A method and device for recommending personalized travel plans for the entire air-rail intermodal journey

Abstract

The present invention discloses a method and device for recommending personalized travel plans for the entire process of air-rail intermodal travel. The method comprises: sending a dynamic questionnaire to a user terminal unit, generating subsequent questions based on the selection results returned by the user terminal unit and obtaining preferences; obtaining a travel plan query request sent by the user terminal unit and generating a set of feasible travel plans; scoring the travel plans in the set of feasible travel plans based on preferences and travel needs; and sorting the plans from high to low according to the travel plan scores and sending them to the user terminal unit. The method provided by the present invention can provide personalized travel plans for the entire process of intercity air-rail intermodal travel based on the stated preferences of the user of the user terminal unit, thereby improving the accuracy of plan recommendations and the user experience of the user of the user terminal unit using intermodal travel.

Description

Technical Field

[0001] The present invention belongs to the technical field of information recommendation, and in particular relates to a method and device for recommending personalized travel plans for the entire process of air-rail intermodal travel. Background Art

[0002] Passenger intermodal transport can coordinate and integrate passenger journeys across different modes of transportation, leveraging the comparative advantages of each mode and promoting the balanced and efficient operation of the transportation system. Intermodal travel, exemplified by air-rail intermodal travel, is characterized by complexity, numerous considerations, and a low degree of personalized travel plans. Existing technologies primarily focus on recommending intercity travel plans, neglecting the intra-city portion of intercity travel and failing to provide comprehensive recommendations for the entire journey. Furthermore, in the area of ​​personalized travel recommendations, existing technologies primarily analyze passenger travel behavior using historical travel data. However, this approach is ineffective for recommending intercity travel plans, given the small amount of data and limited intercommunication between urban and intercity travel data. Summary of the Invention

[0003] In order to solve the technical problems mentioned in the above background technology, the present invention proposes a method and device for recommending personalized travel plans for the entire process of air-rail intermodal travel. By obtaining the travel preferences of users of user terminal units, suitable air-rail intermodal or aviation travel plans for the entire process from departure to destination are recommended based on their travel needs, thereby inducing passengers to adopt air-rail intermodal travel in appropriate scenarios.

[0004] In order to achieve the above technical objectives, the technical solution of the present invention is:

[0005] A first aspect of an embodiment of the present invention provides a method for recommending personalized travel plans for an entire air-rail intermodal journey, comprising the following steps:

[0006] Step S1: Sending a dynamic questionnaire to a user terminal unit, and generating subsequent questions of the dynamic questionnaire based on the dynamic questionnaire selection result returned by the user terminal unit, and finally generating the travel preferences of the user of the user terminal unit;

[0007] Step S2: upon receiving a travel plan query request sent by a user terminal unit, generating a set of feasible air-rail connection or air travel plans;

[0008] Step S3: Score all feasible air-rail connections or air travel options generated in step S2 based on the travel preferences obtained in step S1;

[0009] Step S4: Based on the scoring results obtained in step S3, all feasible air-rail connections or air travel plans generated in step S2 are sorted from high to low according to the scores and sent to the user terminal unit.

[0010] Furthermore, the method for obtaining user travel preferences based on the dynamic questionnaire in step S1 includes the following steps:

[0011] Step S1-1: linearly weighting the factors influencing travel decisions to construct a cost function. The factors influencing travel decisions include total travel time, total travel cost, and travel plan type. A dynamic questionnaire question is pushed to the user terminal unit. The question is to select two more important travel plan factors among the three factors influencing travel decisions.

[0012] Step S1-2: Generate two travel plan options, wherein the two returned travel plan influencing factors have advantages and disadvantages in each of the travel plan options, and the remaining factors remain consistent in the two plans, and push a dynamic questionnaire question to the user terminal unit, wherein the question is to select the plan that better meets the user's travel preferences based on the influencing factors of the two travel plan options;

[0013] Step S1-3: Receive the selection result returned by the user terminal unit, and calculate the range of the ratio of the weights of the two more important travel plan influencing factors returned. If the convergence condition is met, execute step S1-4; if not, execute step S1-2 again;

[0014] Step S1-4: Generate travel plan options and push a dynamic questionnaire question to the user terminal unit, wherein the question is to select a plan that better meets the user's travel preferences based on the influencing factors of the two travel plan options;

[0015] Step S1-5: Receive the selection result returned by the user terminal unit, calculate the range of the weights of the factors affecting the travel plan, and if the convergence condition is met, generate the travel preference of the user of the user terminal unit. If not, execute step S1-4 again and push new dynamic questionnaire questions to the user terminal unit.

[0016] Furthermore, steps S1-2 and S1-4 generate two travel plan options, and provide the values ​​of three influencing factors, namely, total travel time, total cost, and plan type, to the user for selection.

[0017] Furthermore, the method for generating travel plan options for the dynamic questionnaire questions in steps S1-2 and S1-4 includes:

[0018] When step S1-2 is executed for the first time, a selection result returned by the user terminal unit, i.e., two travel plan influencing factors, is received, the geographical location of the user terminal unit is used as the departure point, a destination is randomly selected, and multiple travel plans are generated to form a dynamic questionnaire plan library. Travel plans with mutually advantageous and inferior influencing factors of the two travel plans are screened from the dynamic questionnaire plan library, and the remaining factor of the plan is unified to the same value as the generated plan;

[0019] When step S1-4 is performed for the first time, based on the weight ratio of the two travel plan influencing factors, two travel plans are selected from the dynamic questionnaire plan library, where the weighted sum of the two factors according to the weight ratio has advantages and disadvantages over the remaining factor, and the two travel plans are used as the generated plans;

[0020] When step S1-2 is executed again, based on the upper and lower limits of the ratio of the weights of the influencing factors of the two travel plans, two unselected travel plans with the same influencing factor values ​​and with the selected factors having advantages and disadvantages to each other are generated as the generated plan;

[0021] When step S1-4 is executed again, a plan is generated as the generated plan, in which the unselected factors and the selected factors have advantages and disadvantages relative to each other, based on the ratio of the weights of the two travel plan influencing factors and the upper and lower limits of the weights;

[0022] Preferably, when step S1-2 or S1-4 is executed again, when the upper and lower limits of the ratio of the weights of the influencing factors of the two travel plans only include the upper limit or the lower limit, two plans are selected from the plan library so that the ratio of the weights of the two plans is less than the upper limit or greater than the lower limit. If there are no two plans that meet the requirements in the plan library, a travel plan is generated based on one of the plans in the plan library so that the ratio of the weights is 50% of the upper limit or twice the lower limit, and this is used as the generated plan;

[0023] Preferably, when step S1-2 or S1-4 is executed again, when the upper and lower limit ranges of the ratio of the weights of the influencing factors of the two travel plans include both the upper and lower limits, two plans are selected from the plan library so that the ratio of the weights of the two plans is less than the upper limit and greater than the lower limit. If there are no two plans that meet the requirements in the plan library, a travel plan is generated based on one plan in the plan library so that the ratio of the weights of the two plans is the average of the upper and lower limits, which is used as the generated plan.

[0024] Furthermore, the weights of the factors influencing the travel plan in steps S1-3 and S1-5 are determined based on the values ​​of the factors influencing the two travel plan options and the selection result returned by the user equipment unit, and the determination method includes:

[0025] Construct the travel plan cost function, the formula is as follows:

[0026] C=ω T T+ω P P+ω m m

[0027] Where T is the total time of the travel plan, P is the total cost of the travel plan, m is the type of travel plan represented by a Boolean value, ω T 、ω P、ω m is the weight of each travel decision-making factor;

[0028] According to the selection result received when step S1-2 is performed for the first time, the weights of the influencing factors of the unselected travel options are set to 1;

[0029] According to the selection results received in steps S1-3 and S1-5, the cost functions are C1=ω T T1+ω P P1+ω m m1 and C2 = ω T T2+ω P P2+ω m m2 has two travel options. If the result is option 1, then C1 < C2; if the result is option 2, then C1 > C2;

[0030] Substituting the cost function into the inequality, we can get the upper or lower limit of the ratio of weights.

[0031] Furthermore, the convergence condition in steps S1-3 and S1-5 refers to that the difference between the upper and lower limits of the weights or weight ratios of the two travel plan influencing factors is less than the allowable value, and the average of the upper and lower limits is taken as the value of the weights or weight ratio of the two travel plan influencing factors.

[0032] Furthermore, the method for generating a feasible connecting travel solution in step S2 includes the following steps:

[0033] Step S2-1: Obtain a travel plan query request sent by a user terminal unit, wherein the request includes a departure location, a destination point, a travel date, and departure and arrival time window requirements;

[0034] Step S2-2: Match the departure airport with eligible air-rail intermodal airports in the departure airport list to form a departure airport list, and match the destination airport with eligible air-rail intermodal airports in the destination airport list to form a destination airport list;

[0035] Step S2-3: Calculate the minimum feasible travel time from the departure point to each airport in the departure airport list and the minimum feasible travel time from each airport in the destination airport list to the destination;

[0036] Step S2-4: Filter feasible flights from each departure airport to each arrival airport to form an initial set of travel plans. The take-off and landing times of the air portion of the travel plan must meet the time window requirements. The time window requirements are that the difference between the take-off time and the earliest departure time in the departure time window must be greater than the minimum feasible travel time from the departure point to the airport calculated in step S2-3, and the difference between the latest arrival time and the landing time in the arrival time window must be greater than the minimum feasible travel time from the airport to the destination;

[0037] Step S2-5: For travel plans in the set of travel plans where the departure or arrival airport of the aviation part is an air-rail combined airport, record the plan type as an air-rail combined travel plan and match the railway part of the plan. The types of the remaining plans are recorded as aviation plans;

[0038] Step S2-6: Matching the departure city travel part and the destination city travel part of each travel plan in the travel plan set;

[0039] Step S2-7: Based on the matching results of each part of each travel plan, calculate the plan departure time, estimated arrival time, total travel time and total cost, and only retain the plans whose departure and arrival times meet the time window requirements in the plan set to obtain the feasible connecting travel plan.

[0040] Preferably, the method for matching eligible air-rail transport airports around the departure or destination in step S2-2 includes:

[0041] Match multiple airports that are closest to the departure and destination points;

[0042] Calculate the minimum feasible time from the departure point to the surrounding airports and the minimum time from the destination to the surrounding airports, and calculate the minimum value to obtain the minimum feasible time from the departure point and the destination to the airport;

[0043] Select airports with air-rail transport conditions that can directly connect to the departure or destination city by rail from those with air-rail transport conditions.

[0044] Calculate the minimum feasible time for the departure point to reach each air-rail intermodal airport by transferring to urban transportation via rail, or for each air-rail intermodal airport to reach the destination by transferring to urban transportation via rail;

[0045] Filter out the air-rail intermodal airports where the minimum feasible travel time to the departure or destination is less than a threshold, where the threshold is a specified multiple of the minimum feasible travel time from the departure or destination to the surrounding airports;

[0046] The screened airports are used as matching eligible air-rail transport airports.

[0047] Furthermore, in step S3, the feasible air-rail connection and air travel solutions are scored using a weighted superiority-inferiority distance method, including the following steps:

[0048] Step S3-1: Determine whether the solution type weight in the cost function satisfies ω m ≥0, if it is not satisfied, then a new cost function is established so that ω mThe value of is the absolute value of the original weight, and the value of scheme type m is the value after the original value is not calculated. The form of the cost function, the meaning of the variables and other weights remain unchanged.

[0049] Step S3-2: The total travel time T of each plan in the travel plan set i , total cost P i and solution type m i The three indicators were normalized and converted using the weight method.

[0050] Step S3-3: Calculate the weighted distance of each solution from the positive ideal solution and the negative ideal solution in the solution space and Where T i ′、P i ′、m i ′ are the total time, total cost and solution type value of the i-th solution after normalization, T m ' ax 、P m ' ax 、m′ max 、T m ' in 、P m ' in 、m′ min are the maximum and minimum values ​​of the total time, total cost and solution type of each solution after normalization;

[0051] Step S3-4: Calculate the score of each solution. The score of the i-th solution is S i The calculation formula is:

[0052]

[0053] A second aspect of an embodiment of the present invention provides a device for recommending personalized travel plans for an entire air-rail intermodal journey, including:

[0054] A user terminal unit is configured to receive and display information sent by the preference analysis unit or the solution recommendation unit, and return information to the preference analysis unit or the solution generation unit;

[0055] a preference analysis unit for analyzing preferences of users of the user terminal unit;

[0056] A plan generating unit, configured to generate a feasible travel plan according to a travel request;

[0057] A plan recommendation unit is used to score, sort and recommend travel plans generated by the plan generation unit;

[0058] Optionally, the user terminal unit is specifically configured to receive and display dynamic questionnaire questions pushed by the preference analysis unit, provide them to the user for selection, and feed back the selection results to the preference analysis unit. It is also configured to process the travel plan query request input by the user and send it to the plan generation unit. It is also configured to display the travel plan results sent by the plan recommendation unit.

[0059] Optionally, the preference analysis unit is specifically configured to construct a cost function, generate and send dynamic questionnaire questions to the user terminal unit, receive returned results and calculate a dynamic questionnaire weight range based on the results, and generate new dynamic questionnaire questions to determine the cost function weights;

[0060] Optionally, the solution generation unit is specifically configured to receive a solution query request sent by the user terminal unit, match eligible air-rail intermodal airports, screen feasible flights, mark solution types and match full-process travel solutions, and calculate indicators such as the departure time, estimated arrival time, total time, and total cost of the solution to screen out feasible solutions;

[0061] Optionally, the solution recommendation unit is specifically used to negatively normalize, normalize, and score feasible travel solutions using a weighted superior and inferior solution distance solution scoring method, and sort the solutions by score and send them to the user terminal unit.

[0062] The beneficial effects brought about by adopting the above technical solution are:

[0063] The travel plan recommendation method provided by the present invention adopts a technical solution of obtaining the overall preferences of user terminal unit users for the entire intercity travel process through dynamic questionnaires, and scoring and personalized recommendations for travel plans covering the entire process from the departure point to the destination point, in the absence of relevant data sets. It enhances the integrity and availability of travel planning and recommendation content, and improves the accuracy of recommended travel plans, which can achieve the beneficial effects of improving the level of intercity integrated travel services and enhancing the benefits of users adopting air-rail intermodal travel. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 Flowchart of a method for recommending personalized travel plans for the entire air-rail intermodal journey according to an embodiment of the present invention;

[0065] Figure 2 is a flow chart of a method for obtaining travel preferences of a user terminal unit based on a dynamic questionnaire according to an embodiment of the present invention;

[0066] Figure 3 This is a flowchart of a method for generating a feasible connecting travel solution according to an embodiment of the present invention;

[0067] Figure 4Flowchart of a weighted scoring method for superior and inferior solution distances provided in accordance with an embodiment of the present invention;

[0068] Figure 5 2 is a schematic diagram of the structure of a device for recommending personalized travel plans for the entire air-rail intermodal journey provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0069] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings.

[0070] One application scenario of this embodiment is as follows: When a user of a user terminal unit uses the device for the first time, the preference analysis unit obtains the user's travel preferences by sending a dynamic questionnaire. Subsequently, when the user terminal unit sends a travel request, the solution generation unit calculates a set of feasible travel solutions, and the solution recommendation unit scores the travel solutions based on the travel preferences and sends them to the user terminal unit for recommendation.

[0071] like Figure 1 As shown, the first aspect of the embodiment of the present invention provides a method for recommending personalized travel plans for the entire process of air-rail intermodal travel, including the following steps:

[0072] Step S1: Send the dynamic questionnaire to the user terminal unit, and generate subsequent questions of the dynamic questionnaire based on the dynamic questionnaire selection results returned by the user terminal unit, and finally generate the travel preferences of the user of the user terminal unit;

[0073] This step obtains the travel preferences of the user terminal unit based on the questionnaire selection results obtained by the user terminal unit. The dynamic questionnaire has a fixed structure, but the number of questions and the selection content sent to the user terminal unit in the questionnaire are dynamically generated based on the geographical location of the user terminal unit and the selections made. There is no fixed questionnaire content. Figure 2 As shown, the method for obtaining the travel preference of a user terminal unit based on a dynamic questionnaire provided by an embodiment of the present invention includes the following steps:

[0074] Step S1-1: To reduce the complexity of passenger choices and improve survey accuracy, the three travel decision-making factors (total travel time, total cost, and plan type) are linearly weighted to construct a cost function. A dynamic questionnaire question is pushed to the user terminal unit. The question asks the user to select two more important travel plan factors among the three travel decision-making factors.

[0075] The decision-making objectives of users of user terminal units for travel plans mainly focus on travel time, travel cost, number of transfers, walking distance, comfort, etc. Since the departure points of passengers in intercity travel are relatively fixed, for aviation plans and air-rail travel plans, the city travel parts of each plan in each category are consistent. The only difference is the difference in departure time and the difference in public transportation intervals caused by departure time. Since the number of transfers, walking distance and comfort factors are only related to the type of plan, the type of plan is used instead of factors related only to the number of transfers, walking distance, comfort, etc. Since the indicators are independent of each other, linear addition is used to select and construct the value function C = ω T T+ω P P+ω m m, where T is the total time of the travel plan, P is the total cost of the travel plan, m is the type of travel plan represented by a Boolean value, ω T 、ω P 、ω m is the weight of each factor affecting travel decision.

[0076] In one embodiment, the following questions are pushed to the user terminal unit:

[0077] When choosing between multiple options for intercity travel, which of the three factors—total travel time, total cost, and option type (including air and air-rail combined travel)—is your least important?

[0078] 1. Total time; 2. Total cost; 3. Solution type"

[0079] Step S1-2: Based on the selection results returned by the receiving user terminal unit, the weights of the influencing factors of the unselected travel plans are set to 1 to design a non-inferior solution and give priority to the weights of the two influencing factors; two travel plan options are generated, and a dynamic questionnaire question is pushed to the user terminal unit. The question is to select the plan that better meets the travel preferences based on the influencing factors of the travel plan among the two travel plan options. The options include the values ​​of the three influencing factors of the total travel time, total cost and plan type of the plan. Among them, the two returned travel plan influencing factors have their own advantages and disadvantages in the two plans, and the remaining factor remains consistent in the two plans. The user of the user terminal unit selects the plan that better meets his travel preferences based on the influencing factors of the two plans, that is, the plan with a smaller cost function. For any two plans c1 and c2, their cost functions are C1=ω respectively. T T1+ω P P1+ω m m1 and C2 = ω T T2+ω P P2+ω m m2, if and only if C1<C2, then If plan c1 is strictly better than plan c2 in terms of selection, the range of the ratio of the weights of the influencing factors of the two travel plans obtained in step S1-1 can be obtained.

[0080] Among them, when step S1-2 is executed for the first time, the selection result returned by the user terminal unit, that is, the two travel plan influencing factors, is received, the geographical location of the user terminal unit is used as the departure point, the destination is randomly selected, multiple travel plans are generated, and a dynamic questionnaire plan library is formed. Two travel plans with mutual advantages and disadvantages in the two travel plan influencing factors are screened out from the dynamic questionnaire plan library, and the remaining factor of the two plans is unified to the same value as the generated plan.

[0081] In one embodiment, the following questions are first pushed to the user terminal unit:

[0082] Please choose the one you prefer between the following two intercity travel options:

[0083] 1. Air travel, 6 hours and 48 minutes, 729 yuan; 2. Air travel, 5 hours and 36 minutes, 859 yuan

[0084] When step S1-2 is executed again, two travel plans are generated based on the upper and lower limits of the weight ratio of the two travel plan influencing factors selected by the user terminal unit in step S1-1, meeting the requirements that the influencing factors of the unselected travel plans have the same values ​​and that the two selected factors have advantages and disadvantages. If the upper and lower limits of the weight ratio of the two travel plan influencing factors only include the upper or lower limit, two plans are selected from the plan library so that the weight ratio of the two plans is less than the upper limit or greater than the lower limit. If no two plans meet the requirements in the plan library, a travel plan is generated based on one of the plans in the plan library so that the weight ratio is 50% of the upper limit or twice the lower limit, and the plan is pushed to the user terminal unit. If the upper and lower limits of the weight ratio of the two travel plan influencing factors include both the upper and lower limits, two plans are selected from the plan library so that the weight ratio of the two plans is less than the upper limit and greater than the lower limit. If no two plans meet the requirements in the plan library, a travel plan is generated based on one of the plans in the plan library so that the weight ratio of the two plans is the average of the upper and lower limits, and the plan is pushed to the user terminal unit.

[0085] Step S1-3: Based on the selection result returned by the receiving user terminal unit, that is, the plan with the smaller cost function among the two plan options, calculate the range of the ratio of the weights of the two more important travel plan influencing factors returned in the cost function of the travel plan. If the difference between the upper and lower limits of the ratio of the weights of the two travel plan influencing factors is less than the allowable value, take the average of the upper and lower limits as the value of the ratio of the weights of the two travel plan influencing factors and execute step S1-4. If it is not satisfied, execute step S1-2 again. When the selection result returned by the receiving user terminal unit during the first execution of step S1-2 is 1, let ω T =1, OK range; when the received selection result is 2, let ω P =1, OK range; when the received selection result is 3, let ω m =1, OK range.

[0086] In one embodiment, when the selection result returned by the user terminal unit when step S1-2 is performed for the first time is 1, the result is If the received selection result is 2, you can get

[0087] Step S1-4: Generate two travel plans that meet the requirement that the weighted sum of the two selected factors according to the ratio of their weights has advantages and disadvantages over the remaining factor. Since the value of the weight ratio of the influencing factors of the two travel plans has been determined, it is not necessary to keep some influencing factors consistent in the given plans. A dynamic questionnaire question is pushed to the user terminal unit. The question is to select the plan that better meets the user's travel preferences based on the influencing factors of the two travel plans. The options include the values ​​of the three influencing factors of the total travel time, total cost, and plan type of the plan. The user of the user terminal unit selects the plan that better meets his or her travel preferences based on the influencing factors of the two plans.

[0088] Among them, when step S1-4 is executed for the first time, according to the ratio of the weights of the influencing factors of the two travel plans, two travel plans whose weighted sum of the two factors according to the weight ratio has advantages and disadvantages over the remaining factor are screened out from the dynamic questionnaire plan library as the generated plan.

[0089] In one embodiment, in this step, the following questions are first pushed to the user terminal unit to obtain the range of weights:

[0090] Please choose the one you prefer between the following two intercity travel options:

[0091] 1. Air travel, 6 hours and 25 minutes, 539 yuan; 2. Air-rail combined travel, 5 hours and 52 minutes, 459 yuan

[0092] When step S1-4 is executed again, a plan is generated in which the two unselected factors have advantages and disadvantages over the selected factor based on the ratio of the weights of the two travel plan influencing factors and the upper and lower limits of the weights. The method for generating the plan is the same as that in step S1-2.

[0093] Step S1-5: Receive the selection result returned by the user terminal unit, calculate the range of the weights of the factors affecting the travel plan, and if the convergence condition that the difference between the upper and lower limits of the weights of the factors affecting the travel plan is less than the allowable value is met, take the average of the upper and lower limits as the value of the weights of the factors affecting the travel plan, and obtain the size of each weight in the travel cost function of the user of the user terminal unit. The travel preference of the user of the user terminal unit can be obtained in the form of a cost function. If it is not met, execute step S1-4 again to push new dynamic questionnaire questions to the user terminal unit.

[0094] Step S2: upon receiving a travel plan query request sent by a user terminal unit, generating a set of feasible connecting travel plans;

[0095] This step generates a set of feasible connecting travel solutions based on the travel needs of the user terminal unit, including air travel and air-rail connecting travel solutions. Figure 3 As shown, the method for generating a feasible connecting travel solution provided by an embodiment of the present invention includes the following steps:

[0096] Step S2-1: Obtaining a travel plan query request sent by a user terminal unit, including the specific departure location in the city, the specific destination in the city where the user of the user terminal unit is traveling, the expected travel date, the earliest and latest departure times, and the earliest and latest arrival time requirements;

[0097] In some embodiments, the departure location is determined based on the geographic location of the user terminal unit, and the travel date is defaulted to the current date.

[0098] Step S2-2: Match the departure airport with qualified air-rail transport airports in the departure airport list to form a departure airport list, and match the destination airport with qualified air-rail transport airports in the destination airport list to form a destination airport list. The matching method for the air-rail transport airports includes:

[0099] Based on the geographic location information of all airports, match multiple airports that are closest to the departure and destination points;

[0100] Calculate the minimum feasible time from the departure point to the surrounding airports and the minimum time from the destination to the surrounding airports, and calculate the minimum value to obtain the minimum feasible time from the departure point and the destination to the airport;

[0101] Select airports with air-rail transport conditions that can directly connect to the departure or destination city by rail from those with air-rail transport conditions.

[0102] Calculate the minimum feasible time for the departure point to reach each air-rail intermodal airport by transferring to urban transportation via rail, or for each air-rail intermodal airport to reach the destination by transferring to urban transportation via rail;

[0103] Filter out the air-rail intermodal airports whose minimum feasible travel time to the departure or destination is less than a threshold, where the threshold is a specified multiple of the minimum feasible travel time from the departure or destination to the nearest surrounding airport, for example, 3 in some embodiments;

[0104] The screened airports are used as matching eligible air-rail transport airports.

[0105] Step S2-3: Calculate the minimum feasible travel time from the departure point to each airport in the departure airport list and the minimum feasible travel time from each airport in the destination airport list to the destination;

[0106] In some embodiments, the minimum feasible time is calculated using the following method:

[0107] The time for urban travel is calculated based on the travel time during the peak period when the vehicle frequency is the highest, that is, the minimum value is taken.

[0108] Among them, for air-rail intermodal airports, the minimum feasible time is calculated using the situation where the additional waiting time for air-rail transfer is 0, that is, the air and rail services are just connected to each other, and the rail time takes the minimum value of all rail options.

[0109] Step S2-4: Filter feasible flights from each departure airport to each arrival airport to form an initial set of travel plans. The take-off and landing times of the air portion of the travel plan must meet the time window requirements. The time window requirements are that the difference between the take-off time and the earliest departure time in the departure time window must be greater than the minimum feasible travel time from the departure point to the airport calculated in step S2-3, and the difference between the latest arrival time and the landing time in the arrival time window must be greater than the minimum feasible travel time from the airport to the destination;

[0110] Step S2-5: For travel plans in the set of travel plans where the departure or arrival airport of the aviation part is an air-rail combined airport, record the plan type as an air-rail combined travel plan and match the railway part of the plan. The types of the remaining plans are recorded as aviation plans;

[0111] In some embodiments, to match the rail portion of an air-rail travel plan, an air-rail transfer time is set for the air-rail airport. This time is the difference between flight landing and rail departure, or between rail arrival and flight departure, at that hub. The transfer time is determined based on the specific conditions of the air-rail hub. The rail travel plan with the smallest difference between air and rail times that exceeds the air-rail transfer time for that hub is matched as the rail portion of the plan.

[0112] Step S2-6: Matching the departure city travel part and the destination city travel part of each travel plan in the travel plan set;

[0113] In some embodiments, the travel plan for the city travel part can be queried through the relevant interface. For the air travel plan, the travel plan for the city travel part is the travel plan from the departure point to the departure airport or the destination airport to the destination; for the air-rail travel plan, the travel plan for the city travel part is the travel plan from the departure point to the departure railway station or the destination railway station to the destination.

[0114] Step S2-7: Based on the matching results of each part of each travel plan, calculate the plan departure time, estimated arrival time, total travel time and total cost, and only retain the plans whose departure and arrival times meet the time window requirements in the plan set to obtain the feasible connecting travel plan.

[0115] Step S3: Based on the travel preferences, the feasible air-rail connection and air travel options are scored using a weighted superiority-inferiority distance method based on the three dimensions of total time, total cost and option type, so that the best option can be selected from multiple options and recommended to the user terminal unit for travel. Figure 4 As shown, the weighted superior and inferior solution distance scoring method provided by the embodiment of the present invention includes the following steps:

[0116] Step S3-1: Minimize the data, that is, in the three decision-making objectives involved in the solution, the smaller the value, the better. Since only the solution type may not meet the minimization requirement, it is determined whether the solution type weight in the cost function meets ω m ≥0, if it is not satisfied, then a new cost function is established so that ω m The value of is the absolute value of the original weight, and the value of scheme type m is the value after the original value is negated, that is, the value after multiplying the original value by -1. The form of the cost function, the meaning of the variables and other weights remain unchanged;

[0117] Step S3-2: For the three indicators of each plan in the travel plan set, use the weight method to perform data normalization conversion on their values. Take the total travel time T of the i-th plan as i , total cost P i and solution type m i For example, perform the normalization conversion using the following formula:

[0118]

[0119]

[0120]

[0121] Among them, T′, P′, and m′ are the normalized total time, total cost, and solution type values ​​of each solution, respectively.

[0122] Step S3-3: Calculate the weighted distance of each solution from the positive ideal solution and the negative ideal solution in the solution space, where the positive ideal solution vector is the optimal value of the normalized total time, total cost, and solution type value of each solution, that is, the vector composed of the minimum value, that is:

[0123] z + =[T m ' ax P m ' ax m′ max ]

[0124] The negative ideal solution vector is the worst value of the normalized total time, total cost, and solution type value of each solution, that is, the vector composed of the maximum value, that is:

[0125] z + =[T m ' in P m ' in m′ min ]

[0126] Where T m ' ax 、P m ' ax 、m′ max 、T m ' in 、P m ' in 、m′ min They are the maximum and minimum values ​​of the total time, total cost and solution type of each solution after normalization.

[0127] In the three-dimensional space formed by the three indicators, the three coordinate axes represent the factors affecting the three travel options. Each option is represented by a vector consisting of the values ​​on these three coordinate axes, forming a point in the geometric space. The optimal and worst solutions are also represented as points in this geometric space by the positive and negative ideal solution vectors. For example, for the first option, (T1, P1, m1) is the point in the space representing it.

[0128] The distance between the best and worst solutions represents the weighted spatial distance between each solution and the optimal solution and the worst solution in the geometric space. + The weighted distance is:

[0129]

[0130] Solution i for the worst solution z - The weighted distance is:

[0131]

[0132] For two solutions with the same original value functions, the weight of the weighted distance can satisfy that the standardized value functions of the two solutions are also equal.

[0133] Step S3-4: Calculate the score of each solution. The score of the i-th solution is S i The calculation formula is:

[0134]

[0135] Step S4: Based on the travel plan scores, the feasible air-rail connection or air travel plans are sorted from high to low according to the scores and sent to the user terminal unit.

[0136] In some embodiments, the recommendation method may include pushing in order from high to low according to the plan score, sorting and pushing according to the plan type, pushing the air-rail intermodal travel plan at a fixed location, pushing according to the plan characteristics, etc.

[0137] like Figure 5 As shown, the second aspect of the embodiment of the present invention provides a device for recommending personalized travel plans for the entire process of air-rail intermodal travel, including:

[0138] The user terminal unit 51 is configured to receive and display information sent by the preference analysis unit 52 or the solution recommendation unit 54, and return information to the preference analysis unit 52 or the solution generation unit 54;

[0139] a preference analysis unit 52 for analyzing preferences of a user of the user terminal unit 51;

[0140] A plan generating unit 53 is used to generate a feasible travel plan according to the travel request;

[0141] A plan recommendation unit 54 is used to score, sort and recommend the travel plans generated by the plan generation unit 53;

[0142] Optionally, in one embodiment, the user terminal unit 51 is specifically configured to receive and display dynamic questionnaire questions pushed by the preference analysis unit 52, provide them to the user for selection, and feed back the selection results to the preference analysis unit 52. It is also configured to process the travel plan query request input by the user and send it to the plan generation unit 53. It is also configured to display the travel plan results sent by the plan recommendation unit 54.

[0143] Optionally, in one embodiment, the preference analysis unit 52 is specifically configured to construct a travel plan cost function, generate and send dynamic questionnaire questions to the user terminal unit 51, receive returned results, calculate a dynamic questionnaire weight range based on the results, and generate new dynamic questionnaire questions to determine the cost function weights;

[0144] Optionally, in one embodiment, the plan generation unit 53 is specifically used to receive a plan query request sent by the user terminal unit 51, and match qualified air-rail transport airports, screen flights that meet the time window requirements, mark the plan type and match the railway part and the travel plans within the departure and arrival cities, and calculate the departure time, expected arrival time, total time, total cost and other indicators of the plan to screen out feasible plans; it is also used to generate travel plans based on the departure place and select multiple destinations to form a dynamic questionnaire plan library; it is also used to calculate the minimum feasible time from the departure place or destination to the departure or destination airport based on the travel plan.

[0145] Optionally, in one embodiment, the solution recommendation unit 54 is specifically used to negatively normalize, normalize, and score feasible travel solutions using a weighted superior-inferior solution distance solution scoring method, and sort the solutions by score and send them to the user terminal unit 51.

[0146] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, devices, or computer program products. Therefore, the application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiment of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal translation scripting language JavaScript, etc.

[0147] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (apparatus), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0148] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0149] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0150] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0151] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is intended to include these modifications and variations.

Claims

1. A method for recommending personalized travel plans for the entire air-rail intermodal journey, characterized in that: The following steps are involved: Step S1: Sending a dynamic questionnaire to a user terminal unit, and generating subsequent questions of the dynamic questionnaire based on the dynamic questionnaire selection result returned by the user terminal unit, and finally generating the travel preferences of the user of the user terminal unit; Step S2: receiving a travel plan query request sent by a user terminal unit, and generating a set of feasible air-rail connection or air travel plans; Step S3: Score all feasible air-rail connections or air travel options generated in step S2 based on the travel preferences obtained in step S1; Step S4: Based on the scoring result obtained in step S3, all feasible air-rail connections or air travel plans generated in step S2 are sorted from high to low according to the scores and sent to the user terminal unit; The step S1 specifically includes the following steps: Step S1-1: linearly weighting the factors influencing travel decisions to construct a cost function. The factors influencing travel decisions include total travel time, total travel cost, and travel plan type. A dynamic questionnaire question is pushed to the user terminal unit. The question is to select two more important travel plan factors among the three factors influencing travel decisions. Step S1-2: Generate two travel plan options, wherein the two returned travel plan influencing factors have advantages and disadvantages in each of the travel plan options, and the remaining factors remain consistent in the two plans, and push a dynamic questionnaire question to the user terminal unit, wherein the question is to select the plan that better meets the user's travel preferences based on the influencing factors of the two travel plan options; Step S1-3: Receive the selection result returned by the user terminal unit, and calculate the range of the ratio of the weights of the two more important travel plan influencing factors returned. If the convergence condition is met, execute step S1-4; if not, execute step S1-2 again; Step S1-4: Generate travel plan options and push a dynamic questionnaire question to the user terminal unit, wherein the question is to select a plan that better meets the user's travel preferences based on the influencing factors of the two travel plan options; Step S1-5: Receive the selection results returned by the user terminal unit, calculate the range of weights of the factors affecting the travel plan, and if the convergence conditions are met, generate the travel preferences of the user of the user terminal unit. If not, execute step S1-4 again and push new dynamic questionnaire questions to the user terminal unit.

2. The method for recommending personalized travel plans for the entire air-rail journey according to claim 1, characterized in that: The method for generating travel plan options for the dynamic questionnaire questions in steps S1-2 and S1-4 specifically includes: When step S1-2 is executed for the first time, a selection result returned by the user terminal unit, i.e., two travel plan influencing factors, is received, the geographical location of the user terminal unit is used as the departure point, a destination is randomly selected, and multiple travel plans are generated to form a dynamic questionnaire plan library. Travel plans with mutually advantageous and inferior influencing factors of the two travel plans are screened from the dynamic questionnaire plan library, and the remaining factor of the plan is unified to the same value as the generated plan; When step S1-4 is performed for the first time, based on the weight ratio of the two travel plan influencing factors, two travel plans are selected from the dynamic questionnaire plan library, where the weighted sum of the two travel plan influencing factors according to the weight ratio has advantages and disadvantages over the remaining factor, and the two travel plans are used as the generated plans; When step S1-2 is executed again, based on the upper and lower limits of the ratio of the weights of the influencing factors of the two travel plans, two unselected travel plans with the same influencing factor values ​​and with the selected factors having advantages and disadvantages to each other are generated as the generated plan; When step S1-4 is executed again, a plan is generated as the generated plan, in which the unselected factors and the selected factors have advantages and disadvantages relative to each other, based on the ratio of the weights of the two travel plan influencing factors and the upper and lower limits of the weights; The solution generation method when performing step S1-2 or S1-4 again includes, when the upper and lower limit ranges of the weight ratio include only the upper limit or the lower limit, selecting a solution from the solution library so that the weight ratio of the solution is less than the upper limit or greater than the lower limit; if the solution does not exist in the solution library, generating a travel solution based on a solution in the solution library so that the weight ratio is 50% of the upper limit or twice the lower limit, and using this solution as the generated solution; The solution generation method when executing step S1-2 or S1-4 again also includes, when the upper and lower limit ranges of the weight ratio include both the upper and lower limits, selecting two solutions from the solution library so that the weight ratio of the two solutions is less than the upper limit and greater than the lower limit; if the two solutions do not exist in the solution library, generating a travel solution based on one solution in the solution library so that the weight ratio of the two solutions is the average of the upper and lower limits, as the generated solution.

3. The method for recommending personalized travel plans for the entire air-rail journey according to claim 1, characterized in that: The method for determining the travel plan weight in steps S1-3 and S1-5 specifically refers to: Construct the travel plan cost function, the formula is as follows: C=ω T T+ω P P+ω m m Where T is the total time of the travel plan, P is the total cost of the travel plan, m is the type of travel plan represented by a Boolean value, ω T 、ω P 、ω m is the weight of each travel decision-making factor; According to the selection result received when step S1-2 is performed for the first time, the weights of the influencing factors of the unselected travel options are set to 1; According to the selection results received in steps S1-3 and S1-5, the cost functions are C1=ω T T1+ω P P1+ω m m1 and C2 = ω T T2+ω P P2+ω m m2 has two travel options. If the result is option 1, then C1 < C2; if the result is option 2, then C1 > C2; Substituting the cost function into the inequality, an upper limit or a lower limit range of the ratio of the weights is obtained.

4. The method for recommending personalized travel plans for the entire air-rail journey according to claim 1, characterized in that: The convergence condition in step S1-3 or S1-5 refers to that the difference between the upper and lower limits of the weight or weight ratio is less than the allowable value, and the average of the upper and lower limits is taken as the value of the weight or weight ratio.

5. The method for recommending personalized travel plans for the entire air-rail intermodal journey according to claim 1, characterized in that: The method for generating a feasible connecting travel solution in step S2 includes the following steps: Step S2-1: Obtaining a travel plan query request sent by a user terminal unit, wherein the request includes a departure location, a destination point, a travel date, and departure and arrival time window requirements; Step S2-2: Match the departure airport with eligible air-rail intermodal airports in the departure airport list to form a departure airport list, and match the destination airport with eligible air-rail intermodal airports in the destination airport list to form a destination airport list; Step S2-3: Calculate the minimum feasible travel time from the departure point to each airport in the departure airport list and the minimum feasible travel time from each airport in the destination airport list to the destination; Step S2-4: Filter feasible flights from each departure airport to each arrival airport to form an initial set of travel plans, where the take-off and landing times of the air portion of the travel plans meet the time window requirements. The time window requirements are that the difference between the take-off time and the earliest departure time in the departure time window is greater than the minimum feasible travel time from the departure point to the airport calculated in step S2-3, and the difference between the latest arrival time and the landing time in the arrival time window is greater than the minimum feasible travel time from the airport to the destination; Step S2-5: For travel plans in the set of travel plans where the departure or arrival airport of the aviation part is an air-rail combined airport, record the plan type as an air-rail combined travel plan and match the railway part of the plan. The types of the remaining plans are recorded as aviation plans; Step S2-6: Matching the departure city travel part and the destination city travel part of each travel plan in the travel plan set; Step S2-7: Based on the matching results of each part of each travel plan, calculate the plan departure time, estimated arrival time, total travel time and total cost, and only retain the plans whose departure and arrival times meet the time window requirements in the plan set to obtain the feasible connecting travel plan.

6. The method for recommending personalized travel plans for the entire air-rail journey according to claim 5, characterized in that: The matching method of eligible air-rail transport airports around the departure or destination in step S2-2 includes: Match multiple airports that are closest to the departure and destination points; Calculate the minimum feasible travel time from the departure place to each of the surrounding airports and from the destination to each of the surrounding airports, and calculate the minimum value to obtain the minimum feasible travel time from the departure place and the destination to the airport; Select airports with air-rail transport conditions that can directly connect to the departure or destination city by rail from those with air-rail transport conditions. Calculate the minimum feasible time for the departure point to reach each air-rail intermodal airport by transferring to urban transportation via rail, or for each air-rail intermodal airport to reach the destination by transferring to urban transportation via rail; Filter out the air-rail intermodal airports where the minimum feasible travel time to the departure or destination is less than a threshold, where the threshold is a specified multiple of the minimum feasible travel time from the departure or destination to the surrounding airports; The screened airports are used as matching eligible air-rail transport airports.

7. The method for recommending personalized travel plans for the entire air-rail journey according to claim 3, characterized in that: The weighted superior and inferior solution distance scoring method in step S3 includes the following steps: Step S3-1: Determine whether the solution type weight in the cost function satisfies ω m ≥0, if not satisfied, a new cost function is established, in which ω m The value of is the absolute value of the original weight, and the value of scheme type m is the value after the original value is not calculated. The form of the cost function, the meaning of the variables and other weights remain unchanged. Step S3-2: The total travel time T of each plan in the travel plan set i , total cost P i and solution type m i The three indicators were normalized and converted using the weight method. Step S3-3: Calculate the weighted distance of each solution from the positive ideal solution and the negative ideal solution in the solution space and Where T i '、P i ′、m′ i are the total time, total cost and solution type value of the i-th solution after normalization, T′ max , P′ max 、m′ max , T′ min , P′ min 、m′ min are the maximum and minimum values ​​of the total time, total cost and solution type of each solution after normalization; Step S3-4: Calculate the score of each solution. The score of the i-th solution is S i The calculation formula is:

8. A device for recommending personalized travel plans for the entire process of air-rail intermodal travel, used to execute the method of claims 1-7, characterized in that: include: The user terminal unit is used to receive and display the dynamic questionnaire questions pushed by the preference analysis unit, and send the selection results to the preference analysis unit; a preference analysis unit for analyzing preferences of users of the user terminal unit; A plan generating unit, configured to receive a plan query request sent by a user terminal unit and generate a feasible travel plan; A plan recommendation unit is used to score and sort the travel plans generated by the plan generation unit and send them to the user terminal unit; The user terminal unit is further configured to send a travel plan query request to the plan generating unit and to display the travel plan result sent by the plan recommending unit; The preference analysis unit is further configured to send dynamic questionnaire questions to the user terminal unit and receive returned results; The solution generation unit is also used to match qualified air-rail transport airports; The plan generating unit is further used to calculate the departure time, estimated arrival time, total time, and total cost indicators of the plan, and to screen the plans.

Images