CN109034456B - Method, system, server and computer readable storage medium for scheduling vehicles - Google Patents

Abstract

The application provides a vehicle dispatching method, a system, a server and a computer readable storage medium. The vehicle scheduling method comprises the steps of obtaining the driving time and the driving cost of a driving route between any two areas in a period of time; the region is determined in a partition based on the geographic location; determining the benefits of a first driving route between a first area and a second area in all areas according to the probability distribution of the user vehicle demands on the driving route between any two areas in all areas, and the vehicle price of the first driving route in the time period; sending order information containing fees in the time period to a driver user located in at least one of the first area and the second area, or/and providing preview information containing fees in the time period to a riding user located in the first area or the second area; the cost is obtained based on the vehicle price adjustment.

Description

Method, system, server and computer readable storage medium for scheduling vehicles

Technical Field

The present application relates to the field of data processing technologies, and in particular, to a vehicle dispatching method, a system, a server, and a computer readable storage medium.

Background

With the development of mobile communication technology, order service based on the mobile internet has been widely used, such as meal order service or taxi (car-taking) service. Generally, for a customer or passenger order, the system typically processes the order separately, such as the passenger initiating a drive-up request, the system generates a drive-up order in response to the request, and issues the order to the driver client based on surrounding vehicle conditions, the driver client receiving the drive-up order for the passenger by clicking a confirmation button. However, the problem of unbalanced supply and demand in areas is increasingly prominent due to the influence of the peaks in the morning and evening, weather, and the like. In order to alleviate the unbalance of supply and demand, the driving platform predicts the vehicle demand of the area in a future period of time, and schedules the vehicle to the area with the supply and demand in advance.

In the related art, when a platform schedules vehicles from other areas, one situation is to adopt a form scheduling mode, so that users in areas with large demand are charged with the cost generated by scheduling, which necessarily reduces the willingness of some people to get on a vehicle and is not beneficial to alleviating the people evacuation efficiency in areas with large demand. In another case, a mode of forced dispatching of vehicles by the taxi taking platform is adopted, so that people in corresponding areas can be evacuated, but the cost generated by dispatching is transferred to the taxi taking platform and drivers.

Disclosure of Invention

In view of the above-mentioned drawbacks of the prior art, an object of the present application is to provide a vehicle dispatching method, system, server and computer readable storage medium, so as to solve the problem that the vehicle dispatching means in the prior art cannot be coordinated with the vehicle cost.

To achieve the above and other related objects, a first aspect of the present application provides a vehicle dispatching method, including the steps of: acquiring the driving time and the driving cost of a driving route between any two areas in a period of time; the region is determined in a partition based on the geographic location; determining the benefits of a first driving route between a first area and a second area in all areas according to the probability distribution of the user vehicle demands on the driving route between any two areas in all areas, and the vehicle price of the first driving route in the time period; sending order information containing fees in the time period to a driver user located in at least one of the first area and the second area, or/and providing preview information containing fees in the time period to a riding user located in the first area or the second area; the cost is obtained based on the vehicle price adjustment.

In certain implementations of the first aspect, the driving cost includes at least one of an energy consumption cost, a vehicle depreciation cost, a road bridge cost, and a labor cost.

In certain embodiments of the first aspect, the step of determining the benefit of the first driving route according to the probability distribution of the user's demand for use on the first driving route between the first area and the second area in all areas, and the price of use of the first driving route in the period of time includes: adding a virtual area between the first area and the second area according to the driving time or the driving cost and the probability distribution to form a second driving route connected by the first area, the virtual area and the second area; wherein the benefit of the second travel route is equivalent to the benefit of the first travel route; and obtaining probability distribution of the user vehicle demand on the first driving route so as to predict the vehicle price of the first driving route in the time period.

In certain implementations of the first aspect, the step of obtaining a probability distribution of user vehicular demand on the first vehicular route includes: acquiring historical order data generated by the first driving route in a period of time, wherein the historical order data comprises a vehicle price; preprocessing the historical order data to obtain order data to be fitted; and fitting the order data to be fitted by adopting a preset fitting model, and determining the probability distribution of the vehicle demands on the driving route between the first area and the second area according to the vehicle price.

In certain implementations of the first aspect, the historical order data further includes: one or more of order number, user number, driver number, start point, end point, estimated price of the order, and order generation time stamp information.

In certain implementations of the first aspect, the fitting model is a lognormal fitting function model, and the step of fitting the order data to be fitted with the preset fitting model includes fitting the order data to be fitted with the preset lognormal fitting function model, and determining a lognormal probability distribution of the user demand between the first area and the second area according to the price.

In certain implementations of the first aspect, the step of preprocessing the historical order data includes: performing data improvement processing on historical order data containing driver numbers; and removing invalid historical order data according to a preset field in the historical order data.

In certain implementations of the first aspect, the step of adjusting the cost based on the vehicle price includes: increasing a price based on the vehicle price to obtain the cost or decreasing a price based on the vehicle price to obtain the cost.

In certain implementations of the first aspect, raising a price based on the vehicle price to obtain the fee includes at least one of adding a price, dispensing a subsidy, and awarding a point; reducing the price based on the vehicle price to obtain the fee includes at least one of reducing the price, issuing a coupon, and awarding a point.

A second aspect of the present application provides a vehicular scheduling system comprising: the acquisition module is used for acquiring the driving time and the driving cost of the driving route between any two areas in a period of time; the region is determined in a partition based on the geographic location; the prediction module is used for determining the benefits of the first driving route between the first area and the second area in all areas and the price of the first driving route in the time period according to the probability distribution of the user's driving demand on the driving route between any two areas in all areas; the sending module is used for sending order information containing fees in the time period to a driver user positioned in at least one of the first area and the second area, or/and providing preview information containing fees in the time period to a riding user positioned in the first area or the second area; the cost is obtained based on the vehicle price adjustment.

In certain embodiments of the second aspect, the driving cost includes at least one of energy consumption cost, vehicle depreciation cost, road and bridge cost, and labor cost.

In certain embodiments of the second aspect, the prediction module is configured to add a virtual area between the first area and the second area according to the driving time or the driving cost and the probability distribution to form a second driving route connected by the first area, the virtual area and the second area; wherein the benefit of the second travel route is equivalent to the benefit of the first travel route; and obtaining probability distribution of the user vehicle demand on the first driving route so as to predict the vehicle price of the first driving route in the time period.

In certain implementations of the second aspect, the prediction module is configured to perform: acquiring historical order data generated by the first driving route in a period of time, wherein the historical order data comprises a vehicle price; preprocessing the historical order data to obtain order data to be fitted; and fitting the order data to be fitted by adopting a preset fitting model, and determining the probability distribution of the vehicle demands on the driving route between the first area and the second area according to the vehicle price.

In certain embodiments of the second aspect, the historical order data further comprises: one or more of order number, user number, driver number, start point, end point, estimated price of the order, and order generation time stamp information.

In certain embodiments of the second aspect, the fitting model is a lognormal fitting function model, and the step of fitting the order data to be fitted with a preset fitting model is to fit the order data to be fitted with a preset lognormal fitting function model, and determine a lognormal probability distribution of user vehicle demands in the at least two geographical areas according to prices.

In certain implementations of the second aspect, the prediction module is configured to sort orders in the historical order data according to order generation times to determine completion times for each order; and removing the abnormal order according to the vehicle price to obtain the order data to be fitted.

In certain embodiments of the second aspect, the cost is a cost after raising a price based on the price of the vehicle; or a cost after reducing the price based on the price of the vehicle.

In certain embodiments of the second aspect, raising a price based on the vehicle price to obtain the fee includes at least one of adding a price, dispensing a subsidy, and awarding a point; or reducing the price based on the vehicle price to obtain the fee includes at least one of reducing the price, issuing a coupon, and awarding a point.

A third aspect of the present application provides a server comprising: a memory for storing program code; one or more processors; wherein the processor is configured to invoke the program code stored in the memory to perform the ride-on scheduling method of any of the first aspects above.

A fourth aspect of the application provides a computer readable storage medium having instructions stored therein which, when run on a computer, cause the computer to perform the ride-on scheduling method of any one of the first aspects above.

As described above, the application realizes a price-based vehicle dispatching scheme by generating the vehicle price of the corresponding driving route by considering the vehicle demand of the user and the driving cost of the driver and adjusting the vehicle price according to the dispatching strategy, thereby effectively solving the problem that the vehicle dispatching means in the prior art cannot be coordinated with the driving cost.

Drawings

FIG. 1 is a flow chart illustrating a method for scheduling vehicles according to the present application in one embodiment.

Fig. 2 shows a first driving route between the first area a and the second area B, and virtual areas C1 and C2 and a second driving route disposed between the first area a and the second area B.

Fig. 3 shows a flow chart of a vehicle scheduling method according to the application in a further embodiment.

Fig. 4 is a statistical diagram showing statistics of order data to be fitted from the first area to the second area in order from small to large according to the order price corresponding to each time interval.

Fig. 5 shows a function curve of a log-normal fit function model obtained by fitting the statistics in fig. 4.

Fig. 6 is a schematic diagram of an architecture of a vehicle dispatching system according to an embodiment of the application.

Fig. 7 is a schematic structural diagram of a server according to an embodiment of the present application.

Detailed Description

Further advantages and effects of the present application will become apparent to those skilled in the art from the disclosure of the present application, which is described by the following specific examples.

In the following description, reference is made to the accompanying drawings which describe several embodiments of the application. It is to be understood that other embodiments may be utilized and that mechanical, structural, electrical, and operational changes may be made without departing from the spirit and scope of the present application. The following detailed description is not to be taken in a limiting sense, and the scope of embodiments of the present application is defined only by the claims of the issued patent.

Furthermore, as used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context indicates otherwise. It will be further understood that the terms "comprises," "comprising," "includes," and/or "including" specify the presence of stated features, steps, operations, elements, components, items, categories, and/or groups, but do not preclude the presence, presence or addition of one or more other features, steps, operations, elements, components, items, categories, and/or groups. The terms "or" and/or "as used herein are to be construed as inclusive, or meaning any one or any combination. Thus, "A, B or C" or "A, B and/or C" means "any of the following: a, A is as follows; b, a step of preparing a composite material; c, performing operation; a and B; a and C; b and C; A. b and C). An exception to this definition will occur only when a combination of elements, functions, steps or operations are in some way inherently mutually exclusive.

Stadiums, convention centers, transportation hubs, and areas near corporate parks are areas where there is a shortage of vehicular demand and vehicular supply. Taking a concert as an example, a stadium is used, when the concert ends, a large number of listeners concentrate to fade out, which easily results in that nearby public traffic cannot be dispersed quickly. Currently, taxi platforms may employ forced dispatch capacity to help digest people clusters near stadium and reduce traffic pressure. In addition, a taxi taking platform represented by a network taxi taking platform also provides a taxi taking service to provide taxi taking service for users with urgent taxi taking demands.

However, the forced dispatch mode is prone to problems of overstocked or understocked taxi vehicles. The taxi calling service is not required by most users, and the problem of shortage of supply and demand of vehicles cannot be effectively solved by the method. Based on the above example and the prediction aspect of the supply and demand relation of vehicles to other areas, the application provides a vehicle dispatching method, which aims to dispatch vehicles between two areas by adjusting the vehicle prices between the areas with the vehicle demands larger than the vehicle supply and the areas with the vehicle demands smaller than the vehicle supply, thereby realizing that the areas with unbalanced supply and demand relation of the two vehicles adopts a vehicle getting-on mode to dispatch the vehicles and effectively solving the traffic dredging problem of the local areas.

Referring to fig. 1, a flow chart of a vehicle dispatching method according to an embodiment of the application is shown. The taxi dispatching method is mainly executed by a taxi taking platform. The taxi taking platform may include software and hardware in a computer device.

Here, the computer device includes, but is not limited to: a single server, a server cluster, a distributed server, a cloud server, or the like. The Cloud Service end comprises a Public Cloud (Public Cloud) Service end and a Private Cloud (Private Cloud) Service end, wherein the Public or Private Cloud Service end comprises Software-as-a-Service (SaaS), platform-as-a-Service (Platform-as-Service), infrastructure-as-a-Service (IaaS) and the like. The private cloud service end is, for example, an ali cloud computing service platform, an Amazon (Amazon) cloud computing service platform, a hundred degree cloud computing platform, a Tencel cloud computing platform and the like.

Here, the taxi taking platform includes, but is not limited to, at least one of: a taxi taking platform based on the Internet, a taxi dispatching platform and the like. Wherein the internet-based taxi taking platform comprises, but is not limited to, at least one of the following: a windward driving platform, a carpooling driving platform, a special driving platform and the like. The taxi taking platform further comprises a part for carrying out data analysis based on historical order data, a part for carrying out vehicle searching and sending order confirmation to a driver side based on a taxi taking request and the like.

In some embodiments, the taxi taking platform executes the taxi dispatching method of the application based on a preset time interval to conduct daily vehicle dispatching. In still other embodiments, the taxi platform schedules vehicles based on the taxi scheduling method of the present application when the vehicle demand is greater than the vehicle supply based on the region in which the event is located. Wherein the event includes, but is not limited to, at least one of: isolated events such as singing events, exhibition events, traffic faults and the like; tidal regular events such as commute, up-down, etc.

The taxi taking platform utilizes the taxi taking scheduling method to schedule a plurality of areas in a region. For convenience of description, any two areas or preset two areas to be scheduled in a region are referred to as a first area and a second area in the present application, and the region may further include a third area, a fourth area, and so on. For example, the taxi taking platform utilizes the taxi taking scheduling method to schedule the vehicles between any two areas divided in the preset area. And in addition, the taxi taking platform performs the vehicle dispatching between any two areas in the plurality of areas to be dispatched by using the taxi dispatching method.

In step S110, the driving time and the driving cost of the driving route between any two areas in a period of time are obtained; the region is determined zonally based on geographic location.

The taxi taking platform obtains a first taxi route, a taxi time and a taxi cost between all the areas to be scheduled in a region in the corresponding historical time period according to a mapping relation between a pre-configured time period to be scheduled and the historical time period.

The mapping relation can be configured in a software program by using a configuration file, and can be reflected in search conditions for acquiring various information by the taxi-taking platform according to a default programming rule. For example, according to the time period to be scheduled, the taxi taking platform correspondingly searches the historical data of the same time period. The mapping relationship should be understood broadly, and may be any time period (such as three weeks before) before, and may also include, but is not limited to: the time period to be scheduled is divided from the historical time period based on a moment, a time interval is reserved between the time period to be scheduled and the historical time period, the time period to be scheduled is consistent with the historical time period, and the historical time period corresponding to the time period to be scheduled is constructed based on an event. For example, if the time period to be scheduled by the taxi taking platform is a time period T1 taking the current time as the starting time, the historical time period for acquiring the first taxi route, the taxi time and the taxi cost is a time period T1' before the current time. For another example, if the time period to be scheduled by the taxi taking platform is a future rush hour period, the taxi taking platform obtains a first taxi route, a taxi taking time, a taxi taking cost and the like in rush hour periods of a plurality of workdays. For another example, if the time period to be scheduled by the taxi taking platform is a future concert departure time period, the taxi taking platform obtains a first taxi route, a taxi time, a taxi cost and the like in the departure time period based on the historical multiple concert.

The area to be scheduled is obtained by predicting various data generated by the driving platform according to a historical time period corresponding to the time period to be scheduled. Wherein at least one of the areas to be scheduled includes an area where the vehicle demand is greater than the vehicle supply (or referred to as a high demand area), and/or an area where the vehicle demand is less than the vehicle supply (or referred to as a low demand area). For example, the first region and the second region in the region to be scheduled may each be a high demand region. For another example, the first region and the second region are regions of low demand. For another example, one of the first region and the second region is a low demand region and the other is a high demand region.

Wherein, in some specific examples, the high demand area is based on a determination of the current demand determination. For example, if the time period to be scheduled by the taxi taking platform is a time period started at the current time, the corresponding first area and second area are determined based on the geographical position distribution of the start point and the end point in the taxi taking request acquired in a historical time period from the current time to the current time.

In yet another specific example, the high demand area is determined based on historical order data. For example, the time period to be scheduled by the taxi taking platform is two hours before the start of the concert, and a plurality of corresponding areas are determined based on historical order data, which is stored in a database, of the same place as the concert and of the first two hours, wherein the plurality of areas comprise the area where the place is located.

Wherein the historical order data is generated based on the acquired various types of taxi taking requests. The request to drive up includes, but is not limited to, at least one of: a carpool request, a taxi calling request and a windward taxi calling request. The taxi taking request comprises a starting point, a destination point, a taxi taking time and the like. The generated historical order data includes, but is not limited to, vehicle price, start point, end point, and travel time, etc., and may also include at least one of order number, user number, driver number, etc. The vehicle price comprises an order estimated price, an order actual price and the like. The driving time comprises order generation time, vehicle reservation time, order completion time and the like. For example, a user operates a mobile communication device to send a taxi request to the taxi taking platform and complete payment when a vehicle provided by the taxi taking platform travels to a destination, and the taxi taking platform generates order data. For another example, the user operates the mobile communication device to send a taxi request to the taxi taking platform, and the taxi taking platform actively withdraws after the driver receives the order, so that order data are generated by the taxi taking platform. For another example, the user operates the mobile communication device to send only a request for previewing the taxi-taking price to the taxi-taking platform, so as to expect to view the taxi-taking price, and the taxi-taking platform also generates order data.

Here, in some embodiments, each of the regions may be pre-divided based on a geographic location. In other embodiments, the taxi taking platform obtains all the first taxi routes based on the time period determined by any one of the above methods, and performs clustering processing based on the start point and the end point of each first taxi route, so as to obtain a plurality of areas. The clustering processing includes dividing geographic positions of which the counted distances between starting points and/or ending points on different first driving routes are smaller than a preset distance threshold into a region. For example, the taxi taking platform selects any two areas with different supply and demand relations as a first area and a second area. For another example, the taxi taking platform selects the area and other areas as the first area and the second area according to the area where the geographic position of the generated event is located.

And determining the time period to be scheduled and each area according to any mode, and acquiring a first driving route, driving time and driving cost between the first area and the second area by the driving platform. For example, the taxi taking platform obtains all the first taxi routes with the start point located in the first area and the end point located in the second area, and the data such as the taxi time and the taxi cost corresponding to each first taxi route based on the time period determined by any one of the above modes. For another example, the taxi taking platform obtains all the first taxi routes with the start point located in the second area and the end point located in the first area, and the data such as the taxi time and the taxi cost corresponding to each first taxi route based on the time period determined by any one of the above modes. For another example, the taxi taking platform obtains each data of which the start point is located in the first area and the end point is located in the second area, and obtains each data of which the start point is located in the second area and the end point is located in the first area based on the time period determined by any one of the above modes.

In some embodiments, the taxi taking platform directly obtains the data such as the first taxi route, the taxi taking time and the taxi taking cost from the database between the first area and the second area.

In still other embodiments, the taxi taking platform obtains historical order data between the first area and the second area in a corresponding time period, and obtains data such as a first taxi route, a taxi time, a taxi cost and the like between the first area and the second area by processing the historical order data. The taxi taking platform counts a first taxi route from a first area to a second area and a first taxi route from the second area to the first area based on the starting point and the ending point in each acquired historical order data. And the taxi taking platform counts the driving time and the driving cost from the first area to the second area and/or the driving time and the driving cost from the second area to the first area according to the estimated price of the order or the actual price of the order in each historical order data. In some cases, the taxi taking platform may further obtain road condition information of a corresponding time period, and calculate the driving time and driving cost from the first area to the second area and/or the driving time and driving cost from the second area to the first area by combining the road condition information and the historical order data.

In another embodiment, the driving platform obtains a current time period to be scheduled, a driving request between the first area and the second area, and counts driving time and driving cost from the first area to the second area, and from the second area to the first area according to current road condition information. For example, the taxi taking platform determines a first taxi route, a taxi taking time and a taxi taking cost between a first area and a second area in a time period to be scheduled from the current moment based on a recently acquired taxi taking request.

Wherein the driving cost comprises at least one of energy consumption cost, vehicle depreciation cost, road and bridge cost and labor cost. The driving platform can calculate driving cost unit price according to the at least one driving cost, and determine respective driving costs from the first area to the second area and from the second area to the first area according to the first driving route and driving time between the first area and the second area.

In order to obtain the price of the vehicle which meets the requirements of the passenger user and prompts the driver user to receive the schedule, the taxi taking platform determines the income of a first taxi route between a first area and a second area in all areas according to the probability distribution of the user's demand on the taxi route between any two areas in all areas, and the price of the first taxi route in the time period.

Wherein the probability distribution of the vehicle demand is obtained by counting historical order data in advance. The probability distribution may be described by a curve function. For example, the probability distribution is described by a lognormal distribution function. The taxi taking platform calculates the price of the scheduled taxi taking which is caused to be received by a driver user under the condition that the user requirement is met according to probability distribution. Here, the taxi taking platform can determine the income between any two areas according to the driving direction and the corresponding price of the vehicle.

Wherein the historical order data used to obtain the probability distribution may be independent of or related to the historical order data used in step S110. The probability distribution may be statistically derived based on all of the second historical order data in the database.

In some embodiments, the probability distribution of vehicle demand is based on second historical order data statistics generated in an area where the driving route start and stop points are located. For example, the probability distribution of the vehicle demand includes: a probability distribution determined by counting second historical order data from the first region to the second region, and a probability distribution determined by counting second historical order data from the first region to the second region.

For this purpose, reference is made to fig. 3, which shows a flow chart of a further embodiment of the vehicle scheduling method according to the application, wherein the probability distribution of the vehicle demands in step S120 shown in fig. 3 is obtained by steps S101, S102, S103. It should be noted that, the taxi taking platform obtains probability distribution of the taxi taking requirement in advance. Wherein, according to a specific implementation, the steps S101-S103 may have no necessary timing relationship with the step S110. For example, the taxi taking platform counts the probability distribution of the vehicle demands of any two regions having the driving directions according to regions divided in advance based on the geographic positions, so that when the first region and the second region are determined in step S110, the taxi taking platform obtains the probability distribution from the first region to the second region and from the second region to the first region. As another example, the taxi taking platform obtains probability distributions from the first region to the second region and from the second region to the first region by performing steps S101 to S103 according to the first region and the second region determined in step S110.

In step S101, second historical order data generated in a period of time between the first region and the second region is acquired, wherein the second historical order data includes a vehicle price. Wherein the second historical order data may be to include search criteria including a second region of the first region. Wherein the time field in the acquired second historical order data constitutes a corresponding time period. The second historical order data also includes search conditions for historical time periods corresponding to time periods to be scheduled.

Those skilled in the art will appreciate that the manner in which the second historical order data is obtained is by way of example only and is not limiting. Setting search conditions according to the data analysis requirements of the subsequent steps, and further obtaining second historical order data generated in a period of time. In fact, each piece of second historical order data obtained may be the complete second historical order data in the database, or may be second historical order data obtained by selecting according to the search condition.

In step S102, the second historical order data is preprocessed to obtain order data to be fitted. Here, the acquired second historical order data includes order data having complete field information and order data having incomplete field information. Wherein the order data with incomplete field information is exemplified by, but not limited to, at least one of the following: order data lacking a driver number, order data lacking a completion time, order data lacking an end point, order data having an actual driving range which does not conform to an end point of a starting point corresponding to a vehicle-calling request, and the like. To this end, the acquired second historical order data is pre-processed to at least one of filter, supplement, and modify all of the second historical order data to obtain fitted order data available for fitting processing.

In some embodiments, the taxi-taking platform performs data refinement processing on the second historical order data containing the driver number. Wherein the data perfecting process comprises data supplementation and/or data modification. Here, the second historical order data including the driver number refers to order data with a driver order, and generally, after the driver takes the order, the driver takes the passenger on line according to the start and stop point indicated by the order to complete the order. However, there are some exceptions, such as passengers do not agree with orders, but agree with drivers separately to end points, and only by means of the orders, the vehicle price in the order data differs greatly from the estimated price of the order. For another example, when the passenger cancels the order after the driver takes the order, the time length between the time stamp generated in the order data and the time stamp completed in the order data is far smaller than the time length corresponding to the actual journey. For another example, the driver forgets to click the order completion button after completing the order, resulting in no completion time stamp in the order data.

In some specific examples, the taxi-taking platform orders the orders in the second historical order data containing the same driver number according to the order generation time; and supplementing completion time of the at least one order based on the ordering order. Here, for the situations that the order data with no completion time is included, or the order data with the completion time in the previous order data being greater than the order generation time in the next order data is included, the taxi taking platform orders the same driver number according to the order generation time, and supplements the completion time of the order in the previous order data according to the order generation time in the next order data. For example, the order generation time in the next order data is used as the completion time of the order in the previous order data, and the completion time field in the previous order data is updated. For another example, subtracting a preset taxi taking time interval from the completion time in the previous order data to obtain the completion time in the previous order data and updating the completion time field of the previous order data, wherein the taxi taking time interval can be a time interval or other preset value for obtaining an average taxi taking of a driver through data statistics.

In still other embodiments, the taxi taking platform rejects invalid second historical order data according to a preset field in the second historical order data. Wherein, in some specific examples, the preset field may be only a single field. For example, the taxi taking platform eliminates order data of unmanned order taking according to a driver number field. In still other specific examples, the taxi-taking platform preset determines the invalid second historical order data from a combination of the plurality of fields in the order data based on the plurality of fields. Wherein the combination of the plurality of fields includes, but is not limited to: a combination of at least two of a start point, an end point, an order pre-estimated price, an order generation time, and a completion time field. For example, the taxi-taking platform determines according to the estimated price of the order, the order generation time and the order completion time respectively: an order with the estimated price higher than a preset price and the completion time lower than a preset time is taken as an invalid order; the order with the estimated price lower than a preset price and the completion time higher than a preset time is used as the invalid order. For another example, the taxi taking platform determines according to the starting point, the ending point, the order generation time and the completion time fields: an order whose estimated travel duration differs from the actual order duration by at least n times (n > 1) is taken as an invalid order. Here, the taxi taking platform rejects the determined invalid second historical order data.

In yet another embodiment, the taxi-taking platform categorizes the second historical order data to fit a probability distribution of user vehicular demand under different monovalent mechanisms. The taxi taking platform can firstly screen and perfect the second historical order data according to the implementation mode, and then execute the classifying treatment; the categorization process may also be performed before screening and refining the second historical order data that cannot be categorized or in each category.

In order to grasp the vehicle demands of users with similar starting points and similar ending points more accurately, the taxi taking platform screens second historical order data which are positioned in the same time interval, the starting point (or ending point) in the first area and the ending point (or starting point) in the second area from the second historical order data as order data to be fitted.

The same time interval can be divided according to the time interval corresponding to the unit price; the time interval may be divided based on a preset unit price interval. For example, the vehicle price section is set to be an overlapping-free section of [ a- Δ, a+Δ ], and the time section corresponding to each vehicle price section is set to be the same time section, where a is the vehicle price, Δ is the section threshold value at which the vehicle price is floating up and down, and Δ may be a fixed value or determined based on other fields in the second historical order data (e.g., the Δ is determined based on the travel determined by the same starting point range and the same ending point range, etc.). The same time interval may be obtained by dividing the time period of each acquired second historical order data according to a preset time interval.

Here, the determining manner of the first area and the second area may refer to step S110, which is not described herein.

In step S103, a preset fitting model is adopted to fit the order data to be fitted, and a probability distribution of the vehicle demand on the first vehicle route between the first area and the second area is determined according to the vehicle price.

Here, the driving platform counts the order data to be fitted from step S102 based on the unit price (or unit price interval) corresponding to each time interval. And carrying out fitting processing by statistically selecting a fitting model similar to the statistical graphic trend so as to obtain probability distribution of the user vehicle demand on the first vehicle route between at least two areas, wherein the probability distribution is used for conveniently predicting the change of the user vehicle demand on the first vehicle route between the corresponding two areas reflected by the second historical order data when the unit price is adjusted, so that the vehicle problem of the supply and demand unmatched areas is solved based on the user vehicle supply and demand level.

In some embodiments, the taxi taking platform further performs the step of characterizing the probability distribution of the demand for the vehicle on the first route between the first region and the second region by a function curve. Here, the taxi taking platform characterizes the statistical taxi taking data on the first taxi route between the two areas according to the taxi price by using a function curve of probability distribution of taxi demands described by the fitted lognormal fitting function model. The taxi taking platform can also display the function curve and the corresponding taxi taking statistical data so as to enable technicians to check the fitting effect.

The step S103 includes fitting the order data to be fitted by using a preset lognormal fitting function model, and determining lognormal probability distribution of user vehicle demands in the at least two areas according to prices.

Referring to fig. 4 and 5, fig. 4 is a statistical diagram of statistics of order data to be fitted from a first area to a second area according to a sequential arrangement of vehicle prices corresponding to each time interval from small to large, and fig. 5 is a function curve of a log-normal fit function model obtained by fitting the statistical data in fig. 4, wherein an abscissa of the function curve is characterized by the vehicle prices, and an ordinate of the function curve is characterized by the probability of the vehicle demands of the first area to the second area. The price of the vehicle in each same time interval reflects the unit price of each time interval because the travel routes are similar, and each columnar graph in the diagram can be regarded as the number of passengers or the proportion of the number of passengers from the first area to the second area under each unit price. Taking the statistical diagram shown in fig. 4 as an example, selecting a lognormal fitting function model for fitting, wherein parameters to be determined in the lognormal fitting function model are preset, training the parameters to be determined by using the order data to be fitted so that the lognormal fitting function model constructed by the selected parameters reaches an optimal condition relative to the statistical data fitting degree in the statistical diagram, wherein the optimal condition includes but is not limited to: the error is smaller than a preset error range, etc. The probability distribution of the demand of the vehicle from the first region to the second region, which corresponds to fig. 4, obtained by fitting at different vehicle prices can be represented as a function curve in fig. 5. And predicting the probability of the vehicle demand corresponding to the vehicle price to be scheduled according to the obtained log-normal fit function model of each pair of the first area and the second area.

From this, and more generally, the taxi-taking platform can construct probability distribution of the direction of travel based on any two areas to be scheduled, so as to determine the following price based on different vehicles: probability distribution of user vehicular demand from low demand region to high demand region, and probability distribution of user vehicular demand from high demand region to low demand region. Wherein the taxi taking platform uses a demand function to describe the fitted probability distribution.

In step S120, the benefits of the first driving route between the first area and the second area in all the areas and the price of the first driving route in the time period are determined according to the probability distribution of the user' S demand on the driving route between any two areas in all the areas.

The taxi taking platform determines the taxi price by integrally evaluating the benefits of the taxi route between the areas to be scheduled in a region or evaluating the benefits of the divergent or convergent route taking an area as a starting point or an ending point according to the scheduling requirements. For example, the taxi taking platform obtains the profits and the price of the first taxi taking route between the first area and the second area by using the transfer function of the traffic volume and the local profits of any area arranged on the taxi taking route between the areas. The transfer function comprises parameters such as running time, running cost, probability distribution and the like of the first running route.

In some embodiments, the taxi taking platform adds a virtual area between the first area and the second area according to the taxi taking time or the taxi taking cost so as to form a second taxi path connected by the first area, the virtual area and the second area; wherein the benefit of the second travel route is equivalent to the benefit of the first travel route. And calculating the benefits and the prices of the vehicles of the second driving route by utilizing the added virtual area and the corresponding second driving route, thereby obtaining the benefits and the prices of the vehicles of the corresponding first driving route.

The taxi taking platform determines second taxi taking routes which are equivalent to the total income of the first taxi taking routes between any two areas in all areas to be scheduled and are connected between any two areas and virtual areas positioned on the second taxi taking routes according to preset unit parameters, and the obtained second taxi taking routes connected between the first area and the second area are regarded as having equivalent income with the first taxi taking routes connected between the first area and the second area. Wherein the unit parameters include, but are not limited to: one or more of unit travel time, unit travel cost, unit benefit. In addition, according to parameters such as driving time and driving cost, the virtual area on the second driving route from the first area to the second area may be the same as or different from the virtual area on the second driving route from the second area to the first area. The parameters also include, but are not limited to, global parameters capable of reflecting vehicle density parameters of any area in a time period to be scheduled, probability distribution of user vehicle demands between any two areas, road condition information and the like, and the global parameters can be used for calculating an additional virtual area between any two areas in the whole area.

The virtual area is selected according to each road between the first area and the second area on the map. The virtual area forms a second route with the first area and the second area. The benefit of the second route formed based on the set virtual area and the benefit of the first driving route are determined for an optimization objective based on at least one of driving cost, driving profit, and throughput of the driving platform of the driver user or the driving platform in a period to be scheduled. In order to conveniently select virtual areas on each road between the first area and the second area, the taxi taking platform sets the virtual areas between the two areas according to preset unit benefits. For example, please refer to fig. 2, which shows a first driving route between the first area a and the second area B, and virtual areas C1 and C2 and a second driving route disposed between the first area a and the second area B. The driving platform determines that the passing time of the first driving route between the first area A and the second area B is 3 unit time through calculation, and sets two virtual areas C1 and C2 between the first area A and the second area B, so that the passing time from the first area A to the virtual area C1, the virtual areas C1 to C2 and the passing time from the virtual area C2 to the second area B are all 3 unit time.

In some embodiments, the taxi taking platform selects a virtual area for the optimization objective based on any one of a cost of driving, profit of driving, and time of driving of the driver user or the taxi taking platform. For example, the taxi taking platform selects to set at least one virtual area on at least one road according to each road between any two areas to be scheduled on the map and according to road condition information of each road in a historical time period corresponding to the time period to be scheduled, so that the taxi taking platform counts the deviation of the total cost of the first taxi taking route between any two areas and the total cost of the second taxi taking route formed by any two areas and the set virtual area within a preset cost error range, and the taxi taking platform regards the benefit of the first taxi taking route between the first area and the second area and the benefit of the second taxi taking route between the first area and the second area as equivalent.

In other embodiments, the taxi-taking platform selects a virtual region for the optimization objective based on a linear weighted combination of at least two of a cost of driving, profit of driving, time of driving, probability distribution, and throughput of the taxi-taking platform for the driver user or the taxi-taking platform. The taxi taking platform selects a second driving route between any two areas in all areas to be scheduled according to a decision model which takes unit parameters as step length in advance, so that deviation between a profit target corresponding to all the second driving routes and a profit target corresponding to all the first driving routes in a time period to be scheduled meets a preset optimization condition, wherein the taxi taking platform sets a corresponding virtual area according to the step length. The taxi taking platform determines the virtual area arranged between the first area and the second area in the mode. Wherein the decision model comprises a Markov Decision Process (MDP) model or the like. The optimization conditions include, but are not limited to: at least one of the deviation is minimum and the deviation is within a preset error range.

Wherein the benefits used in the decision model may be calculated using a linear weighted combination of at least two of the acquired travel time and travel cost, travel profits, probability distribution, and throughput. The decision model can also be constructed based on parameters such as empty vehicle density distribution, action space, state transition rules and the like which are subjected to statistical prediction.

Taking a Markov decision process model as an example, describing each second driving route which predicts the equivalent benefits of the first driving route among all areas within the dividing range of all areas to be scheduled, and the price of the vehicle of each first driving route under the condition of meeting the benefits. The preset Markov decision process model is expressed as (G, D, G, S, A, W). Wherein G represents an input map; d represents a demand function, i.e. an estimated number of passengers or orders at a certain moment (departure time) willing to accept a certain price to get in from one area to another; g represents a reward function, i.e. an optimization objective; s represents a state space, namely all possible density distribution of empty vehicles on the map; a represents an action space; w represents a state transition rule: the empty density of one zone at this time minus the sum of the flows from that zone to the other zones plus the sum of the flows from the other zones to that zone equals the empty density of the point at the next time. Since the scheduling strategy requires determining how many vehicles to schedule from one area to another, the pricing strategy requires determining the prices for vehicles on different routes. Wherein each of the vehicle prices includes a vehicle price from the first area to the second area. To reduce the variables, the Markov decision process model describes the scheduling policy and the pricing policy in terms of traffic. The preset action space A is all possible A 'of road running of the vehicles to be scheduled among all areas, and the reward function is replaced by the optimized objective function G' after the salifying treatment, so that the new MDP is (G, D, G ', S, A', W). And determining each second driving route and each virtual area which can be equivalent to each first driving route by using the constructed MDP. Wherein in the new MDP (G, D, G ', S, A ', W), G ' is a concave function and the transfer rule is linear, so that the sum of the flow from one area to the other area at a certain moment of time of the taxi taking platform is smaller than the empty density of the point or the area at the moment of time to limit the condition I; and calculating the vehicle price and the scheduling strategy of the objective function MDP (G, D, G ', S, A', W) when the limit condition I, II is met by subtracting the empty vehicle density of one area at the moment from the sum of the flow of the other area from the point and adding the sum of the flow of the other area to the first area to be equal to the empty vehicle density of one area at the next moment. Wherein the vehicle price comprises a vehicle price from the first area to the second area and a vehicle price from the second area to the first area. The scheduling policy includes, but is not limited to: the area (including the virtual area and the area to be scheduled) in which the order information is transmitted is selected based on the area to be scheduled, and the cost in the order information.

The cost may be the resulting price of the vehicle. In some specific examples, the taxi taking platform is further adjusted based on the resulting price of the vehicle, comprising: increasing the price based on the vehicle price or decreasing the price based on the vehicle price to obtain a fee that can be sent to the driver user and the ride user for viewing. Wherein raising a price based on the vehicle price to obtain the fee includes at least one of adding a price, issuing a subsidy, and awarding a point; reducing the price based on the vehicle price to obtain the fee includes at least one of reducing the price, issuing a coupon, and awarding a point. For example, when the first area is in high demand and the second area is in low demand in the time period to be scheduled, the taxi taking platform increases the proportion of the monovalent price based on the obtained price of the first taxi route from the first area to the second area, and provides at least part of the increased price to the taxi taking user in a subsidized manner (such as an electronic coupon). For another example, when the first area is in low demand and the second area is in high demand in the time period to be scheduled, the taxi taking platform reduces the proportion of the monovalent price based on the obtained taxi price of the first taxi route from the first area to the second area, and provides at least part of the reduced price for the driver user in a subsidy mode (such as electronic coupons).

It should be noted that the above examples are only examples, and the rule that the technician may set the taxi taking platform to adjust the price of the vehicle according to the scheduling policy is not described herein one by one.

In step S130, order information including a fee in the time period is sent to a driver user located in at least one of the first area and the second area, or/and preview information including a fee in the time period is provided to a riding user located in the first area or the second area; the cost is obtained based on the vehicle price adjustment.

The taxi taking platform determines the journey information from the first area to the second area in the time period to be scheduled by analyzing the taxi taking request from the taxi taking user or the preview request of the driver user, and sends order information containing the calculated cost (the taxi price if no price adjustment is made) to the mobile communication device of the driver user (or feeds back to the driver user) located in the first area and at least one virtual area nearest to the first area. Wherein, the order information also comprises a destination point, even a recommended route and the like according to the type of the request. Wherein the recommended route includes a second driving route that runs along the virtual area. For example, when a riding user operates a mobile communication device to issue a request for getting a ride in order to expect a driver user to provide a riding service, the transmitting module pushes order information containing the fee to the driver user in an idle state located in at least one of the first area and the second area in order from the passenger's distance from near to far, to be accepted by the driver user. The driver user in the idle state is determined based on the fact that the attribute of the driver number which is registered by the transmitting module and uniquely corresponds to the driver user is idle, wherein the idle state indicates that the driver user is not in a passenger carrying state and can receive orders.

And the taxi taking platform determines the travel information from the first area to the second area in the time period to be scheduled by analyzing the taxi taking request from the taxi taking user, and feeds back the preview information containing the cost to the taxi taking user so as to enable the taxi taking user to preview the taxi taking cost. For example, when a ride user operates a mobile communication device to issue a request for a ride to expect a preview fee, the ride platform confirms whether the ride user further issues a request for a ride invitation by feeding back preview information containing the fee to the mobile communication device held by the corresponding ride user.

It should be noted that the foregoing examples are only for illustration and not limitation of the present application, and in fact, a riding user may get a car from the second area to the first area, and the car taking platform sends corresponding preview information and order information containing corresponding fees to the riding user and the driver user according to the calculated fee for getting the car from the second area to the first area. The above examples are further drawn to the scheduling of vehicles in the whole city or region, and the taxi taking platform can provide the scheduling of vehicles in the region where the starting point is located and the region where the ending point is located in any time period, which will not be described herein.

In summary, the vehicle price of the corresponding driving route is generated by considering the vehicle demand of the user and the driving cost of the driver, and the driving price is adjusted according to the scheduling strategy, so that the price-based vehicle scheduling scheme is realized, and the problem that the vehicle scheduling means cannot be coordinated with the driving cost in the prior art is effectively solved.

In addition, the vehicle demand prediction method provided by the application realizes the prediction of the vehicle demand by constructing the probability distribution of the user demand and the vehicle cost, namely, the price sensitivity of the user is analyzed by combining price factors when the demand is predicted, which is favorable for relieving unbalance of supply and demand, and effective available data is provided for pricing and scheduling of a driving platform, thereby solving the problem that the scheduling cost is not easy to control because the vehicle demand of the user is not considered in a mode of processing the supply and demand mismatch of the vehicle in the prior art.

The application also provides a vehicle dispatching system. The taxi dispatching system is software and hardware running in the taxi taking platform. The taxi taking platform utilizes the taxi dispatching system to dispatch a plurality of areas in a region. For convenience of description, any two regions in a region will be referred to as a first region and a second region, and the region may further include a third region, a fourth region, and so on. For example, the taxi taking platform utilizes the taxi dispatching system to dispatch the vehicles between any two areas divided in the preset area. And in addition, the taxi taking platform utilizes the taxi taking scheduling system to schedule the vehicles between any two areas in the plurality of areas to be scheduled.

The car dispatching system provided by the application is arranged in the car taking platform and is used for dispatching among a plurality of areas in a region. For convenience of description, any two areas or preset two areas to be scheduled in a region are referred to as a first area and a second area in the present application, and the region may further include a third area, a fourth area, and so on. For example, the vehicle dispatching system provided by the application is arranged in the taxi taking platform and is used for dispatching vehicles between any two areas divided in a preset area. In another example, the vehicle dispatching system provided by the application is used for dispatching vehicles between any two areas in a plurality of areas to be dispatched.

Referring to fig. 6, a schematic diagram of an architecture of a vehicle dispatching system according to an embodiment of the application is shown. The vehicle dispatching system comprises: an acquisition module 11, a prediction module 12 and a transmission module 13. The modules in the car dispatching system can be executed by hardware such as a memory, a processor, a network interface and the like in the operation server based on a data flow.

The acquiring module 11 is configured to acquire a driving time and a driving cost of a driving route between any two areas in a period of time; the region is determined zonally based on geographic location.

The taxi taking platform obtains a first taxi route, a taxi time and a taxi cost between all the areas to be scheduled in a region in the corresponding historical time period according to a mapping relation between a pre-configured time period to be scheduled and the historical time period.

The mapping relationship may be configured in a software program by using a configuration file, or may be reflected in a search condition for acquiring each piece of information by the acquisition module 11 according to a default programming rule. For example, according to the time period to be scheduled, the obtaining module 11 correspondingly searches for the historical data of the same time period. The mapping relationship should be understood broadly, and may be any time period (such as three weeks before) before, and may also include, but is not limited to: the time period to be scheduled is divided from the historical time period based on a moment, a time interval is reserved between the time period to be scheduled and the historical time period, the time period to be scheduled is consistent with the historical time period, and the historical time period corresponding to the time period to be scheduled is constructed based on an event. For example, if the time period to be scheduled by the obtaining module 11 is a time period T1 with the current time as the starting time, the historical time period for obtaining the first driving route, the driving time and the driving cost is a time period T1' before the current time. For another example, if the time period to be scheduled by the obtaining module 11 is a future rush hour period, the obtaining module 11 obtains a first driving route, driving time, driving cost, etc. of rush hour periods of a plurality of working days. For another example, if the time period to be scheduled by the obtaining module 11 is a future concert diverging time period, the obtaining module 11 obtains a first driving route, driving time, driving cost, and the like in the concert diverging time period based on a plurality of historic singing.

The area to be scheduled is obtained by the obtaining module 11 according to various data predictions generated by a historical time period corresponding to the time period to be scheduled. Wherein at least one of the areas to be scheduled includes an area where the vehicle demand is greater than the vehicle supply (or referred to as a high demand area), and/or an area where the vehicle demand is less than the vehicle supply (or referred to as a low demand area). For example, the first region and the second region in the region to be scheduled may each be a high demand region. For another example, the first region and the second region are regions of low demand. For another example, one of the first region and the second region is a low demand region and the other is a high demand region.

Wherein, in some specific examples, the high demand area is based on a determination of the current demand determination. For example, the obtaining module 11 determines the corresponding first area and the second area based on the geographical location distribution of the start point and the end point in the driving request obtained in a historical period from the current time to the current time when the period to be scheduled is the period from the current time.

In yet another specific example, the high demand area is determined based on historical order data. For example, the time period to be scheduled by the obtaining module 11 is two hours before the start of the concert, and a plurality of corresponding areas are determined based on the historical order data stored in the database and having the same place as the concert and two hours before the start, wherein the plurality of areas include the area where the place is located.

Wherein the historical order data is generated based on the acquired various types of taxi taking requests. The request to drive up includes, but is not limited to, at least one of: a carpool request, a taxi calling request and a windward taxi calling request. The taxi taking request comprises a starting point, a destination point, a taxi taking time and the like. The generated historical order data includes, but is not limited to, vehicle price, start point, end point, and travel time, etc., and may also include at least one of order number, user number, driver number, etc. The vehicle price comprises an order estimated price, an order actual price and the like. The driving time comprises order generation time, vehicle reservation time, order completion time and the like. For example, the user operates the mobile communication device to send a taxi request to the acquisition module 11, and finishes payment while traveling to a destination while riding the vehicle provided by the acquisition module 11, the acquisition module 11 generates an order data. For another example, the user operates the mobile communication device to send a taxi request to the obtaining module 11, and actively cancel the taxi request after the driver receives the order, and the obtaining module 11 generates order data. For another example, the user operates the mobile communication device to send only a request for previewing the driving price to the acquisition module 11, so that the driving price is expected to be checked, and the acquisition module 11 also generates an order data.

Here, in some embodiments, each of the regions may be pre-divided based on a geographic location. In other embodiments, the obtaining module 11 obtains all the first driving routes based on the time period determined in any one of the foregoing manners, and performs clustering processing based on the start point and the end point of each of the first driving routes, so as to obtain a plurality of areas. The clustering processing includes dividing geographic positions of which the counted distances between starting points and/or ending points on different first driving routes are smaller than a preset distance threshold into a region. For example, the obtaining module 11 selects any two areas of different supply-demand relationships as the first area and the second area. As another example, the acquiring module 11 selects the area and other areas as the first area and the second area according to the area where the geographic location of the generated event is located.

The time period to be scheduled and each area are determined according to any mode, and the obtaining module 11 obtains a first driving route, driving time and driving cost between the first area and the second area. For example, the obtaining module 11 obtains, based on the time period determined in any one of the above manners, all the first driving routes having the start point located in the first area and the end point located in the second area, and the driving time and the driving cost corresponding to each first driving route. For another example, the obtaining module 11 obtains, based on the time period determined by any one of the above methods, all the first driving routes with the start point located in the second area and the end point located in the first area, and the driving time and the driving cost corresponding to each first driving route. For another example, the acquiring module 11 acquires each data having a start point located in the first area and an end point located in the second area, and acquires each data having a start point located in the second area and an end point located in the first area, based on the time period determined by any one of the above methods.

In some embodiments, the obtaining module 11 directly obtains the data such as the first driving route, the driving time, and the driving cost from the database between the first area and the second area.

In still other embodiments, the acquiring module 11 acquires the historical order data between the first area and the second area in the corresponding time period, and obtains the data such as the first driving route, the driving time, the driving cost and the like between the first area and the second area by processing the historical order data. Wherein the obtaining module 11 counts a first driving route from the first area to the second area and a first driving route from the second area to the first area based on the start point and the end point in each obtained historical order data. The acquiring module 11 counts the driving time and driving cost from the first area to the second area and/or the driving time and driving cost from the second area to the first area according to the estimated price of the order or the actual price of the order in each historical order data. In some cases, the obtaining module 11 may further obtain road condition information of a corresponding time period, and calculate the driving time and driving cost from the first area to the second area and/or the driving time and driving cost from the second area to the first area by combining the road condition information and the historical order data.

In another embodiment, the obtaining module 11 obtains a current waiting time period, a driving request from the first area to the second area, and statistics of driving time and driving cost from the first area to the second area, and from the second area to the first area according to current road condition information. For example, the obtaining module 11 determines a first driving route, a driving time and a driving cost between a first area and a second area in a time period to be scheduled from a current time based on a recently obtained driving request.

Wherein the driving cost comprises at least one of energy consumption cost, vehicle depreciation cost, road and bridge cost and labor cost. The obtaining module 11 may calculate a driving cost unit price according to the at least one driving cost, and determine respective driving costs from the first area to the second area and from the second area to the first area according to the first driving route and the driving time between the first area and the second area.

In order to obtain a price for a vehicle that both meets the needs of the passenger user and prompts the driver user to receive a schedule, the prediction module 12 determines the return of the first driving route between the first area and the second area in all areas and the price for the first driving route during the period of time based on the probability distribution of the user's demand for the vehicle on the driving route between any two areas in all areas.

Wherein the probability distribution of the vehicle demand is obtained by counting historical order data in advance. The probability distribution may be described by a curve function. For example, the probability distribution is described by a lognormal distribution function. The taxi taking platform calculates the price of the scheduled taxi taking which is caused to be received by a driver user under the condition that the user requirement is met according to probability distribution. Here, the taxi taking platform can determine the income between any two areas according to the driving direction and the corresponding price of the vehicle.

Wherein the historical order data used to obtain the probability distribution may be independent of or related to the historical order data used in step S110. In some embodiments, the probability distribution of vehicle demand is based on second historical order data statistics generated in an area where the driving route start and stop points are located. For example, the probability distribution of the vehicle demand includes: a probability distribution determined by counting second historical order data from the first region to the second region, and a probability distribution determined by counting second historical order data from the first region to the second region.

For this purpose, the prediction module 12 obtains the probability distribution of the vehicle demand via steps S101, S102, S103. The prediction module 12 obtains a probability distribution of the vehicle demand in advance. Wherein, according to a specific implementation, the prediction module 12 may not necessarily have a timing relationship with the acquisition module 11. For example, the prediction module 12 calculates the probability distribution of the vehicle demands of any two areas having the driving directions according to the areas divided in advance based on the geographical position, so that when the first area and the second area are determined in the acquisition module 11, the driving platform acquires the probability distribution from the first area to the second area and from the second area to the first area. As another example, the prediction module 12 obtains probability distributions from the first region to the second region and from the second region to the first region by performing steps S101 to S103 according to the first region and the second region determined in the obtaining module 11.

In step S101, second historical order data generated in a period of time between the first region and the second region is acquired, wherein the second historical order data includes a vehicle price. Wherein the second historical order data may be to include search criteria including a second region of the first region. Wherein the time field in the acquired second historical order data constitutes a corresponding time period. The second historical order data also includes search conditions for historical time periods corresponding to time periods to be scheduled.

Those skilled in the art will appreciate that the manner in which the second historical order data is obtained is by way of example only and is not limiting. Setting search conditions according to the data analysis requirements of the subsequent steps, and further obtaining second historical order data generated in a period of time. In fact, each piece of second historical order data obtained may be the complete second historical order data in the database, or may be second historical order data obtained by selecting according to the search condition.

In step S102, the second historical order data is preprocessed to obtain order data to be fitted. Here, the acquired second historical order data includes order data having complete field information and order data having incomplete field information. Wherein the order data with incomplete field information is exemplified by, but not limited to, at least one of the following: order data lacking a driver number, order data lacking a completion time, order data lacking an end point, order data having an actual driving range which does not conform to an end point of a starting point corresponding to a vehicle-calling request, and the like. To this end, the acquired second historical order data is pre-processed to at least one of filter, supplement, and modify all of the second historical order data to obtain fitted order data available for fitting processing.

In some embodiments, the prediction module 12 performs data refinement processing on the second historical order data containing the driver number. Wherein the data perfecting process comprises data supplementation and/or data modification. Here, the second historical order data including the driver number refers to order data with a driver order, and generally, after the driver takes the order, the driver takes the passenger on line according to the start and stop point indicated by the order to complete the order. However, there are some exceptions, such as passengers do not agree with orders, but agree with drivers separately to end points, and only by means of the orders, the vehicle price in the order data differs greatly from the estimated price of the order. For another example, when the passenger cancels the order after the driver takes the order, the time length between the time stamp generated in the order data and the time stamp completed in the order data is far smaller than the time length corresponding to the actual journey. For another example, the driver forgets to click the order completion button after completing the order, resulting in no completion time stamp in the order data.

In some specific examples, the prediction module 12 orders in the second historical order data containing the same driver number according to order generation time; and supplementing completion time of the at least one order based on the ordering order. Here, for the case of the order data including no completion time, or the order data having a completion time greater than the order generation time in the next order data, the prediction module 12 orders the same driver number according to the order generation time, and supplements the completion time of the order in the previous order data according to the order generation time in the next order data. For example, the order generation time in the next order data is used as the completion time of the order in the previous order data, and the completion time field in the previous order data is updated. For another example, subtracting a preset taxi taking time interval from the completion time in the previous order data to obtain the completion time in the previous order data and updating the completion time field of the previous order data, wherein the taxi taking time interval can be a time interval or other preset value for obtaining an average taxi taking of a driver through data statistics.

In still other embodiments, the prediction module 12 eliminates invalid second historical order data according to a preset field in the second historical order data. Wherein, in some specific examples, the preset field may be only a single field. For example, the prediction module 12 eliminates the order data of the unmanned order according to the driver number field. In still other specific examples, the prediction module 12 presets determining invalid second historical order data based on a combination of fields in the order data based on a plurality of fields. Wherein the combination of the plurality of fields includes, but is not limited to: a combination of at least two of a start point, an end point, an order pre-estimated price, an order generation time, and a completion time field. For example, the prediction module 12 determines from the order forecast price, the order generation time, and the order completion time, respectively: an order with the estimated price higher than a preset price and the completion time lower than a preset time is taken as an invalid order; the order with the estimated price lower than a preset price and the completion time higher than a preset time is used as the invalid order. As another example, the prediction module 12 determines from the start point, end point, order generation time, and completion time fields: an order whose estimated travel duration differs from the actual order duration by at least n times (n > 1) is taken as an invalid order. Here, the prediction module 12 rejects the determined invalid second historical order data.

In yet another embodiment, the prediction module 12 categorizes the second historical order data to fit a probability distribution of user vehicular demand under different monovalent mechanisms. Wherein, the prediction module 12 may first screen and perfect the second historical order data according to the foregoing embodiment, and then execute the classifying process; the categorization process may also be performed before screening and refining the second historical order data that cannot be categorized or in each category.

In order to more accurately grasp the vehicle demands of users with similar starting points and similar ending points, the prediction module 12 screens the second historical order data, which are located in the same time interval, the starting point (or ending point) in the first area, and the ending point (or starting point) in the second area, from the second historical order data as the order data to be fitted.

The same time interval can be divided according to the time interval corresponding to the unit price; the time interval may be divided based on a preset unit price interval. For example, the vehicle price section is set to be an overlapping-free section of [ a- Δ, a+Δ ], and the time section corresponding to each vehicle price section is set to be the same time section, where a is the vehicle price, Δ is the section threshold value at which the vehicle price is floating up and down, and Δ may be a fixed value or determined based on other fields in the second historical order data (e.g., the Δ is determined based on the travel determined by the same starting point range and the same ending point range, etc.). The same time interval may be obtained by dividing the time period of each acquired second historical order data according to a preset time interval.

Here, the determining manner of the first area and the second area may refer to step S110, which is not described herein.

In step S103, a preset fitting model is adopted to fit the order data to be fitted, and a probability distribution of the vehicle demand on the first vehicle route between the first area and the second area is determined according to the vehicle price.

Here, the prediction module 12 counts the order data to be fitted from step S102 based on the unit price (or unit price interval) corresponding to each time interval. And carrying out fitting processing by statistically selecting a fitting model similar to the statistical graphic trend so as to obtain probability distribution of the user vehicle demand on the first vehicle route between at least two areas, wherein the probability distribution is used for conveniently predicting the change of the user vehicle demand on the first vehicle route between the corresponding two areas reflected by the second historical order data when the unit price is adjusted, so that the vehicle problem of the supply and demand unmatched areas is solved based on the user vehicle supply and demand level.

In some embodiments, the prediction module 12 further performs the step of characterizing the probability distribution of the demand for vehicles on the first route between the first region and the second region by a function curve. Here, the prediction module 12 characterizes the statistical representation of the function curve of the probability distribution of the vehicle demand described by the fitted lognormal fit function model as a function of the vehicle price to determine the driving data on the first driving route between the two regions. The prediction module 12 may also display the function curve and the corresponding driving statistics for a technician to check the fitting effect.

The step S133 includes fitting the order data to be fitted by using a preset lognormal fitting function model, and determining a lognormal probability distribution of the user' S vehicle demands in the at least two areas according to the price.

Referring to fig. 4 and 5, fig. 4 is a statistical diagram of statistics of order data to be fitted from a first area to a second area according to a sequential arrangement of vehicle prices corresponding to each time interval from small to large, and fig. 5 is a function curve of a log-normal fit function model obtained by fitting the statistical data in fig. 4, wherein an abscissa of the function curve is characterized by the vehicle prices, and an ordinate of the function curve is characterized by the probability of the vehicle demands of the first area to the second area. The price of the vehicle in each same time interval reflects the unit price of each time interval because the travel routes are similar, and each columnar graph in the diagram can be regarded as the number of passengers or the proportion of the number of passengers from the first area to the second area under each unit price. Taking the statistical diagram shown in fig. 4 as an example, selecting a lognormal fitting function model for fitting, wherein parameters to be determined in the lognormal fitting function model are preset, training the parameters to be determined by using the order data to be fitted so that the lognormal fitting function model constructed by the selected parameters reaches an optimal condition relative to the statistical data fitting degree in the statistical diagram, wherein the optimal condition includes but is not limited to: the error is smaller than a preset error range, etc. The probability distribution of the demand of the vehicle from the first region to the second region, which corresponds to fig. 4, obtained by fitting at different vehicle prices can be represented as a function curve in fig. 5. And predicting the probability of the vehicle demand corresponding to the vehicle price to be scheduled according to the obtained log-normal fit function model of each pair of the first area and the second area.

It follows that, more generally, the prediction module 12 can construct a probability distribution of driving directions based on any two areas to be scheduled, to determine, based on different vehicle prices: probability distribution of user vehicular demand from low demand region to high demand region, and probability distribution of user vehicular demand from high demand region to low demand region. Wherein the prediction module 12 uses a demand function to describe the fitted probability distribution.

The prediction module 12 is configured to determine a benefit of a first driving route between a first area and a second area in all areas according to a probability distribution of a user's demand on the driving route between any two areas in all areas, and a price of the first driving route in the time period.

The taxi taking platform determines the taxi price by integrally evaluating the benefits of the taxi route between the areas to be scheduled in a region or evaluating the benefits of the divergent or convergent route taking an area as a starting point or an ending point according to the scheduling requirements. For example, the prediction module 12 obtains the profit and the price of the vehicle for the first driving route between the first area and the second area by using the transfer function of the traffic volume and the local profit of any area set on the driving route between the two areas. The transfer function comprises parameters such as running time, running cost, probability distribution and the like of the first running route.

In some embodiments, the prediction module 12 adds a virtual area between the first area and the second area according to the driving time or the driving cost to form a second driving route connected by the first area, the virtual area and the second area; wherein the benefit of the second travel route is equivalent to the benefit of the first travel route. And calculating the benefits and the prices of the vehicles of the second driving route by utilizing the added virtual area and the corresponding second driving route, thereby obtaining the benefits and the prices of the vehicles of the corresponding first driving route.

Here, the prediction module 12 determines, according to preset unit parameters, a second driving route that is equivalent to the total profit of the first driving route between any two areas in all the areas to be scheduled and connects between any two areas, and a virtual area located on each second driving route, and regards the obtained second driving route that connects between the first area and the second area as having equivalent profit to the first driving route that also connects between the first area and the second area. Wherein the unit parameters include, but are not limited to: one or more of unit travel time, unit travel cost, unit benefit. In addition, according to parameters such as driving time and driving cost, the virtual area on the second driving route from the first area to the second area may be the same as or different from the virtual area on the second driving route from the second area to the first area. The parameters also include, but are not limited to, global parameters capable of reflecting vehicle density parameters of any area in a time period to be scheduled, probability distribution of user vehicle demands between any two areas, road condition information and the like, and the global parameters can be used for calculating an additional virtual area between any two areas in the whole area.

The virtual area is selected according to each road between the first area and the second area on the map. The virtual area forms a second route with the first area and the second area. The benefit of the second route formed based on the set virtual area and the benefit of the first driving route are determined for the optimization target based on at least one of driving cost, driving profit of the driver user or the prediction module 12, and throughput of the prediction module 12 during the period to be scheduled. In order to facilitate selection of the virtual area on each road between the first area and the second area, the prediction module 12 sets the virtual area between the two areas according to a preset unit benefit. For example, referring to fig. 2, a first driving route from a first area a to a second area B, and virtual areas C1 and C2 and a second driving route disposed between the first area a and the second area B are shown. The prediction module 12 determines that the traffic time of the first driving route between the first area a and the second area B is 3 unit times, and the driving platform sets two virtual areas C1, C2 between the first area a and the second area B such that the traffic time from the first area a to the virtual area C1, the virtual areas C1 to C2, and the virtual area C2 to the second area B is 3 unit times.

In some embodiments, the prediction module 12 selects a virtual area for the optimization objective based on any of the cost of driving, profit of driving, and time of driving of the driver user or the prediction module 12. For example, the prediction module 12 selects to set at least one virtual area on at least one road according to each road between any two areas to be scheduled on the map and according to the road condition information of each road in the historical time period corresponding to the time period to be scheduled, so that the deviation between the total cost of the first driving route between any two areas and the total cost of the second driving route formed by any two areas and the set virtual area is counted by the prediction module 12 to be within the preset cost error range, and the prediction module 12 regards the benefit of the first driving route between the first area and the second area and the benefit of the second driving route between the first area and the second area as equivalent.

In other embodiments, the prediction module 12 selects the virtual region for the optimization objective based on a linear weighted combination of at least two of the cost of driving, profit of driving, time of driving, probability distribution, and throughput of the taxi-platform for the driver user or the prediction module 12. The taxi taking platform selects a second driving route between any two areas in all areas to be scheduled according to a decision model which takes unit parameters as step length in advance, so that deviation between a profit target corresponding to all the second driving routes and a profit target corresponding to all the first driving routes in a time period to be scheduled meets a preset optimization condition, wherein the taxi taking platform sets a corresponding virtual area according to the step length. The taxi taking platform determines the virtual area arranged between the first area and the second area in the mode. Wherein the decision model comprises a Markov Decision Process (MDP) model or the like. The optimization conditions include, but are not limited to: at least one of the deviation is minimum and the deviation is within a preset error range. Wherein the benefit used in the decision model may be calculated using a linear weighted combination of at least two of the acquired travel time and travel cost, travel profits, and throughput.

The decision model can also be constructed based on parameters such as empty vehicle density distribution, action space, state transition rules and the like which are subjected to statistical prediction.

Taking a Markov decision process model as an example, describing each second driving route which predicts the equivalent benefits of the first driving route among all areas within the dividing range of all areas to be scheduled, and the price of the vehicle of each first driving route under the condition of meeting the benefits. The preset Markov decision process model is expressed as (G, D, G, S, A, W). Wherein G represents an input map; d represents a demand function, i.e. an estimated number of passengers or orders at a certain moment (departure time) willing to accept a certain price to get in from one area to another; g represents a reward function, i.e. an optimization objective; s represents a state space, namely all possible density distribution of empty vehicles on the map; a represents an action space; w represents a state transition rule: the empty density of one zone at this time minus the sum of the flows from that zone to the other zones plus the sum of the flows from the other zones to that zone equals the empty density of the point at the next time. Since the scheduling strategy requires determining how many vehicles to schedule from one area to another, the pricing strategy requires determining the prices for vehicles on different routes. Wherein each of the vehicle prices includes a vehicle price from the first area to the second area. To reduce the variables, the Markov decision process model describes the scheduling policy and the pricing policy in terms of traffic. The preset action space A is all possible A 'of road running of the vehicles to be scheduled among all areas, and the reward function is replaced by the optimized objective function G' after the salifying treatment, so that the new MDP is (G, D, G ', S, A', W). And determining each second driving route and each virtual area which can be equivalent to each first driving route by using the constructed MDP. Wherein in the new MDP (G, D, G ', S, a ', W), G ' is a concave function and the transfer rule is linear, so that the prediction module 12 limits the condition I that the sum of the flows from one area to the other area at a time will be less than the empty density of that point or area at that time; and calculating the vehicle price and the scheduling strategy of the objective function MDP (G, D, G ', S, A', W) when the limit condition I, II is met by subtracting the empty vehicle density of one area at the moment from the sum of the flow of the other area from the point and adding the sum of the flow of the other area to the first area to be equal to the empty vehicle density of one area at the next moment. Wherein the vehicle price comprises a vehicle price from the first area to the second area and a vehicle price from the second area to the first area. The scheduling policy includes, but is not limited to: the area (including the virtual area and the area to be scheduled) in which the order information is transmitted is selected based on the area to be scheduled, and the cost in the order information.

The cost may be the resulting price of the vehicle. In some specific examples, the taxi taking platform is further adjusted based on the resulting price of the vehicle, comprising: increasing the price based on the vehicle price or decreasing the price based on the vehicle price to obtain a fee that can be sent to the driver user and the ride user for viewing. Wherein raising a price based on the vehicle price to obtain the fee includes at least one of adding a price, issuing a subsidy, and awarding a point; reducing the price based on the vehicle price to obtain the fee includes at least one of reducing the price, issuing a coupon, and awarding a point. For example, when the first area is in high demand and the second area is in low demand in the time period to be scheduled, the taxi taking platform increases the proportion of the monovalent price based on the obtained price of the first taxi route from the first area to the second area, and provides at least part of the increased price to the taxi taking user in a subsidized manner (such as an electronic coupon). For another example, when the first area is in low demand and the second area is in high demand in the time period to be scheduled, the taxi taking platform reduces the proportion of the monovalent price based on the obtained taxi price of the first taxi route from the first area to the second area, and provides at least part of the reduced price for the driver user in a subsidy mode (such as electronic coupons).

It should be noted that the above examples are only examples, and the rule that the technician may set the taxi taking platform to adjust the price of the vehicle according to the scheduling policy is not described herein one by one.

The sending module 13 is configured to send order information including a fee in the time period to a driver user located in at least one of the first area and the second area, or/and provide preview information including a fee in the time period to a riding user located in the first area or the second area; the cost is obtained based on the vehicle price adjustment.

The transmitting module 13 determines travel information from the first area to the second area in a time period to be scheduled by analyzing a taxi taking request from a passenger or a preview request of a driver user, and transmits order information including the calculated fee (a taxi price if there is no price adjustment) to a mobile communication device of the driver user (or feeds back to the driver user) located in the first area and at least one virtual area nearest to the first area. Wherein, the order information also comprises a destination point, even a recommended route and the like according to the type of the request. For example, when a riding user operates a mobile communication device to issue a request for getting a ride in order to expect a driver user to provide a riding service, the transmission module pushes order information containing the fee to the driver user in an idle state located in at least one of the first area, the second area, and the virtual area in order from near to far from the passenger to wait for the driver user to accept. The driver user in the idle state is determined based on the fact that the attribute of the driver number which is registered by the transmitting module and uniquely corresponds to the driver user is idle, wherein the idle state indicates that the driver user is not in a passenger carrying state and can receive orders.

The sending module 13 determines the travel information from the first area to the second area in the time period to be scheduled by analyzing the taxi taking request from the taxi taking user, and feeds back the preview information containing the fee to the taxi taking user so as to enable the taxi taking user to preview the taxi taking fee. For example, when the riding user operates the mobile communication device to issue a request for getting a car for a preview fee, the transmitting module 13 confirms whether the riding user further issues the request for getting a car invitation by feeding back preview information containing the fee to the mobile communication device held by the corresponding riding user.

It should be noted that the foregoing examples are only for illustration and not limitation of the present application, and in fact, the riding user may get a car from the second area to the first area, and the sending module 13 sends corresponding preview information and order information containing corresponding fees to the riding user and the driver user according to the calculated fee for getting a car from the second area to the first area. The above examples are further drawn to the scheduling of vehicles in the whole city or region, and the sending module 13 may provide the scheduling of vehicles in the region where the starting point is located and the region where the ending point is located in any time period, which will not be described herein.

The application also provides a server. The server is used for running the taxi taking platform disclosed by the application and any taxi taking platform capable of executing the taxi taking demand prediction method. Referring to fig. 7, a schematic diagram of the structure of the server in an embodiment is shown. The server includes a memory, and one or more processors.

The memory 31 may include, among other things, high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic disk storage devices, flash memory devices, or other non-volatile solid-state storage devices. In some embodiments, the memory may also include memory remote from the one or more processors, such as network-attached memory accessed via a communication network (not shown), which may be the Internet, one or more intranets, a Local Area Network (LAN), a wide area network (WLAN), a Storage Area Network (SAN), etc., or a suitable combination thereof. The memory also includes memory controller that can control access to the memory by other components of the device, such as the CPU and peripheral interfaces. The memory is used for storing program codes.

The processor 32 is operatively coupled to the memory 31. The processor may execute program code stored in the memory 31, such as steps of data reception and transmission with the mobile communication device, and steps of calculating a first car share price and a second car share price from the trip information, etc. More specifically, the processor is configured to invoke program code stored in the memory to perform a ride-on scheduling method. For example, the processor executes the vehicle dispatching method designed with reference to fig. 1 and the corresponding text descriptions, and will not be described herein. As such, the processor may include one or more general purpose microprocessors, one or more application specific processors (ASICs), one or more field programmable logic arrays (FPGAs), or any combinations thereof.

The processor 32 is also operably coupled to a network interface to communicatively couple the subject computing device to a network. For example, the network interface may connect the computing device to a wide area network (WAN, or injected 4G, 5G, or LTE cellular network).

It should be further noted that, from the above description of the embodiments, it will be apparent to those skilled in the art that part or all of the present application may be implemented by means of software in combination with a necessary general hardware platform. Based on such understanding, the present application also provides a computer readable storage medium having stored therein at least one program or instruction which, when executed on a computer, causes the computer to perform any of the aforementioned methods of scheduling use of a vehicle.

Based on such understanding, the technical solutions of the present application may be embodied essentially or in part in the form of a software product, which may include one or more machine-readable media having stored thereon machine-executable program code which, when executed by one or more machines such as a computer, computer network, or other electronic device, may cause the one or more machines to perform operations in accordance with embodiments of the present application. For example, performing steps in a call service, etc. The machine-readable medium may include, but is not limited to, floppy diskettes, optical disks, CD-ROMs, magneto-optical disks, ROMs (read-only memory), RAMs (random access memory), EPROMs (erasable programmable read-only memory), EEPROMs (electrically erasable programmable read-only memory), magnetic or optical cards, flash memory, or other type of media/machine-readable medium suitable for storing machine-executable program code. Wherein the storage medium may be located on a machine or in a third party server, such as in a server providing cloud storage.

It should be further noted that, from the above description of the embodiments, it will be apparent to those skilled in the art that part or all of the present application may be implemented by means of software in combination with a necessary general hardware platform. Based on such understanding, the present application also provides a storage medium of a computer device, the storage medium storing at least one program that, when called, performs any of the aforementioned vehicular scheduling methods.

Based on such understanding, the technical solutions of the present application may be embodied essentially or in part in the form of a software product, which may include one or more machine-readable media having stored thereon machine-executable program code which, when executed by one or more machines such as a computer, computer network, or other electronic device, may cause the one or more machines to perform operations in accordance with embodiments of the present application. For example, performing steps in a call service, etc. The machine-readable medium may include, but is not limited to, floppy diskettes, optical disks, CD-ROMs, magneto-optical disks, ROMs (read-only memory), RAMs (random access memory), EPROMs (erasable programmable read-only memory), EEPROMs (electrically erasable programmable read-only memory), magnetic or optical cards, flash memory, or other type of media/machine-readable medium suitable for storing machine-executable program code. The storage medium can be located in the robot or a third-party server, such as a cloud server. The specific application mall is not limited herein, such as an alicloud, a cloud server for Huazhi, and the like.

The application may be described in the general context of computer-executable program code, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

The above embodiments are merely illustrative of the principles of the present application and its effectiveness, and are not intended to limit the application. Modifications and variations may be made to the above-described embodiments by those skilled in the art without departing from the spirit and scope of the application. Accordingly, it is intended that all equivalent modifications and variations of the application be covered by the claims, which are within the ordinary skill of the art, be within the spirit and scope of the present disclosure.

Claims (16)

1. The vehicle dispatching method is characterized by comprising the following steps of:

acquiring the driving time and the driving cost of a driving route between any two areas in a period of time; the region is determined in a partition based on the geographic location;

Determining the benefits of the first driving route between the first area and the second area in all areas according to the probability distribution of the user vehicle demands on the driving route between any two areas in all areas, and the vehicle price of the first driving route in the time period, wherein the method comprises the following steps: adding a virtual area between the first area and the second area according to the driving time or the driving cost and the probability distribution to form a second driving route connected by the first area, the virtual area and the second area; wherein the benefit of the second travel route is equivalent to the benefit of the first travel route; obtaining probability distribution of user vehicle demands on the first driving route so as to predict vehicle prices of the first driving route in the time period; wherein the probability distribution of the user vehicular demand is associated with a vehicular price in the historical order data of the first vehicular route; the step of obtaining the probability distribution of the user vehicle demand on the first driving route comprises the following steps: acquiring historical order data generated by the first driving route in a period of time, wherein the historical order data comprises a vehicle price; preprocessing the historical order data to obtain order data to be fitted; fitting the order data to be fitted by adopting a preset fitting model, and determining probability distribution of vehicle demands on a driving route between the first area and the second area according to the vehicle price;

sending order information containing fees in the time period to a driver user located in at least one of the first area and the second area, or/and providing preview information containing fees in the time period to a riding user located in the first area or the second area; the cost is obtained based on the vehicle price adjustment.

2. The ride-on scheduling method of claim 1, wherein the driving cost comprises at least one of an energy consumption cost, a vehicle depreciation cost, a road-bridge cost, and a labor cost.

3. The method of scheduling use of claim 1, wherein the historical order data further comprises: one or more of order number, user number, driver number, start point, end point, estimated price of the order, and order generation time stamp information.

4. The vehicle dispatching method according to claim 1, wherein the fitting model is a lognormal fitting function model, and the step of fitting the order data to be fitted by using a preset fitting model is to fit the order data to be fitted by using a preset lognormal fitting function model, and determine a lognormal probability distribution of user vehicle demands between the first area and the second area according to prices.

5. The method of scheduling vehicles according to claim 1, wherein the step of preprocessing the historical order data comprises:

performing data improvement processing on historical order data containing driver numbers; and

And removing invalid historical order data according to a preset field in the historical order data.

6. The vehicle scheduling method of claim 1, wherein the step of adjusting the cost based on the vehicle price comprises: increasing a price based on the vehicle price to obtain the cost or decreasing a price based on the vehicle price to obtain the cost.

7. The ride-scheduling method of claim 6, wherein raising a price based on the ride price to obtain the fee comprises at least one of adding a price, issuing a subsidy, and a point prize; reducing the price based on the vehicle price to obtain the fee includes at least one of reducing the price, issuing a coupon, and awarding a point.

8. A vehicular scheduling system, comprising:

The acquisition module is used for acquiring the driving time and the driving cost of the driving route between any two areas in a period of time;

The region is determined in a partition based on the geographic location;

The prediction module is used for determining the benefits of the first driving route between the first area and the second area in all areas and the price of the first driving route in the time period according to the probability distribution of the user's driving demand on the driving route between any two areas in all areas; wherein the probability distribution of the user vehicular demand is associated with a vehicular price in the historical order data of the first vehicular route; the prediction module is used for adding a virtual area between the first area and the second area according to the driving time or the driving cost and the probability distribution so as to form a second driving route connected by the first area, the virtual area and the second area; wherein the benefit of the second travel route is equivalent to the benefit of the first travel route; obtaining probability distribution of user vehicle demands on the first driving route so as to predict vehicle prices of the first driving route in the time period; the prediction module is further configured to perform:

Acquiring historical order data generated by the first driving route in a period of time, wherein the historical order data comprises a vehicle price;

Preprocessing the historical order data to obtain order data to be fitted; and

Fitting the order data to be fitted by adopting a preset fitting model, and determining probability distribution of vehicle demands on a driving route between the first area and the second area according to the vehicle price; and

The sending module is used for sending order information containing fees in the time period to a driver user positioned in at least one of the first area and the second area, or/and providing preview information containing fees in the time period to a riding user positioned in the first area or the second area; the cost is obtained based on the vehicle price adjustment.

9. The ride vehicle scheduling system of claim 8, wherein the driving costs comprise at least one of energy costs, vehicle depreciation costs, road and bridge costs, and labor costs.

10. The ride vehicle scheduling system of claim 8, wherein the historical order data further comprises: one or more of order number, user number, driver number, start point, end point, estimated price of the order, and order generation time stamp information.

11. The vehicular scheduling system according to claim 8, wherein the fitting model is a lognormal fitting function model, and the step of fitting the order data to be fitted using a preset fitting model is to fit the order data to be fitted using a preset lognormal fitting function model, and determine a lognormal probability distribution of user vehicular demands in the at least two geographical areas according to prices.

12. The ride vehicle scheduling system of claim 8, wherein the prediction module is configured to sort orders in the historical order data according to order generation time to determine completion time for each order; and removing the abnormal order according to the vehicle price to obtain the order data to be fitted.

13. The vehicle scheduling system according to claim 8, wherein the fee is a fee after raising a price on the basis of the vehicle price; or a cost after reducing the price based on the price of the vehicle.

14. The ride vehicle scheduling system of claim 13, wherein raising a price based on the vehicle price to obtain the fee comprises at least one of adding a price, issuing a subsidy, and a point prize; or reducing the price based on the vehicle price to obtain the fee includes at least one of reducing the price, issuing a coupon, and awarding a point.

15. A server, comprising:

A memory for storing program code;

one or more processors;

Wherein the processor is configured to invoke program code stored in the memory to perform the ride-on scheduling method of any of claims 1-7.

16. A computer readable storage medium having instructions stored therein which, when run on a computer, cause the computer to perform the ride-on scheduling method of any one of the preceding claims 1-7.