transport transfer station with the minimum carpooling and intermodal transport cost according to the generalized travel cost of passenger carpooling and intermodal transport and the set of effective intermodal transport paths; According to the passengers' carpooling transfer stations, carpooling passengers are divided into three sub-types: walking-subway-online carpooling WSR, online carpooling-subway-walking RSW, and online carpooling-subway-online carpooling RSR. Eliminate the passengers who do not meet the itinerary similarity constraints and travel cost saving constraints among the WSR, RSW and RSR type passengers, and obtain the feasible matching sets of WSR, RSW and RSR type carpooling passengers respectively; Based on the feasible matching set of carpooling passengers, carpooling passengers are matched with the maximum motor vehicle road mileage saving as the optimization goal, a carpooling passenger matching model is established, and the carpooling passenger matching model is solved using an optimization solver to obtain the optimal matching result information of online carpooling and subway combined use; Based on the travel information of passengers using the shared transport service, the shortest paths between all subway stations in the subway station set are searched according to the spatial proximity principle to form a valid intermodal transport path set P, including: For passengers who use carpooling services, they publish their itinerary information on the carpooling platform in advance. The itinerary information includes the latitude and longitude information of the starting and ending points of the trip, the planned departure time of the passenger, and the planned arrival time of the passenger. Based on the obtained passenger travel information and the principle of spatial proximity, a set of subway stations within a certain range of the starting and ending points of the travel is found, and the shortest paths between all subway stations in the subway station set are searched to form a set of effective intermodal routes; The calculation of the generalized travel cost of passenger carpooling taking into account the time value includes: Calculate the cost from the starting point to the transfer station C(o,so), that is, the front-end connection cost. When the passenger belongs to the WSR type, the front-end connection cost C(o,so)=C w (o,so)=γT w (o,so), where o represents the starting point, so represents the entry station of the intermodal subway, and γ is the unit time value of walking; T w is the walking time. When the passenger is of RSW or RSR type, the front connection cost of the first passenger is: C(o,so)=C rs (o1,so1)=μ(T rs (o1,o2)+T rs (o2,so1))+β(D rs (o1,o2)+D rs (o2,so1)) The forward connection cost of the second passenger is C(o,so)=C rs (o2,so2)=μT rs (o2,so2)+βD rs (o2, so2), where o1 and o2 represent the starting points of the first and second passengers respectively; so1 and so2 represent the entry points of the two passengers to the subway; T represents the time in the car; D represents the travel distance; μ represents the unit time value of the online carpooling; β represents the unit mileage cost; Calculate the cost from the transfer station to the terminal, C(d, sd), which is the back-end connection cost. When the passenger belongs to the RSW type, the back-end connection cost C(d, sd) = C w (sd,d)=γT w (sd,d), where d is the end point, sd is the exit station of the intermodal subway, and γ is the unit time value of walking; T w is the walking time. When the passenger belongs to WSR or RSR type, the rear-end connection cost for the first passenger getting off is: C(d,sd)=C rs (sd1,d1)=αT rs (sd1,d1)+βD rs (sd1,d1), The cost of the second passenger getting off the bus is: C(d,sd)=C rs (sd2,d2)=α(T rs (sd2,d1)+T rs (d1,d2))+β(D rs (sd2,d1)+D rs (d1,d2)); Where α and β represent the unit time value parameter and the unit mileage cost respectively; d1 and d2 represent the destinations of the first and second passengers who get off the bus respectively; sd1 and sd2 represent the exit stations where the two passengers connect to the subway respectively; The travel cost of the subway section is calculated, which consists of three parts: travel time cost, subway fare and carriage congestion cost. The in-train time and subway fare considering congestion penalty are summed up to obtain the generalized travel cost C of the subway section. s (so, sd), by the travel time cost θT on the subway s (so, sd), the time cost of transferring in the subway εT change (so, sd) and subway fare F s (so, sd), converted into the expression: In the formula, so, sd represent the intermodal transfer entry and exit stations, T run,e represents the train running time on the subway network edge e; T dwell,v represents the train's stop time at subway station v; T change,e represents the time required for the virtual transfer edge e at the subway station; E so,sd represents the set of interval network edges on the path from so station to sd station; V so,sd represents the set of subway stations on the path from so station to sd station; represents the set of virtual transfer edges on the path from so station to sd station, f c (ω e ) represents the penalty coefficient for carriage congestion, θ represents the unit time cost of the subway in-train time; ε represents the time cost of passengers transferring in the subway; F s It represents the subway fare between so and sd stations; Passengers walk from the ride-hailing drop-off point to the station or from the station to the ride-on point. The cost function C tr (o,d)=λT tr , where λ is the unit time value of walking, T tr is the walking time, take the walking time as a constant; Calculate the time deviation cost based on the advance and delay of departure time, as well as the advance and delay of arrival time: C td (o,d)=ηΔT o +ζΔT d , among which ΔT o =the + max{(t o -t' o ),0}+η - max{(t' o -t o ),0} zDT d =ζ + max{(t d -t' d ),0}+ζ - max{(t' d -t d ),0} Where ΔT o and ΔT d They represent departure time deviation and arrival time deviation respectively; t o and t d Respectively represent the planned departure and arrival time; t' o and t' d represents the actual departure and arrival time; parameter η + , η - , + and - The unit time costs of early departure, delayed departure, early arrival and delayed arrival are respectively. When the trip belongs to the WSR type, t' o =t o , t' d =t o +T w (o,so)+T s (so,sd)+T rs (sd,d), when the trip belongs to RSW or RSR, assuming that passengers i and j share a ride, pick up i first and meet his on-time departure, t' oi =t oi , t' di =t' oi +T rs (o i ,o j )+T rs (o j ,so i )+T s (so i ,sd i )+T w (sd i ,d i ), at this time t' oj =t' oi +T rs (o i ,o j ), t' dj =t' dj +T rs (o j ,so j )+T s (so j ,sd j )+T w (sd j ,d j ); The generalized travel cost C of carpooling considering time driving rss (o,d) is the sum of the above five parts, that is, C rss (o,d)=C(o,so)+C(d,sd)+C s (so,sd)+C tr (o,d)+C td (o, d), where so represents the entrance station of the intermodal subway; sd represents the exit station of the intermodal subway; the subscripts "rss", "s", "tr", and "td" represent the intermodal transfer between online car-hailing and subway, subway, intermodal transfer, and time deviation, respectively; The step of searching for a carpooling transfer station with the minimum carpooling cost based on the passenger carpooling generalized travel cost and the effective carpooling path set includes: Step 4.1: For a given trip of a certain passenger, the shortest network paths between all subway stations within a certain range of the starting and ending points are taken as the effective intermodal paths for the trip. Assuming that the given trip is from the starting point o to the end point d, the process of calculating the effective intermodal paths for the given trip is as follows: Step 1: Search for all subway stations in the neighborhood of the starting point o and the end point d to form the starting station set Meet at the final station Step 2: Arrange the station set obtained in the first step to obtain a total of m×n station pairs. Step 3: For each station pair in the previous step, use the Dijkstra algorithm to solve the shortest path on the subway network, taking into account the on-board time and transfer perception time, and obtain the effective intermodal path set P = {p1, p2...p m×n }, where the path P is saved as a sequence of sites passed through in sequence; Step 4.2: Search for intermodal transfer stations. The basic principle of searching for transfer stations is to minimize the generalized travel cost of passenger carpooling. Select a valid intermodal route, replace any two stations on the route with carpooling travel in sequence, update the generalized travel cost of passenger carpooling, compare it with the original taxi cost, and obtain the transfer station with the minimum carpooling cost. The original taxi cost is the cost required to take a taxi without changing the itinerary. The passengers are divided into three sub-types of passengers according to their carpooling transfer stations: walking-subway-online carpooling WSR, online carpooling-subway-walking RSW and online carpooling-subway-online carpooling RSR, including: For any passenger trip, the carpooling and intermodal transfer search algorithm is used to determine the transfer stations one by one. According to the distance between the transfer station and the starting and ending points of each passenger, the carpooling and intermodal passengers are divided into three intermodal sub-type passengers: WSR, RSW and RSR, and the passenger set is obtained: Ω RSW ={i|D(o i ,so i )>D threshold ,D(d i ,sd i )≤D threshold } Ω WSR ={i|D(o i ,so i )≤D threshold ,D(d i ,sd i )>D threshold } Ω RSR ={i|D(o i ,so i )>D threshold ,D(d i ,sd i )>D threshold } In the formula, Ω RSW ,Ω WSR ,Ω RSR is a user subset of the three carpooling subtypes RSW, WSR and RSR. o, d, so and sd are the starting point, end point, front-end transfer station and back-end transfer station of the trip respectively. threshold is the threshold of the walkable distance; For the WSR type, two passengers walk to the same subway station, take the subway to the transfer station, and after exiting the station, take the same online car-hailing vehicle to their respective destinations; for the RSW type, passengers first use online car-hailing vehicles to go to the same connecting subway station, and then take the subway to their respective destinations; for the RSR type, passengers first use online car-hailing vehicles to reach the connecting front subway station, take the subway to the connecting back subway station, and after exiting the station, share a ride again to their respective trip destinations. Only passengers belonging to the same subcategory can be matched together for intermodal transportation; The passengers of WSR, RSW and RSR types that do not meet the itinerary similarity constraint and the travel cost saving constraint are eliminated respectively, and feasible matching sets of WSR, RSW and RSR type carpooling passengers are obtained respectively, including: For RSW and RSR passengers, calculate the absolute value of the difference between the planned departure times of shared passengers i and j. If the absolute value is greater than the threshold t threshold , then the shared passengers i and j are judged as unconstrained passenger matching pairs and are eliminated; for WSR type passengers, the absolute value of the time difference between shared passengers i and shared passengers j arriving at the front station is calculated If the absolute value is greater than the threshold t threshold , then the shared passengers i and j are judged as a passenger matching pair that does not meet the constraint and are eliminated. After the above operation of eliminating passengers that do not meet the constraint, a feasible matching set of shared passengers that meets the itinerary similarity is obtained; Based on the feasible matching set of carpool passengers that meet the itinerary similarity, the cost from the starting point to the transfer station, the cost from the transfer station to the destination, the travel cost of the subway section, the conversion cost between online carpooling and subway, and the deviation cost of departure / arrival time of carpooling passenger i and carpooling passenger j are calculated respectively, and the travel cost of the online carpooling and subway combined transport mode of carpooling passenger i is obtained. Get the travel cost of the online carpooling and subway combined mode for the shared passenger j If travel costs Greater than the travel cost of a shared passenger i using a cruising taxi or online ride-hailing car alone or travel costs Greater than the travel cost of carpooling passenger j using a cruising taxi or online ride-hailing car alone Then, it is judged that the carpooling passengers i and carpooling passengers j are a passenger matching pair that does not meet the travel cost saving constraint and are eliminated; after the above operation of eliminating the unmatched carpooling passenger combinations, the feasible matching sets of carpooling passengers of types WSR, RSW and RSR are obtained respectively; The method of matching the shared passengers based on the feasible matching set of shared passengers with the maximum motor vehicle road mileage saving as the optimization goal, establishing a shared passenger matching model, solving the shared passenger matching model using an optimization solver, and obtaining the optimal matching result information of online carpooling and subway combined use includes: Based on the feasible matching sets of carpool passengers of the WSR, RSW and RSR types, carpool passengers are matched with the maximum motor vehicle road mileage saving as the optimization goal, and a carpool passenger matching model is established. The construction method of the carpool passenger matching model is as follows: Step 7.1: Shared passenger A is passenger i, shared passenger B is passenger j, subject to the acceptance order requirement, (i,j)≠(j,i), the feasible matching set Ψ of the shared passengers is composed of the feasible sets Ψ of its three subtypes RSW, WSR, and RSR RSW , WSR , RSR Composition, that is, Ψ=Ψ RSW ∪Ψ WSR ∪Ψ RSR ; The feasible matching set of RSW type carpool passengers is: The front-end transfer station so of passenger i and passenger j is the same, and the absolute value of the difference in planned departure time Less than the threshold t threshold , the travel cost of using online carpooling and subway combined transport mode C rss Should be less than the travel cost of using a cruising taxi or online car-hailing service alone C t / r , passenger i and passenger j are two different passengers, and for any passenger i and passenger j, they belong to the RSW subtype carpooling passenger set; The feasible matching set of WSR type carpooling passengers is: The rear transfer station sd of passenger i and passenger j is the same, and the absolute value of the time difference between arriving at the front station Less than the threshold t threshold , the remaining constraints are the same as those of the RSW type; The feasible matching set of RSR type carpool passengers is: The front connecting transfer station so and the back connecting transfer station sd of passenger i and passenger j are the same, and the other constraints are the same as the RSW type; Step 7.2: Taking the maximum motor vehicle road mileage saving as the optimization goal, establish the optimization equation, namely Max∑ (i,j)∈Ψ c ij x ij ; The constraints of the optimization equation include: (1) Carpooling Passenger Matching Constraints (2) Decision variable constraints In the above equation, Ψ is the feasible matching set, x ij is the decision-side variable constraint, the purpose of which is to ensure that each passenger in the feasible set can only be matched to one carpool combination. The decision variable constraint is a 0-1 Boolean value constraint on the matching decision variable, 0 represents a failed match, and 1 represents a successful match. R is the first passenger set to get on the bus in the feasible matching combination, and R' is the second passenger set to get on the bus in the feasible matching combination. The coefficient c ij The road mileage that can be reduced by the passenger combination is calculated as follows: c ij =D t / r (o i ,d i )+D t / r (o j ,d j )-[D rs (o i ,o j )+min(D rs (o i ,so i ),D rs (o j ,so j )+min(D rs (d i ,sd i ),D rs (d j ,sd j ))+D rs (d i ,d j )] Step 7.3: Model the online carpooling matching problem from passengers to drivers as a mixed integer programming problem, and use the optimization solver to solve the carpooling passenger matching model to ensure that each passenger in the feasible set can only be matched to one carpooling combination, and obtain the optimal online carpooling and subway combined matching result information, which includes the matching results of carpooling passengers, the nearest boarding and alighting points to the passengers, the online carpooling and subway transfer points, and the passengers' planned departure / arrival time information. The service platform sends the above matching result information to the driver and carpooling passengers.

Images