Method and system for controlling surge pricing for transportation request to transportation service network
The method and system optimize surge pricing by considering pick-up and drop-off locations and real-time market conditions, enhancing driver utilization and revenue optimization in transportation services.
Patent Information
- Application Number
- PCT/CN2024/123448
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-10-08
- Publication Date
- 2026-04-16
AI Technical Summary
Conventional surge pricing methods for transportation services, such as ride-hailing, are inefficient as they do not account for drop-off locations and rely solely on historical data, leading to inefficiencies in driver utilization and revenue optimization.
A method and system that considers both pick-up and drop-off locations, along with real-time market conditions, to dynamically adjust surge pricing by using a booking conversion control parameter to optimize network efficiency and revenue.
Improves efficiency in transportation networks by optimizing driver utilization and revenue through dynamic surge pricing adjustments based on current demand and booking probabilities.
Smart Images

Figure CN2024123448_16042026_PF_FP_ABST
Abstract
Description
METHOD AND SYSTEM FOR CONTROLLING SURGE PRICING FOR TRANSPORTATION REQUEST TO TRANSPORTATION SERVICE NETWORKTECHNICAL FIELD
[0001] The present invention generally relates to a method of controlling a surge pricing for a transportation request to a transportation service network, and a system thereof, such as for a ride-hailing service or a delivery service.BACKGROUND
[0002] In on-demand transportation service industries, such as the ride-hailing service industry, a price surge engine plays a critical role to dynamically adjust a surge pricing (e.g., a price surge parameter or multiplier / factor) for a transportation request on a spatial-temporal basis (which may also be referred to as an area-time basis) to control demand on the transportation service network (e.g., a ride-hailing service network) . Conventionally, a surge pricing for a transportation request (which includes a price check, such as a fare check for a ride-hailing service) may be computed based on only a pick-up location (e.g., a geographic location or region, such as represented by a geohash (e.g., geohash 6) ) of the transportation and a time period (e.g., over a 1-minute time period) . In this regard, a surge pricing may thus be computed per geohash for each time period (e.g., for each 1-minute time period) . Therefore, the same surge pricing may be computed for all transportation requests with the same pick-up location (e.g., same geohash) within the same time period. However, such a conventional price surging technique for determining the surge pricing for a transportation request to a transportation service network may be inefficient (or may not be optimal) for the transportation network since it determines the same surge pricing for transportation requests as long as they belong to the same pick-up location and the same time period, regardless of other characteristics or parameters of the transportation requests. For example, in practice, there may be a large number of transportation requests with the same pick-up location within the same time period but having various other characteristics or parameters (e.g., different drop-off locations) .
[0003] There has also been disclosed a price surging technique which additionally takes into account the drop-off location of the transportation request, which may be referred to as the dynamic drop-off surcharge (DDoS) technique. Compared to the above-mentioned conventional price surging technique, this DDoS technique adjusts the surge pricing further based on the drop-off location so as to promote or encourage the conversion of transportation requests to actual bookings for transportation requests with a drop-off location having associated therewith a shorter driver idle / waiting time based on historical data. However, the DDoS technique still suffers from various drawbacks / deficiencies, resulting in inefficiencies in the transportation service network. For example, the DDoS technique adjusts the surge pricing further based on the drop-off location of the transportation request by purely relying on historical data (e.g., past eight weeks of historical data) , and thus does not take into account current or real-time market conditions (e.g., demand condition fluctuations) . Therefore, the manner in which the DDoS technique adjusts / controls the surge pricing can be ineffective.
[0004] A need therefore exists to provide a method of controlling a surge pricing for a transportation request to a transportation service network, as well as a system thereof, that seeks to overcome, or at least ameliorate, one or more deficiencies in conventional methods of controlling a surge pricing, and more particularly, with improved effectiveness in controlling the surge pricing, resulting in improved efficiency (e.g., with respect to driver utilization and revenue optimization) in the transportation service network. It is against this background that the present invention has been developed.SUMMARY
[0005] According to a first aspect of the present invention, there is provided a method of controlling a surge pricing for a transportation request to a transportation service network using at least one processor, the method comprising:
[0006] receiving a transportation request having associated therewith a pick-up location, a drop-off location and a pick-up time for a transportation;
[0007] obtaining a first network contribution measure for the transportation request with respect to the drop-off location, the first network contribution measure indicating a contribution measure of the transportation request to the transportation service network with respect to the drop-off location;
[0008] determining a base network improvement measure for the transportation request for using a base surge pricing based on the base surge pricing and the first network contribution measure for the transportation request;
[0009] setting a value of a booking conversion control parameter for the transportation request, the booking conversion control parameter being configured to control a threshold for implementing a surge pricing adjustment to the base surge pricing for the transportation request;
[0010] determining a candidate surge pricing for the transportation request based on the base surge pricing and the surge pricing adjustment for the transportation request, the surge pricing adjustment being determined based on the base network improvement measure and the value of the booking conversion control parameter for the transportation request;
[0011] determining a first booking probability for the transportation request for using the candidate surge pricing based on the pick-up location, the drop-off location and the pick-up time for the transportation and the candidate surge pricing for the transportation request;
[0012] determining a first booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing based on the first booking probability, the first network contribution measure and the candidate surge pricing for the transportation request; and
[0013] determining the surge pricing for the transportation request based on the value of the booking conversion control parameter and the first booking probability adjusted network improvement measure for the transportation request.
[0014] According to a second aspect of the present invention, there is provided a system for controlling a surge pricing for a transportation request to a transportation service network, the system comprising:
[0015] at least one memory; and
[0016] at least one processor communicatively coupled to the at least one memory and configured to:
[0017] receive a transportation request having associated therewith a pick-up location, a drop-off location and a pick-up time for a transportation;
[0018] obtain a first network contribution measure for the transportation request with respect to the drop-off location, the first network contribution measure indicating a contribution measure of the transportation request to the transportation service network with respect to the drop-off location;
[0019] determine a base network improvement measure for the transportation request for using a base surge pricing based on the base surge pricing and the first network contribution measure for the transportation request;
[0020] set a value of a booking conversion control parameter for the transportation request, the booking conversion control parameter being configured to control a threshold for implementing a surge pricing adjustment to the base surge pricing for the transportation request;
[0021] determine a candidate surge pricing for the transportation request based on the base surge pricing and the surge pricing adjustment for the transportation request, the surge pricing adjustment being determined based on the base network improvement measure and the value of the booking conversion control parameter for the transportation request;
[0022] determine a first booking probability for the transportation request for using the candidate surge pricing based on the pick-up location, the drop-off location and the pick-up time for the transportation and the candidate surge pricing for the transportation request;
[0023] determine a first booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing based on the first booking probability, the first network contribution measure and the candidate surge pricing for the transportation request; and
[0024] determine the surge pricing for the transportation request based on the value of the booking conversion control parameter and the first booking probability adjusted network improvement measure for the transportation request.
[0025] According to a third aspect of the present invention, there is provided a computer program product, embodied in one or more non-transitory computer-readable storage mediums, comprising instructions executable by at least one processor to perform the method of controlling a surge pricing for a transportation request to a transportation service network according to the above-mentioned first aspect of the present invention.BRIEF DESCRIPTION OF THE DRAWINGS
[0026] Embodiments of the present invention will be better understood and readily apparent to one of ordinary skill in the art from the following written description, by way of example only, and in conjunction with the drawings, in which:
[0027] FIG. 1 depicts a schematic flow diagram of a method of controlling a surge pricing for a transportation request to a transportation service network, according to various embodiments of the present invention;
[0028] FIG. 2 depicts a schematic block diagram of a system for controlling a surge pricing for a transportation request to a transportation service network, according to various embodiments of the present invention;
[0029] FIG. 3 depicts a schematic block diagram of an exemplary computer system which may be used to realize or implement the system for controlling a surge pricing for a transportation request to a transportation service network, according to various embodiments of the present invention;
[0030] FIG. 4 depicts a drawing illustrating the decomposition of the technical problem of how to control the surge pricing for a transportation request to maximize / optimize network efficiency into three sub-problems, according to various example embodiments of the present invention;
[0031] FIG. 5 depicts a tree diagram showing an overview of a method of controlling a surge pricing for a transportation request, according to various example embodiments of the present invention;
[0032] FIG. 6 depicts a schematic flow diagram of the method of controlling a surge pricing for a transportation request, according to various example embodiments of the present invention;
[0033] FIGs. 7A and 7B show plots of the booking conversion control parameter (αi) and the budget ratio for a PI (proportional-integral) controller and a PID (proportional-integral-derivative) controller to illustrate the rule-based online thresholding demand pacing, according to various example embodiments of the present invention;
[0034] FIG. 8 depicts a schematic flow diagram illustrating an example training of the booking probability model (which may be referred to as the BTR (booked through rate) model) , according to various example embodiments of the present invention;
[0035] FIG. 9 depicts a schematic flow diagram for an example simulation of the method of controlling a surge pricing for each time period, according to various example embodiments of the present invention; and
[0036] FIG. 10 illustrates example network value results obtained for geographic regions (e.g., geohashes) , according to various example embodiments of the present invention.DETAILED DESCRIPTION
[0037] Various embodiments of the present invention provide a method and a system for controlling a surge pricing for a transportation request to a transportation service network, such as for a ride-hailing service or a delivery service (e.g., a food delivery service) .
[0038] As explained in the background, conventional price surging techniques suffer from a number of drawbacks / deficiencies, resulting in inefficiencies in the transportation service network, such as but not limited to, poor or inefficient driver utilization and poor or ineffective revenue optimization (e.g., gross merchandise value (GMV) ) in the transportation service network. In this regard, various embodiments of the present invention provide a method of controlling a surge pricing for a transportation request to a transportation service network, as well as a system thereof, that seeks to overcome, or at least ameliorate, one or more deficiencies in conventional methods of controlling a surge pricing, and more particularly, with improved effectiveness in controlling the surge pricing, resulting in improved efficiency (e.g., with respect to driver utilization and revenue optimization) in the transportation service network.
[0039] FIG. 1 depicts a schematic flow diagram of a method 100 of controlling a surge pricing for a transportation request (e.g., for each transportation request) to a transportation service network using at least one processor, according to various embodiments of the present invention. The method 100 comprising: receiving (at 106) a transportation request having associated therewith a pick-up location, a drop-off location and a pick-up time for a transportation (i.e., a transportation request comprising pick-up location information, drop-off location information and pick-up time information for the transportation) ; obtaining (at 108) a first network contribution measure for the transportation request with respect to the drop-off location, the first network contribution measure indicating a contribution measure of the transportation request to the transportation service network with respect to the drop-off location; determining (at 110) a base network improvement measure for the transportation request for using a base surge pricing based on the base surge pricing and the first network contribution measure for the transportation request; setting (at 112) a value of a booking conversion control parameter for the transportation request, the booking conversion control parameter being configured to control a threshold for implementing a surge pricing adjustment to the base surge pricing for the transportation request; determining (at 114) a candidate surge pricing for the transportation request based on the base surge pricing and the surge pricing adjustment for the transportation request, the surge pricing adjustment being determined based on the base network improvement measure and the value of the booking conversion control parameter for the transportation request; determining (at 116) a first booking probability for the transportation request for using the candidate surge pricing based on the pick-up location, the drop-off location and the pick-up time for the transportation and the candidate surge pricing for the transportation request; determining (at 118) a first booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing based on the first booking probability, the first network contribution measure and the candidate surge pricing for the transportation request; and determining (at 120) the surge pricing for the transportation request based on the value of the booking conversion control parameter and the first booking probability adjusted network improvement measure for the transportation request.
[0040] It will be appreciated by a person skilled in the art that the present invention is not limited to any particular or specific type of contribution measure to the transportation service network, which may be defined or configured for measuring or representing a type of contribution to the transportation service network as desired or as appropriate, such as an efficiency of the transportation service network that may be desired to be improved or optimized to benefit the transportation service network. Accordingly, the contribution measure (and thus, the network contribution measure) may be a measure of network value (e.g., benefit) to the transportation service network. Therefore, for example, depending on the particular or specific type of efficiency of the transportation service network desired to be improved or optimized by controlling the surge pricing, the contribute measure (and thus, the network contribution measure) may be defined or configured as appropriate for measuring or representing such a type of efficiency of the transportation service network. As an illustrative example, in the case of the transportation service network being a ride-hailing service network, a type of efficiency of the transportation service network that may be improved or optimized by controlling the surge pricing may be with respect to driver utilization (e.g., maximizing driver in-transit time or minimizing driver idle / waiting time) in the ride-hailing service network. Therefore, depending on the particular or specific type of efficiency of the transportation service network desired to be improved or optimized, the manner of defining the contribution measure to represent such a type of efficiency may also be different. Therefore, it will be appreciated by a person skilled in the art that various contribution measures may be provided or defined as desired or as appropriate for measuring or representing various types of efficiencies of the transportation service network. In various embodiments, a contribution measure may also be provided or defined for measuring or representing a combination of two or more types of efficiencies of the transportation service network desired to be improved or optimized by controlling the surge pricing.
[0041] As explained above, the network contribution measure may be a measure of network value (e.g., benefit) to the transportation service network with respect to an efficiency of the transportation service network. Therefore, the contribution measure (and thus, the network contribution measure) may also be referred to, or regarded as, a network value. In various embodiments, for each time period or window (e.g., each 1-minute time period) , each geographical region (e.g., each geohash) may have a respective network value assigned thereto relating to an efficiency of the transportation service network. For example, referring back to the above illustrative example of the type of efficiency of the transportation service network being driver utilization in the ride-hailing service network, for each time period or window, each geographical region may have a respective network value assigned thereto indicating a degree or level of benefit of a driver being available to pick up passenger (s) at the geographical region at the time period (e.g., sending passenger (s) to the geographical region as the drop-off location and thus being available to pick up passenger (s) at the geographical region after drop-off) with respect to driver utilization of the transportation service network. For example, a high network value at a geographical region at a time period may indicate a high degree or level of benefit of a driver being available to pick up passenger (s) at the geographical region at the time period with respect to driver utilization of the transportation service network. In various embodiments, the degree or level of benefit to driver utilization is considered or measured over a predetermined time period, such as over the next two to six hours or one day. Therefore, the network value at a geographical region at a time period may indicate the degree of level of benefit of a driver being available to pick up passenger (s) at the geographical region at the time period with respect to driver utilization of the transportation service network considered or measured over the predetermined time period, such as over the next two to six hours or one day.
[0042] In various embodiments, the above-mentioned first network contribution measure for the transportation request with respect to the drop-off location may be obtained based on the network contribution measure (e.g., network value) assigned to the drop-off location at the drop-off time. In this regard, the drop-off time may be determined based on the pick-up time and the transit time (e.g., estimated) from the pick-up location to the drop-off location.
[0043] In various embodiments, the above-mentioned network improvement measure determined for the transportation request indicates an improvement measure to the transportation service network which takes into account a potential revenue generated by the transportation request in addition to the first network contribution measure for the transportation request. Therefore, the network improvement measure may be an overall improvement value determined for the transportation request to the transportation service network which takes into account the value from the first network contribution measure and the value from the potential revenue generated by the transportation request. Therefore, the above-mentioned base network improvement measure for the transportation request for using the base surge pricing may be determined based on the base surge pricing (or more specifically the potential revenue generated from the base fare and the base surge pricing) and the first network contribution measure for the transportation request.
[0044] In various embodiments, the surge pricing may be in the form of a price surge parameter or multiplier / factor.
[0045] Therefore, the method 100 of controlling a surge pricing for a transportation request according to various embodiments of the present invention advantageously has improved effectiveness in controlling the surge pricing, resulting in improved efficiency in the transportation service network. In particular, the base network improvement measure for the transportation request for using the base surge pricing is determined. In this regard, the method 100 does not merely determine the surge pricing according to the base network improvement measure (e.g., not merely the higher the base network improvement measure the lower the surge pricing) but further determines the surge pricing based on a booking conversion control parameter for the transportation request for controlling a threshold for implementing a surge pricing adjustment to the base surge pricing for the transportation request. In particular, the surge pricing adjustment (to the base surge pricing) is determined based on the base network improvement measure and the value of the booking conversion control parameter for the transportation request. Therefore, the value of the booking conversion control parameter for the transportation request can be dynamically set to modify the contribution of the base network improvement measure in determining the surge pricing adjustment. For example, the booking conversion control parameter may be configured to pace demand and the value thereof may be dynamically set to indicate a degree or level of desire to convert the transportation request to actual booking for pacing demand. As a result, for example, even if the base network improvement measure for a transportation request indicates a high value, but if the level of desire to convert the transportation request to actual booking is low (e.g., due to demand pacing, for example, a target number of transportation bookings for the pick-up location at the particular time period has already been achieved) , the booking conversion control parameter may be set to modify the contribution of the base network improvement measure in determining the surge pricing adjustment to result in a higher surge pricing adjustment than otherwise would be obtained without the booking conversion control parameter. Furthermore, the booking probability of the transportation request being converted into actual booking for using the candidate surge pricing is determined for evaluating or considering the impact or benefit of using the candidate surge pricing to transportation service network. In particular, the network improvement measure for the transportation request for using the candidate surge pricing is determined taking into account the booking probability for using the candidate surge pricing, which may thus be referred to as a booking probability adjusted network improvement measure. Therefore, the method 100 further takes into account the booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing in determining whether to implement the candidate surge pricing determined. Accordingly, in determining the surge pricing for the transportation request, the method 100 does not merely determine the candidate surge pricing according to the base network improvement measure but further determines the candidate surge pricing based on the booking conversion control parameter (e.g., for demand pacing) and then further evaluates the impact or benefit of the transportation request using the candidate surge pricing to transportation service network based on the booking probability for the transportation request for using the candidate surge pricing, for improving the effectiveness in controlling the surge pricing, resulting in improved efficiency in the transportation service network, such as with respect to driver utilization and revenue optimization. These advantages or technical effects, and / or other advantages or technical effects, will become more apparent to a person skilled in the art as the method 100 of controlling a surge pricing for a transportation request, as well as the corresponding system for controlling a surge pricing for a transportation request, is described in more detail according to various embodiments and example embodiments of the present invention.
[0046] In various embodiments, the method 100 further comprises obtaining a second network contribution measure for the transportation request with respect to the pick-up location. The second network contribution measure indicates the contribution measure of the transportation request to the transportation service network (the same contribution measure as described above with respect to the first network contribution measure) with respect to the pick-up location. In various embodiments, the second network contribution measure for the transportation request with respect to the pick-up location may be obtained based on the network contribution measure (e.g., network value) assigned to the pick-up location at the pick-up time. In this regard, the above-mentioned base network improvement measure for the transportation request for using the base surge pricing is determined further based on the second network contribution measure for the transportation request. For example, the base network improvement measure for the transportation request for using the base surge pricing may be determined further based on a difference between the first and second network contribution measures (e.g., a difference between the network value obtained with respect to the drop-off location and the network value obtained with respect to the pick-up location) for the transportation request.
[0047] In various embodiments, as explained hereinbefore, the contribution measure of the transportation request to the transportation service network (i.e., indicated by the first and second network contribution measures) is with respect to an efficiency of the transportation service network, such as with respect to driver utilization (e.g., percentage of time drivers are in-transit) in the transportation service network over a certain time period (e.g., over the next two to six hours or one day) .
[0048] In various embodiments, the value of the booking conversion control parameter for the transportation request is determined based on a target number of transportation bookings for the pick-up location in a predefined time period (e.g., a 1-minute time period) including the pick-up time.
[0049] In various embodiments, the value of the booking conversion control parameter for the transportation request is determined further based on a number of transportation bookings that has been made for the pick-up location in the predefined time period and based on an amount of time lapsed in the predefined time period.
[0050] In various embodiments, the value of the booking conversion control parameter for the transportation request is determined based on a difference between a ratio between the number of transportation bookings that has been made and the target number of transportation bookings for the pick-up location in the predefined time period and a ratio between the amount of time lapsed in the predefined time period and the predefined time period.
[0051] In various embodiments, the surge pricing adjustment for the transportation request is determined based on a difference between the base network improvement measure and the value of the booking conversion control parameter for the transportation request.
[0052] In various embodiments, the first booking probability for the transportation request for using the candidate surge pricing is determined based on the pick-up location, the drop-off location and the pick-up time for the transportation and the candidate surge pricing for the transportation request using on a trained booking probability machine learning model.
[0053] In various embodiments, the method 100 further comprises determining a second booking probability for the transportation request for using the base surge pricing based on the pick-up location, the drop-off location and the pick-up time for the transportation and the base surge pricing for the transportation request. For example, the second booking probability for the transportation request may also be determined using the trained booking probability machine learning model.
[0054] In various embodiments, the method 100 further comprises: determining a second booking probability adjusted network improvement measure for the transportation request for using the base surge pricing based on the second booking probability, the first network contribution measure and the base surge pricing for the transportation request; and determining a net booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing based on the first and second booking probability adjusted network improvement measures for the transportation request, for example, based on a difference between the first booking probability and the second booking probability adjusted network improvement measures for the transportation request. In this regard, the above-mentioned surge pricing for the transportation request is determined based on the value of the booking conversion control parameter and the net booking probability adjusted network improvement measure for the transportation request.
[0055] In various embodiments, the above-mentioned determining (at 120) the surge pricing for the transportation request comprises determining whether to implement the candidate surge pricing for the transportation request based on a first surge adjustment condition configured based on the base network improvement measure and the booking conversion control parameter for the transportation request and a second surge adjustment condition configured based on the net booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing.
[0056] In various embodiments, the above-mentioned determining whether to implement the candidate surge pricing for the transportation request comprises determining to implement the candidate surge pricing for the transportation request based on determining that the first and second surge adjustment conditions are satisfied.
[0057] In various embodiments, the first surge adjustment condition is satisfied if a magnitude of a difference between the base network improvement measure and the value of the booking conversion control parameter for the transportation request satisfies a first predetermined threshold; and the second surge adjustment condition is satisfied if the net booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing satisfies a second predetermined threshold. It will be appreciated by a person skilled in the art that the first and second predetermined thresholds may be configured or set as desired or as appropriate, and thus, the present invention is not limited to any particular or specific first and second predetermined thresholds.
[0058] In various embodiments, the transportation request to the transportation service network is for a ride-hailing service or a delivery service (e.g., a food delivery service) .
[0059] FIG. 2 depicts a schematic block diagram of a system 200 for controlling a surge pricing for a transportation request (e.g., each transportation request) to a transportation service network according to various embodiments of the present invention, corresponding to the above-mentioned method 100 of controlling a surge pricing for a transportation request as described hereinbefore with reference to FIG. 1 according to various embodiments of the present invention. The system 200 comprises: at least one memory 202; and at least one processor 204 communicatively coupled to the at least one memory 202 and configured to perform the method 100 of controlling a surge pricing for a transportation request as described hereinbefore according to various embodiments of the present invention. Accordingly, the at least one processor 204 is configured to: receive a transportation request having associated therewith a pick-up location, a drop-off location and a pick-up time for a transportation; obtain a first network contribution measure for the transportation request with respect to the drop-off location, the first network contribution measure indicating a contribution measure of the transportation request to the transportation service network with respect to the drop-off location; determine a base network improvement measure for the transportation request for using a base surge pricing based on the base surge pricing and the first network contribution measure for the transportation request; set a value of a booking conversion control parameter for the transportation request, the booking conversion control parameter being configured to control a threshold for implementing a surge pricing adjustment to the base surge pricing for the transportation request; determine a candidate surge pricing for the transportation request based on the base surge pricing and the surge pricing adjustment for the transportation request, the surge pricing adjustment being determined based on the base network improvement measure and the value of the booking conversion control parameter for the transportation request; determine a first booking probability for the transportation request for using the candidate surge pricing based on the pick-up location, the drop-off location and the pick-up time for the transportation and the candidate surge pricing for the transportation request; determine a first booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing based on the first booking probability, the first network contribution measure and the candidate surge pricing for the transportation request; and determine the surge pricing for the transportation request based on the value of the booking conversion control parameter and the first booking probability adjusted network improvement measure for the transportation request.
[0060] It will be appreciated by a person skilled in the art that the at least one processor 204 may be configured to perform various functions or operations through set (s) of instructions (e.g., software modules) executable by the at least one processor 204 to perform various functions or operations. Accordingly, as shown in FIG. 2, the system 200 may comprise: a transportation request module (or a transportation request circuit) 206 configured to receive a transportation request having associated therewith a pick-up location, a drop-off location and a pick-up time for a transportation; a network contribution measure obtaining module (or a network contribution measure obtaining circuit) 208 configured to determine a first network contribution measure for the transportation request with respect to the drop-off location, the first network contribution measure indicating a contribution measure of the transportation request to the transportation service network with respect to the drop-off location; a network improvement measure determining module (or a network improvement measure determining circuit) 210 configured to determine a base network improvement measure for the transportation request for using a base surge pricing based on the base surge pricing and the first network contribution measure for the transportation request; a booking conversion control parameter setting module (or a booking conversion control parameter setting circuit) 212 configured to set a value of a booking conversion control parameter for the transportation request, the booking conversion control parameter being configured to control a threshold for implementing a surge pricing adjustment to the base surge pricing for the transportation request; a candidate surge pricing determining module (or a candidate surge pricing determining circuit) 214 configured to determine a candidate surge pricing for the transportation request based on the base surge pricing and the surge pricing adjustment for the transportation request, the surge pricing adjustment being determined based on the base network improvement measure and the value of the booking conversion control parameter for the transportation request; a booking probability determining module (or a booking probability determining circuit) 216 configured to determine a first booking probability for the transportation request for using the candidate surge pricing based on the pick-up location, the drop-off location and the pick-up time for the transportation and the candidate surge pricing for the transportation request; a booking probability adjusted network improvement measure determining module (or a booking probability adjusted network improvement measure determining circuit) 218 configured to determine a first booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing based on the first booking probability, the first network contribution measure and the candidate surge pricing for the transportation request; and a surge pricing determining module (or a surge pricing determining circuit) 220 configured to determine the surge pricing for the transportation request based on the value of the booking conversion control parameter and the first booking probability adjusted network improvement measure for the transportation request.
[0061] It will be appreciated by a person skilled in the art that the above-mentioned modules are not necessarily separate modules, and two or more modules may be realized by or implemented as one functional module (e.g., a circuit or a software program) as desired or as appropriate without deviating from the scope of the present invention. For example, two or more of the transportation request module 206, the network contribution measure obtaining module 208, the network improvement measure determining module 210, the booking conversion control parameter setting module 212, the candidate surge pricing determining module 214, the booking probability determining module 216, the booking probability adjusted network improvement measure determining module 218 and the surge pricing determining module 220 may be realized (e.g., compiled together) as one executable software program (e.g., software application or simply referred to as an “app” ) , which for example may be stored in the at least one memory 202 and executable by the at least one processor 204 to perform the corresponding functions or operations as described herein according to various embodiments.
[0062] In various embodiments, the system 200 for controlling a surge pricing for a transportation request corresponds to the method 100 of controlling a surge pricing for a transportation request as described hereinbefore with reference to FIG. 1, therefore, various operations, functions or steps configured to be performed by the least one processor 204 may correspond to various operations, functions or steps of the method 100 described hereinbefore according to various embodiments, and thus need not be repeated with respect to the system 200 for clarity and conciseness. In other words, various embodiments described herein in context of methods (e.g., the method 100 of controlling a surge pricing for a transportation request) are analogously valid for the corresponding systems or devices (e.g., the system 200 for controlling a surge pricing for a transportation request) , and vice versa. For example, in various embodiments, the at least one memory 202 may have stored therein the transportation request module 206, the network contribution measure obtaining module 208, the network improvement measure determining module 210, the booking conversion control parameter setting module 212, the candidate surge pricing determining module 214, the booking probability determining module 216, the booking probability adjusted network improvement measure determining module 218 and / or the surge pricing determining module 220, which respectively correspond to various operations, functions or steps of the method 200 of controlling a surge pricing for a transportation request as described hereinbefore according to various embodiments, which are executable by the at least one processor 204 to perform the corresponding operations, functions or steps as described herein.
[0063] A computing system, a controller, a microcontroller or any other system providing a processing capability may be provided according to various embodiments in the present invention. Such a system may be taken to include one or more processors and one or more computer-readable storage mediums. For example, the system 200 described hereinbefore may include at least one processor (or controller) 204 and at least one computer-readable storage medium (or memory) 202 which are for example used in various processing carried out therein as described herein. A memory or computer-readable storage medium used in various embodiments may be a volatile memory, for example a DRAM (Dynamic Random Access Memory) or a non-volatile memory, for example a PROM (Programmable Read Only Memory) , an EPROM (Erasable PROM) , EEPROM (Electrically Erasable PROM) , or a flash memory, e.g., a floating gate memory, a charge trapping memory, an MRAM (Magnetoresistive Random Access Memory) or a PCRAM (Phase Change Random Access Memory) .
[0064] In various embodiments, a “circuit” may be understood as any kind of a logic implementing entity, which may be special purpose circuitry or a processor executing software stored in a memory, firmware, or any combination thereof. Thus, in an embodiment, a “circuit” may be a hard-wired logic circuit or a programmable logic circuit such as a programmable processor, e.g., a microprocessor (e.g., a Complex Instruction Set Computer (CISC) processor or a Reduced Instruction Set Computer (RISC) processor) . A “circuit” may also be a processor executing software, e.g., any kind of computer program, e.g., a computer program using a virtual machine code, e.g., Java. Any other kind of implementation of various functions or operations may also be understood as a “circuit” in accordance with various other embodiments. Similarly, a “module” may be a portion of a system according to various embodiments in the present invention and may encompass a “circuit” as above, or may be understood to be any kind of a logic-implementing entity therefrom.
[0065] Some portions of the present disclosure are explicitly or implicitly presented in terms of algorithms and functional or symbolic representations of operations on data within a computer memory. These algorithmic descriptions and functional or symbolic representations are the means used by those skilled in the data processing arts to convey most effectively the substance of their work to others skilled in the art. An algorithm is here, and generally, conceived to be a self-consistent sequence of steps leading to a desired result. The steps are those requiring physical manipulations of physical quantities, such as electrical, magnetic or optical signals capable of being stored, transferred, combined, compared, and otherwise manipulated.
[0066] The present specification also discloses a system (e.g., which may also be embodied as one or more devices or apparatuses) , such as the system 200, for performing various operations, functions or steps of various methods described herein. Such a system may be specially constructed for the required purposes or may comprise a general purpose computer system selectively activated or reconfigured by a computer program stored in the computer system. In general, various algorithms that may be presented herein are not limited to being implemented or executed by any particular computer system. Alternatively, the construction of more specialized computer system to perform various operations, functions or steps of various methods described herein may be provided as desired or as appropriate without going beyond the scope of the present invention.
[0067] In addition, the present specification also at least implicitly discloses computer program (s) or software / functional module (s) , in that it would be apparent to a person skilled in the art that various operations, functions or steps of various methods described herein may be put into effect by computer code. The computer program (s) is not intended to be limited to any particular programming language and implementation thereof, and it will be appreciated by a person skilled in the art that a variety of programming languages and coding thereof may be used to implement the computer program (s) . Moreover, the computer program (s) is not intended to be limited to any particular control flow as there are a variety of programming languages which can use different control flows. It will be appreciated by a person skilled in the art that a computer program may be stored on any computer-readable storage medium (non-transitory computer-readable storage medium) , such as but not limited to, a magnetic disk, an optical disk or a memory chip. For example, a computer program stored on a computer-readable storage medium may be loaded and executed on a computer system to implement various operations, functions or steps of various methods described herein according to various embodiments of the present invention.
[0068] Accordingly, in various embodiments, there is provided a computer program product, embodied in one or more computer-readable storage mediums (non-transitory computer-readable storage medium) , comprising instructions (e.g., the transportation request module 206, the network contribution measure obtaining module 208, the network improvement measure determining module 210, the booking conversion control parameter setting module 212, the candidate surge pricing determining module 214, the booking probability determining module 216, the booking probability adjusted network improvement measure determining module 218 and / or the surge pricing determining module 220) executable by one or more computer processors to perform a method 100 of controlling a surge pricing for a transportation request as described hereinbefore with reference to FIG. 1 according to various embodiments of the present invention. Accordingly, various computer programs or software modules described herein may be stored in a computer program product receivable by a system therein, such as the system 200 as shown in FIG. 2, for execution by at least one processor 204 of the system 200 to perform various operations, functions or steps of various methods described herein according to various embodiments of the present invention.
[0069] It will be appreciated by a person skilled in the art that various modules described herein (e.g., the transportation request module 206, the network contribution measure obtaining module 208, the network improvement measure determining module 210, the booking conversion control parameter setting module 212, the candidate surge pricing determining module 214, the booking probability determining module 216, the booking probability adjusted network improvement measure determining module 218 and / or the surge pricing determining module 220) may be software module (s) realized by computer program (s) or set (s) of instructions executable by a computer processor to perform various functions or operations. Various modules described herein (e.g., the transportation request module 206, the network contribution measure obtaining module 208, the network improvement measure determining module 210, the booking conversion control parameter setting module 212, the candidate surge pricing determining module 214, the booking probability determining module 216, the booking probability adjusted network improvement measure determining module 218 and / or the surge pricing determining module 220) may also be implemented as hardware module (s) being functional hardware unit (s) designed to perform various functions or operations. More particularly, in the hardware sense, a module is a functional hardware unit designed for use with other components or modules. For example, a module may be implemented using discrete electronic components, or it can form a portion of an entire electronic circuit such as an Application Specific Integrated Circuit (ASIC) . Numerous other possibilities exist. It will also be appreciated by a person skilled in the art that a combination of hardware and software modules may be implemented. Furthermore, various operations, functions or steps of various methods described herein may be performed in parallel rather than sequentially as desired or as appropriate (e.g., as long as it does not render the method (s) inoperable or unsatisfactory for its intended purpose) .
[0070] In various embodiments, the system 200 for controlling a surge pricing for a transportation request may be realized by any computer system (e.g., desktop or portable computer system) including at least one processor and at least one memory, such as an example computer system 300 as schematically shown in FIG. 3 as an example only and without limitation. Various methods / steps or functional modules may be implemented as software, such as a computer program being executed within the computer system 300, and instructing the computer system 300 (in particular, one or more processors therein) to conduct various functions or operations as described herein according to various embodiments. The computer system 300 may comprise a system unit 302, one or more input devices 304 such as a keyboard, a touchscreen and / or a mouse, and a plurality of output devices such as a display 308. The system unit 32 may be connected to a computer network 312 via a suitable transceiver device 314, to enable access to e.g., the Internet or other network systems such as Local Area Network (LAN) or Wide Area Network (WAN) . The system unit 302 may include a processor 318 for executing various instructions, a Random Access Memory (RAM) 320 and a Read Only Memory (ROM) 322. The system unit 302 may further include a number of Input / Output (I / O) interfaces, for example I / O interface 324 to the display device 308 and I / O interface 326 to the one or more input devices 304. The components of the system unit 302 typically communicate via an interconnected bus 328 and in a manner known to a person skilled in the art.
[0071] For example, the system 200 may receive a transportation request over a wireless communication network (e.g., cellular network (e.g., 4G, 5G or a future generation cellular network) ) from a computing device or system (e.g., a portable computing device, such as a mobile communication device) of a user requesting a transportation, such as for a ride-hailing service. The system 200 may then operate to control or determine the surge pricing for the transportation request according to the method 100 as described herein according to various embodiments of the present invention, and a price for the transportation request (e.g., based on the base price and the surge pricing determined) may then be transmitted from the system 200 over the wireless communication network to the device of the user for displayed thereat for consideration or confirmation by the user on whether to make the booking.
[0072] It will be appreciated by a person skilled in the art that the terminology used herein is for the purpose of describing various embodiments only and is not intended to be limiting of the present invention. As used herein, the singular forms “a” , “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises” and / or “comprising, ” when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.
[0073] Any reference to an element or a feature herein using a designation such as “first” , “second” and so forth does not limit the quantity or order of such elements or features, unless stated or the context requires otherwise. For example, such designations may be used herein as a convenient way of distinguishing between two or more elements or instances of an element. Thus, a reference to first and second elements does not necessarily mean that only two elements can be employed, or that the first element must precede the second element, unless stated or the context requires otherwise. In addition, a phrase referring to “at least one of” a list of items refers to any single item therein or any combination of two or more items therein.
[0074] In order that the present invention may be readily understood and put into practical effect, various example embodiments of the present invention will be described hereinafter by way of examples only and not limitations. It will be appreciated by a person skilled in the art that the present invention may, however, be embodied in various different forms or configurations and should not be construed as limited to the example embodiments set forth hereinafter. Rather, these example embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the present invention to those skilled in the art.
[0075] In particular, for better understanding of the present invention and without limitation or loss of generality, and unless stated or the context requires otherwise, various example embodiments of the present invention will now be described with respect to an example practical application of the transportation request being for a ride-hailing service and the transportation service network being a ride-hailing service network for illustration purposes only. It will be understood by a person skilled in the art that the present invention is not limited to such an example practical application and may be employed in a variety of other practical applications as desired or as appropriate without going beyond the scope of the present invention, as long as a surge pricing for a transportation request to a transportation service network is desired to be controlled (or dynamically controlled) , such as but not limited to, a delivery service (e.g., a food delivery service, where a surge pricing for a food delivery request to a food delivery service network may be controlled) .
[0076] As explained in the background, conventional price surging techniques suffer from a number of drawbacks / deficiencies, resulting in inefficiencies in the transportation service network, such as but not limited to, poor or inefficient driver utilization (e.g., poor driver in-transit time or high driver idle / waiting time) and poor or ineffective revenue optimization (e.g., in relation to gross merchandise value (GMV) ) in the transportation service network. For example, conventionally, a surge pricing for a transportation request (which includes a price check, such as a fare check (which may be referred to herein as FC) for a ride-hailing service) may be computed based on only a pick-up location (e.g., a geographic location or region, such as represented by a geohash (e.g., geohash 6) ) of the transportation and a time period (e.g., over a 1-minute time period) . In this regard, a surge pricing may thus be computed per geohash for each time period (e.g., for each 1-minute time period) . Therefore, the same surge pricing may be computed for all transportation requests with the same pick-up location (e.g., same geohash) within the same time period. However, various example embodiments find that transportation requests should not have the same surge pricing simply because they have the same pick-up location within the same time period. In this regard, various example embodiments find that, for transportation requests having the same pick-up location within the same time period, transportation requests having certain characteristics or parameters may be preferred over transportation requests having certain other characteristics or parameters. For example, a transportation request to go to a central business district (CBD) area just before peak hour thereat may be preferred. On the other hand, for example, a transportation request to go to the outskirts just before peak hours at the CBD area may be less desirable. Accordingly, the above-mentioned conventional price surging technique for determining the surge pricing for a transportation request to a transportation service network may be inefficient (or may not be optimal) for the transportation network. In contrast, various example embodiments seek to control (e.g., adjust or fine-tune) the surge pricing for a transportation request (e.g., each transportation request) based on additional characteristics or parameters of the transportation requests, such as to reflect one or more preferences for improving efficiency in the transportation service network. In this regard, various example embodiments seek to quantify such preference (s) and control the surge pricing for each incoming transportation request (e.g., including a fare check for a transportation) in real-time for improving efficiency in the transportation service network, such as with respect to driver utilization.
[0077] Accordingly, various example embodiments provide a method of controlling a surge pricing for a transportation request to a transportation service network, as well as a system thereof, that seeks to overcome, or at least ameliorate, one or more deficiencies in conventional methods of controlling a surge pricing, and more particularly, with improved effectiveness in controlling the surge pricing, resulting in improved efficiency (e.g., with respect to driver utilization and revenue optimization) in the transportation service network. Various example embodiments seek to control the surge pricing for each transportation request based on a network contribution measure obtained for the transportation request with respect to the drop-off location. In this regard, the network contribution measure indicates a contribution measure (e.g., network value) of the transportation request to the transportation service network with respect to the drop-off location. In this regard, the contribution measure is with respect to an efficiency of the transportation service network, such as driver utilization (e.g., maximize driver in-transit time or minimize drive idle / waiting time) in the transportation service network. In various example embodiments, the method of controlling a surge pricing for a transportation request may be implemented to dynamically fine-tune the surge pricing for the transportation request in real-time to shape or influence passenger / consumer behavior to advantageously maximize / optimize network efficiency, such as maximizing driver utilization, under the constraint of a target number of transportation bookings (which may herein be referred to a target bookings to convert (BTC) ) for the pick-up location over a predefined time period (e.g., a 1-minute time period) including the pick-up time (e.g., achieving an actual number of transportation bookings as close to the target number as possible for the pick-up location over the predefined time period) .
[0078] According to various example embodiments, the technical problem of how to control the surge pricing for a transportation request to maximize / optimize network efficiency may be decomposed into three sub-problems, as illustrated in FIG. 4. In particular, the technical problem may be decomposed into a spatial-temporal surge for demand pacing or dampening (e.g., determines the target number of transportation bookings (or the target BTC) for a pick-up location over a predefined time period) , a fine-tuned individual surge for transportation request selection (e.g., determines which transportation requests (e.g., each including a fare check) are more preferable to convert into actual bookings) and an elasticity-based surge for revenue optimization / maximization (e.g., determines the final surge pricing for the transportation request taking into account the impact of the candidate surge pricing to the potential revenue generated by the transportation request) .
[0079] In various example embodiments, based on the target number of transportation bookings (e.g., the target BTC) for a pick-up location over a predefined time period associated with an incoming transportation request, which may be output by an upstream spatial-temporal surge module / component (e.g., target of 8 out of 20 (or 40%) of transportation requests to be converted into actual bookings) , the surge pricing for the incoming transportation request may be individually controlled based on the characteristics or parameters associated with the transportation request. In this regard, various example embodiments may seek to improve (e.g., softly nudge up) the booked-through-rate (BTR) of high value transportation requests and decrease the BTR of low value transportation requests. In this regard, for each transportation request, various example embodiments seek to determine a network contribution measure for a transportation request based the characteristics or parameters associated with the transportation request, such as with respect to the drop-off location. In this regard, the network contribution measure may be a measure of network value (e.g., benefit) to the transportation service network.
[0080] As explained hereinbefore, it will be appreciated by a person skilled in the art that the present invention is not limited to any particular or specific type of contribution measure (or network value) to the transportation service network, which may be defined or configured for measuring or representing a type of contribution to the transportation service network as desired or as appropriate, such as an efficiency of the transportation service network that may be desired to be improved or optimized to benefit the transportation service network. Therefore, for example, depending on the particular or specific type of efficiency of the transportation service network desired to be improved or optimized by controlling the surge pricing, the contribute measure may be defined or configured as appropriate for measuring or representing such a type of efficiency of the transportation service network. Therefore, depending on the particular or specific type of efficiency of the transportation service network desired to be improved or optimized, the manner of defining the contribution measure (and thus, the network contribution measure) to represent such a type of efficiency may also be different. As an illustrative example, in the case of the transportation service network being a ride-hailing service network, a type of efficiency of the transportation service network that may be improved or optimized by controlling the surge pricing may be with respect to driver utilization in the ride-hailing service network (e.g., maximizing driver in-transit time or minimizing driver idle / waiting time) over a time period (e.g., over the next two to six hours or one day) . Therefore, it will be appreciated by a person skilled in the art that various contribution measures (or network values) may be provided or defined as desired or as appropriate for measuring or representing various types of efficiencies of the transportation service network. In various example embodiments, a contribution measure (or network value) may also be provided or defined for measuring or representing a combination of two or more types of efficiencies of the transportation service network desired to be improved or optimized by controlling the surge pricing.
[0081] In various example embodiments, the contribution measure for measuring or representing an efficiency of the transportation service network may measure or represent an expected / estimated / predicted long-term benefit with respect to the efficiency at a given location (geohash or area) at a given time window and with certain given contexts (e.g., real-time and historical) . As described above, the network contribution measure may be a measure of network value (e.g., benefit) to the transportation service network with respect to an efficiency of the transportation service network. In this regard, for each time period or window (e.g., each 1-minute time period) , each geographical region (e.g., each geohash) may have a respective network value assigned thereto relating to an efficiency of the transportation service network. In various example embodiments, the network value may be defined to measure an efficiency of the transportation service network with respective to driver utilization. In this regard, for each time period or window, each geographical region may have a respective network value assigned thereto indicating a degree or level of benefit of a driver being available to pick up passenger (s) at the geographical region at the time period (e.g., sending passenger (s) to the geographical region as the drop-off location and thus being available to pick up passenger (s) at the geographical region after drop-off) with respect to driver utilization of the transportation service network. For example, a high network value at a geographical region at a time period may indicate a high degree or level of benefit of a driver being available to pick up passenger (s) at the geographical region at the time period with respect to driver utilization of the transportation service network.
[0082] In various example embodiments, a first network contribution measure for the transportation request with respect to the drop-off location may be obtained based on the network contribution measure (e.g., network value) assigned to the drop-off location at the drop-off time. In this regard, the drop-off time may be determined based on the pick-up time and the transit time (e.g., estimated) from the pick-up location to the drop-off location. In various example embodiments, a second network contribution measure for the transportation request with respect to the pick-up location may be obtained based on the network contribution measure (e.g., network value) assigned to the pick-up location at the pick-up time. Accordingly, various example embodiments may control the surge pricing for a transportation request based on the first and second network contribution measures obtained for the transportation request with respect to the pick-up and drop-off locations relating to an efficiency of the transportation service network (network efficiency) with respective to driver utilization, to optimize or maximize driver utilization in the transportation service network over a time period (e.g., over the next two to six hours or one day) .
[0083] As described in the background, there has been disclosed a price surging technique which additionally takes into account the drop-off location of the transportation request, which may be referred to as the dynamic drop-off surcharge (DDoS) technique. This DDoS technique adjusts the surge pricing based on the drop-off location so as to promote or encourage the conversion of transportation requests to actual bookings for transportation requests with a drop-off location having associated therewith a shorter driver idle / waiting time based on historical data. Therefore, the DDoS technique reduces the surge pricing when the transportation request results in a shorter driver idle / waiting time than expected, and vice versa, based on historical data. However, various example embodiment note that the DDoS technique at least lack the demand pacing module / function as described above. Therefore, in various example embodiments, the above-mentioned technical problem may be formulated as how to control (e.g., adjust or fine-tune) the surge pricing for a transportation request for improving an efficiency of the transportation service network while pacing the demand optimally.
[0084] A method of controlling a surge pricing for a transportation request (e.g., each transportation request) to a transportation network will now be described according to various example embodiments. FIG. 5 depicts a tree diagram showing an overview of the method of controlling a surge pricing (e.g., the resultant surge pricing may be referred to as the network value-based individual surge (NVIS) pricing) for a transportation request. As shown, the surge pricing may be controlled or determined based on network efficiency optimization (or maximization) (e.g., based on the network contribution measure (s) obtained for the transportation request as described above according to various example embodiments) , network revenue optimization (or maximization) and demand pacing, e.g., corresponding to the three sub-problems described hereinbefore with reference to FIG. 4, namely, the fine-tuned individual surge module (or component) , the elasticity-based surge module (or component) and the spatial-temporal surge module (or component) , respectively. The demand pacing may be based on optimally encouraging or discouraging demand (conversion of transportation requests into actual bookings) for a pick-up location over a given time period.
[0085] In various example embodiments, to achieve / implement network efficiency optimization (or maximization) , network revenue optimization (or maximization) and demand pacing, the method of controlling a surge pricing for a transportation request may be implemented by a dynamic surge (DS) model comprising a network contribution measure obtaining module (or component) 608 for obtaining network contribute measures for each transportation request based on network values assigned to pick-up and drop-off locations, respectively, relating to an efficiency of the transportation service network (network efficiency) for optimizing (or maximizing) network efficiency, a network revenue determining module (or component) 610 for determining the potential revenue (e.g., from the base fare and the base surge pricing) generated by each transportation request for optimizing (or maximizing) a network revenue of the transportation service network and a demand pacing module (or component) 612 for facilitating demand pacing for a pick-up location over a given time period.
[0086] FIG. 6 depicts a schematic flow diagram of the method 600 of controlling a surge pricing for a transportation request (e.g., each transportation request) according to various example embodiments of the present invention (e.g., corresponding to the method 100 of controlling a surge pricing as described hereinbefore with reference to FIG. 1 according to various embodiments) , based on the network contribution measure obtaining module 608, the network revenue determining module 610 and the demand pacing module 612.
[0087] The method 600 comprises receiving a transportation request (which includes a price check, such as a fare check (FC) for a ride-hailing service) having associated therewith a pick-up location, a drop-off location and a pick-up time for a transportation (i.e., comprising pick-up location information, drop-off location information and pick-up time information for the transportation) . It will be appreciated by a person skilled in the art that the present invention is not limited to these characteristics or parameters and that additional one or more characteristics or parameters may be provided as desired or as appropriate, such as the transit time (from the pick-up location to the drop-off location) , the number of passengers, the selected type of ride-hailing service (e.g., standard, premium, large size vehicle, and so on) .
[0088] The method 600 further comprises obtaining a first network contribution measure (by the network contribution measure obtaining module 608, e.g., corresponding to the network contribution measure obtaining module 208 as described hereinbefore according to various embodiments) for the transportation request for a ride-hailing service with respect to the drop-off location, the first network contribution measure indicating a contribution measure of the transportation request to the ride-hailing service network with respect to the drop-off location.
[0089] The method 600 further comprises determining a base network improvement measure for the transportation request (e.g., by a transportation request network improvement measure determining module 614, e.g., corresponding to the network improvement measure determining module 210 as described hereinbefore according to various embodiments) for using a base surge pricing based on a base fare, a base surge pricing and the first network contribution measure for the transportation request.
[0090] The method 600 further comprises setting a value of a booking conversion control parameter (e.g., a target booking-to-convert (BTC) parameter) for the transportation request (e.g., by the demand pacing module 612, e.g., corresponding to the booking conversion control parameter setting module 212 as described hereinbefore according to various embodiments) , the booking conversion control parameter being configured to control a threshold for implementing a surge pricing adjustment to the base surge pricing for the transportation request.
[0091] The method 600 further comprises determining a candidate surge pricing for the transportation request based on the base surge pricing and the surge pricing adjustment for the transportation request, the surge pricing adjustment being determined based on the base network improvement measure and the value of the booking conversion control parameter for the transportation request.
[0092] The method 600 further comprises determining a first booking probability for the transportation request (e.g., by a booking probability determining module 616 (which may be referred to as a BTR module) , e.g., corresponding to the booking probability determining module 216 as described hereinbefore according to various embodiments) using the candidate surge pricing based on the pick-up location, the drop-off location and the pick-up time for the transportation and the candidate surge pricing for the transportation request.
[0093] The method 600 further comprises determining a first booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing based on the first booking probability, the first network contribution measure and the candidate surge pricing for the transportation request.
[0094] The method 600 further comprises determining the surge pricing for the transportation request based on the value of the booking conversion control parameter and the first booking probability adjusted network improvement measure for the transportation request.
[0095] In various example embodiments, the first network contribution measure for the transportation request with respect to the drop-off location is obtained based on the network contribution measure (e.g., network value) assigned to the drop-off location at the drop-off time.
[0096] In various example embodiments, the method 600 further comprises obtaining a second network contribution measure for the transportation request with respect to the pick-up location, the second network contribution measure indicating a contribution measure of the transportation request to the transportation service network with respect to the pick-up location. In this regard, the base network improvement measure for the transportation request for using the base surge pricing is determined further based on the second network contribution measure for the transportation request, such as based on a difference between the first and second network contribution measures for the transportation request.
[0097] As an illustrative example according to various example embodiments, at a given time upon receiving a transportation request (including a fare check request) , various example embodiments determine (or define) a base network improvement measure for the transportation request (e.g., determining a value of the fare check request (i.e., how much value it brings to the transportation service network) ) for using a base surge pricing. In this regard, various example embodiments seek to optimize (or maximize) not only the network revenue (e.g., the GMV) , but also optimize (or maximize) the overall network efficiency. In various example embodiments, the GMV may be the sum of all the bookings fares at a geographical region over a time period (e.g., 1 day) .
[0098] In various example embodiments, the network improvement measure determined for the transportation request indicates an improvement measure to the transportation service network which takes into account a potential revenue generated by the transportation request in addition to the network contribution measure for the transportation request. Therefore, the network improvement measure may be an overall improvement value determined for the transportation request to the transportation service network which takes into account the value from the network contribution measure and the value from the potential revenue generated by the transportation request. Therefore, the base network improvement measure for the transportation request for using the base surge pricing may be determined based on the base surge pricing (or more specifically, the potential revenue generated from the base fare and the base surge pricing, e.g., determined by the network revenue determining module 610) and the first and second network contribution measures for the transportation request obtained with respect to the drop-off and pick-up locations (e.g., by the network contribution measure obtaining module 608) . By way of an illustrative example and without limitation, given the additive nature of the network contribution measure, the base network improvement measure (Vi, where i denotes an ith transportation request) for a transportation request may be defined or determined by the following expression:
[0099] where si denotes the base surge pricing (which may be in the form of a surge parameter or a surge multiplier / factor) and Fi denotes the base fare (e.g., the non-surcharge non-surge (NSNS) fare) for the transportation request. The product siFi thus represents the fare with base surge (which corresponds to the potential revenue generated from the base fare and the base surge pricing) for the transportation request. In the illustrative example, the first network contribution measure for the transportation request with respect to the drop-off location may be obtained based on the pick-up location pui, the drop-off location doi, the pick-up time ( which may also be referred to as the start time) and the transit time from the pick-up location to the drop-off location which may be expressed as In various example embodiments, the second network contribution measure for the transportation request with respect to the drop-off location may be obtained based on the pick-up location pui, the pick-up time and the transit time from the pick-up location to the pick-up location tpu, pu (which is effectively zero) , which may be expressed as N (pui, pui, tpu, pu) .
[0100] In various example embodiments, the first network contribution measure for the transportation request with respect to the drop-off location doi may be obtained based on the network contribution measure (e.g., network value) assigned to the drop-off location doi (e.g., the geohash including the drop-off location doi) at the drop-off time (e.g., the predefined time period including the drop-off time) . In this regard, the drop-off time may be determined based on the pick-up time and the transit time (e.g., estimated) from the pick-up location to the drop-off location The second network contribution measure for the transportation request with respect to the pick-up location pui may be obtained based on the network contribution measure (e.g., network value) assigned to the pick-up location pui (e.g., the geohash including the pick-up location pui) at the pick-up time (e.g., the predefined time period including the pick-up time ) . For example, these network values assigned to the drop-off and pick-up locations may be obtained or retrieved from a lookup table storing, for each geographical region (e.g., each geohash) and each time period or window (e.g., each 1-minute time period) , the network value assigned to the geographical region for the time period. In various example embodiments, the respectively network value assigned to a geographical region for a time period may be a signal that indicates an estimated / predicted long-term benefit with respect to the efficiency of the transportation service network for a driver being available to pick up passenger (s) at the geographical region at the time period.
[0101] In various example embodiments, the first network contribution measure may be obtained (or determined) by applying a time discount factor γ to the network value obtained with respect to the drop-off location, where the time refers to the transit time taken to travel from the pickup location pui to the drop-off location doi. The time discount factor γ seeks to discount the network value obtained with respect to the drop-off location at the drop-off time with respect to the pick-up time (or with respect to the transit time from the pickup location pui to the drop-off location doi) . The time discount factor γ may be set as desired or as appropriate, such as but not limited to, a discount of 5%for every 5 minutes of transit time from the pickup location pui to the drop-off location doi. Accordingly, a net network contribution measure for the transportation request (e.g., indicating a net contribution (or incremental contribution) of the transportation request) with respect to an efficiency of the transportation service network may be determined based on a difference between the first and second network contribution measures obtained for the transportation request (i.e., In this regard, the net network contribution measure may be a positive or negative value. In Equation (1) , λ denotes a tunable parameter for balancing (or weighting) the contribution of the network revenue siFi component and the contribution of the net network contribution measure component to the value of the base network improvement measure (Vi) for the transportation request. It will be appreciated by a person skilled in the art that the tunable parameter λ may be tuned as appropriate such as through an offline simulation. As an illustrative example, a grid search may be performed over various values for the tunable parameter λ (e.g., 0.5, 1, 1.5) and simulate the number of rides and the GMV, and select the value of the tunable parameter λ that results in the highest rides and / or GMV.
[0102] For example, various services from supply shaping, dispatch and pricing domains can consume a network value signal to support various decision processes. For example, the network value signal can support various verticals including transport, food and mart. In various example embodiments, the respective network value assigned to each geographical region and each time period or window may be determined (or estimated or predicted) by using a machine learning model (which may be referred to as a network value machine learning model) trained based on various historical data such as historical driver trajectories, including in-transit time.
[0103] In various example embodiments, the value of the above-mentioned booking conversion control parameter (e.g., denoted by αi, where i denotes the ith transportation request) for the transportation request is determined (e.g., by the demand pacing module 612) based on a target number of transportation bookings (e.g., a target booking-to-convert (BTC) parameter or signal) for the pick-up location pui in a predefined time period (e.g., over a 1-min time period) including the pick-up time In various example embodiments, the value of the booking conversion control parameter for the transportation request is determined further based on a number of transportation bookings that has been made (e.g., denoted by a booking counter signal (BCS) or parameter) for the pick-up location pui in the predefimed time period and based on an amount of time lapsed (e.g., denoted by τi) in the predefined time period.
[0104] In various example embodiments, the value of the booking conversion control parameter αi for the transportation request is determined based on a difference between a ratio between the number of transportation bookings that has been made (BCS) and the target number of transportation bookings (BTC) for the pick-up location in the predefined time period and a ratio between the amount of time lapsed τi in the predefined time period and the predefined time period. In various example embodiments, the surge pricing adjustment (e.g., denoted by δi) for the transportation request is determined based on a difference between the base network improvement measure Vi for the transportation request and the value of the booking conversion control parameter αi for the transportation request.
[0105] In various example embodiments, the demand pacing module 612 is configured to facilitate demand pacing for enabling the optimal conversion of demands (transportation requests) into actual bookings. In this regard, demand pacing may be achieved by setting the value of the booking conversion control parameter αi for controlling a threshold for implementing a surge pricing adjustment to the base surge pricing for the transportation request. In particular, the surge pricing adjustment (to the base surge pricing) may be determined based on the base network improvement measure Vi and the value of the booking conversion control parameter αi for the transportation request. Therefore, the value of the booking conversion control parameter αi for the transportation request can be dynamically set to modify the contribution of the base network improvement measure in determining the surge pricing adjustment. For example, the value thereof may be set (based on the number of transportation bookings that has been made (BCS) , the target number of transportation bookings (BTC) for the pick-up location in the predetermined time period and the amount of time lapsed τi) to indicate or reflect a degree or level of desire to convert the transportation request to actual booking for pacing demand. As an illustrative example, in various example embodiments, demand pacing is implemented based on a rule-based online thresholding performed by the rule-based online thresholding module 620. The rule-based online thresholding demand pacing is configured to use the booking conversion control parameter αi output by the demand pacing module 612 to threshold the base network improvement measure Vi determined for the transportation requests.
[0106] As an illustrative example, the rule-based online thresholding demand pacing may be implemented based on a PI (proportional-integral) controller or a PID (proportional-integral-derivative) controller, which is effective when the underlying stochastic process is unknown. For example, it will be understood by a person skilled in the art that the PI or PID controller is based on classic control design theory whereby a target may be set and the control algorithm would restore the threshold (which is a function of the budget ratio) towards the target regardless of the threshold being above or under the target.
[0107] To illustrate the rule-based online thresholding demand pacing, FIGs. 7A and 7B illustrate plots of the booking conversion control parameter αi and the budget ratio for the PI controller and the PID controller. In particular, FIGs. 7A and 7B show six time steps on the y axis, and αi being the threshold that the PI or PID controller controls. In this regard, the budget ratio may be defined as the ratio of the number of transportation bookings that has been made (BCS) to a predetermined target number of transportation bookings (BTC) , that is, BCS / BTC. For example, at time 2 in FIG. 7B, the PID controller determines that the budget ratio is not increasing fast enough according to a predetermined target rate. As a result, the PID controller may decrease the αi threshold. Lowering this threshold αi results in a higher number of simulated transportation booking counts, since more transportation requests would have a value (e.g., the base network improvement measure) which passes the lowered threshold αi. Accordingly, it can be seen that the PI or PID controller is able to control the pace demand without knowing the underlying stochastic process, although each controller exceeds the budget ratio slightly.
[0108] In various example embodiments, the above-mentioned first booking probability (e.g., denoted as b (ni, Ui) ) for the transportation request for using the candidate surge pricing is determined (e.g., by the booking probability determining module 616 (which may be referred to as the BTR module) ) based on the pick-up location pui, the drop-off location doi and the pick-up time for the transportation and the candidate surge pricing ni for the transportation request using on a trained booking probability machine learning model (which may be referred to as the BTR model) . In various example embodiments, the BTR module 616 (or the BTR model) is applied to auto-adapt the surge pricing. For example, in the DDoS technique discussed in the background, its model was fitted only one time. As a result, the model's perception of the market stay the same as when it was fitted. In contrast, the BTR model according to various example embodiments is configured to keep up to date with the market. For example, the BTR model may be fitted at regular intervals using Chimera ML pipelines to keep up to date with the market. In various example embodiments, historical data may be used to fit or train the BTR model to determine (or predict) the booking probability (BTR) for the transportation request.
[0109] As an illustrative example, the XGBoost model may be employed to train the BTR model. For example, the prediction function may be expressed as:
[0110] where yi denotes the predicted probability for an ith transportation request to proceed or convert into a transportation booking.
[0111] FIG. 8 depicts a schematic flow diagram illustrating an example training of the BTR model according to various example embodiments of the present invention. As shown in FIG. 8, the BTR model may be trained based on a number of characteristics or parameters of transportation requests and the corresponding outcomes (i.e., transportation booking decisions, i.e., whether transportation booking made or not) of such transportation requests. For example, the characteristics or parameters of a transportation request may include an overall fare / price (e.g., siFi or niFi) for the transportation request, a pick-up location, a drop-off location and a pick-up time. The BTR model may be trained based on various characteristics or parameters of the transportation request as desired or as appropriate and are not limited to the above-mentioned types of characteristics or parameters. For example, other characteristics or parameters may include the distance from the pick-up location to the drop-off location, the transit time from the pick-up location to the drop-off location, the day of the week, and so on. Furthermore, various historical features may be extracted from historical bookings, such as the average bookings'fare per unit distance (e.g., per km) and the average BTR from previous T-1 to T-30 days (where T denotes the current day) for each group of features or characteristics (e.g., pick-up geohash, drop-off geohash, day of week, hour of day) . In this regard, the BTR may be defined as a ratio of the number of transportation bookings to the number of transportation requests for a given segment, whereby each segment is defined by the above-mentioned group of characteristics or features. By using both transport request's characteristics and historical bookings'characteristics, the XGBoost model can obtain more information about for which transportation request the overall BTR will be higher. For example, the use of such historical bookings'characteristics as input features have been found to improve the BTR model.
[0112] In various example embodiments, to obtain historical characteristics or features (e.g., average fare per km, average BTR) but not import any future information, a historical features pipeline may first be created. For example, this pipeline may collect data from previous T-1 to T-30 days and calculate features (e.g., trend features such as average fare per km, average BTR) for each feature type of the group of feature types (e.g., the pick-up location (e.g., geohash) , the drop-off location (e.g., geohash) , the day of week, hour of day) for each day. Then for a transportation request, the historical features used are from, for example, the day before this transportation request's creation date so that the historical features will include the most recent and relevant historical features for the machine learning model. In this way, the offiine data set has a more similar distribution as the online data, and more accurate model performance metrics can be achieved. As for the model update technique, the model may be trained and uploaded to a model bank daily.
[0113] In various example embodiments, the method 600 further comprising determining a second booking probability (e.g., denoted as b (si, Ui) ) for the transportation request for using the base surge pricing si based on the pick-up location pui, the drop-off location doi and the pick-up time for the transportation and the base surge pricing si for the transportation request. In this regard, the second booking probability b (si, Ui) for the transportation request may also be determined (or estimated) using the trained BTR model. For example, the price of the transportation request input to the BTR model may be determined from the base surge pricing si and the base fare Fi for the transportation request.
[0114] In various example embodiments, the method 600 further comprises: determining a second booking probability adjusted network improvement measure for the transportation request for using the base surge pricing si based on the second booking probability b (si, Ui) , the first network contribution measure and the base surge pricing si for the transportation request; and determining a net booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing ni based on the first booking probability b (ni, Ui) and the second booking probability b (si, Ui) for the transportation request. In this regard, the surge pricing for the transportation request is determined based on the value of the booking conversion control parameter αi and the net booking probability adjusted network improvement measure (e.g., denoted as Mi) for the transportation request.
[0115] In various example embodiments, the above-mentioned determining the surge pricing for the transportation request comprises determining whether to implement the candidate surge pricing ni for the transportation request based on a first surge adjustment condition configured based on the base network improvement measure Vi and the booking conversion control parameter αi for the transportation request and a second surge adjustment condition configured based on the net booking probability adjusted network improvement measure Mi for the transportation request for using the candidate surge pricing ni. In various example embodiments, the above-mentioned determining whether to implement the candidate surge pricing ni for the transportation request comprises determining to implement the candidate surge pricing ni for the transportation request based on determining that the first and second surge adjustment conditions are satisfied. In various example embodiments, the first surge adjustment condition is satisfied if a magnitude of a difference between the base network improvement measure Vi and the value of the booking conversion control parameter αi for the transportation request satisfies a first predetermined threshold (e.g., is greater than the first predetermined threshold, which may be denoted as ψ) ; and the second surge adjustment condition is satisfied if the net booking probability adjusted network improvement measure Mi for the transportation request for using the candidate surge pricing ni satisfies a second predetermined threshold (e.g., is greater than 0) .
[0116] For illustration purpose, an example method for controlling a surge pricing for each transportation request will now be described below according to various example embodiments of the present invention. For each transportation request i, including pick-up location information pui, drop-off location information doi and pick-up time information received (which includes a price check, such as a fare check for a ride-hailing service) , the following example operations (e.g., steps) may be performed.
[0117] Step 1: obtain the bookings-to-convert (BTC) parameter (or signal, e.g., corresponding to the target number of transportation bookings described hereinbefore) for the transportation request i for the pick-up location pui and the predefined time period (e.g., 60 seconds) including the pick-up time from upstream.
[0118] Step 2: obtain first and second network contribution measures (e.g., first and second network values) and with respect to the pick-up location pui and the drop-off location doi, respectively. For example, the first network contribution measure may be determined as follows: where the first network contribution measure (e.g., first network value) may be retrieved from a lookup table (e.g., as described hereinbefore) for the geographic region (e.g., geohash) including the pick-up location pui and the predefined time period including the drop-off time (which can be determined or estimated by ), which is subjected to a time discount factor γ (e.g., where tpu, do / 5 denotes the transit time (or trip duration) from the pick-up location pui to the drop-off location doi, for example, in multiples of 5 minutes. For example, if the trip duration is 10 minutes, then tpu, do / 5 would be 2. The 5 minutes is an example predefined length of time period that is aligned with the time index of the network value machine learning model described hereinbefore where the network value may be determined every 5 minutes) .
[0119] Step 3: calculate the base network improvement measure Vi for the transportation request for using the base surge pricing si:
[0120] where λ denotes a tunable parameter for balancing (or weighting) the contribution of the network revenue siFi component and the contribution of the net network contribution measure component to the value of the base network improvement measure Vi) for the transportation request. It will be appreciated by a person skilled in the art that the tunable parameter λ may be tuned as appropriate and is not limited to any particular or specific value.
[0121] Step 4: update / set the value of the booking conversion control parameter αi (which may also be referred to as the alpha threshold) for the transportation request, for example, as follows:
[0122] where Im refers to the set of transportation requests in minute m. αm denotes the initial booking conversion control parameter determined for the predefined time period (e.g., for minute m) , and it may be determined based on the average value of all the base network improvement measures Vi over the past 30 minutes, or more specifically, the sum of all the base network improvement measures Vi over the past 30 minutes divided by the count of all the transportation requests over the past 30 min. Accordingly, the value of the booking conversion control parameter αi for the transportation request is determined based on a difference between a ratio between the number of transportation bookings that has been made and the target number of transportation bookings for the pick-up location (BCS / BTC) in the predefined time period and a ratio between the amount of time lapsed in the predefined time period and the predefined time period (e.g., τi / 60) .
[0123] Step 5: calculate a candidate surge pricing ni for the transportation request, assuming the transportation request is selected, as follows: ni = si + δi
[0124] Accordingly, the surge pricing adjustment δi for the transportation request is determined based on a difference between the base network improvement measure Vi and the value of the booking conversion control parameter αi for the transportation request. For example, the surge pricing adjustment δi is determined based on applying an exponential function to the difference between the base network improvement measure Vi and the booking conversion control parameter αi, multiplied by the original or base surge si, and bounded by the minimum and maximum surge deltas δmin, δmax. These minimum and maximum bounds are predefined and for example, may be -0.1 for the minimum delta δmin and 0.1 for the maximum delta δmax. Accordingly, if the base network improvement measure Vi is much bigger than the booking conversion control parameter αi, then δi would be negative so as to decrease the original or base surge si and encourage the booking of the corresponding transportation request, since the transportation request is determined to highly contribute to the efficiency of the transportation service network, and vice versa.
[0125] Step 6: determine first and second booking probabilities (b (ni, Ui) and b (si, Ui) , which may be referred to as BTR values) for the transportation request for using the candidate surge pricing ni and the original surge pricing si, respectively.
[0126] Step 7: determine the net booking probability adjusted network improvement measure Mi for the transportation request for using the candidate surge pricing ni, including determining the first booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing ( [b (ni, Ui) (N (doi, pui, tpu, do) + niFi) + (1-b (ni, Ui) N (pui, pui, tpu, pu) ] ) and the second booking probability adjusted network improvement measure for the transportation request for using the base surge pricing ( [b (si, Ui) (N (doi, pui, tpu, do) + siFi) + (1-b (si, Ui) N (pui, pui, tpu, pu) ] ) as follows: Mi = [b (ni, Ui) (N (doi, pui, tpu, do) + niFi) + (1 -b (ni, Ui) N (pui, pui, tpu, pu) ] - [b (si, Ui) (N (doi, pui, tpu, do) + siFi) + (1-b (si, Ui) N (pui, pui, tpu, pu) ] (Equation 6)
[0127] As can be seen from Equation (6) , the first booking probability adjusted network improvement measure comprises a first component relating to the network improvement measure for the transportation request for using the candidate surge pricing ni for the probability that it is booked and a second component relating to the network improvement measure for the transportation request for using the candidate surge pricing ni for the probability that it is not booked. Similarly, the second booking probability adjusted network improvement measure comprises a first component relating to the network improvement measure for the transportation request for using the base surge pricing si for the probability that it is booked and a second component relating to the network improvement measure for the transportation request for using the base surge pricing si for the probability that it is not booked. Accordingly, the net booking probability adjusted network improvement measure Mi for the transportation request for using the candidate surge pricing ni would be positive if the first booking probability adjusted network improvement measures for the transportation request for using the candidate surge pricing ni is higher than the second booking probability adjusted network improvement measures for the transportation request for using the base surge pricing si.
[0128] Step 8: determine the surge pricing for the transportation request based on surge adjustment conditions, such as based on the following example two threshold conditions: If |Vi-αi| >ψ and Mi > 0, then ni= si + δi Else ni = si (Equation 7)
[0129] Accordingly, whether to implement the candidate surge pricing ni for the transportation request is determined based on a first surge adjustment condition configured based on the base network improvement measure Vi and the value of the booking conversion control parameter αi for the transportation request and a second surge adjustment condition configured based on the net booking probability adjusted network improvement measure Mi for the transportation request for using the candidate surge pricing ni. For example, the first surge adjustment condition is satisfied if a magnitude of a difference between the base network improvement measure and the value of the booking conversion control parameter for the transportation request (e.g., |Vi-αi| ) satisfies a first predetermined threshold (e.g., > ψ) . The second surge adjustment condition is satisfied if the net booking probability adjusted network improvement measure Mi for the transportation request for using the candidate surge pricing ni satisfies a second predetermined threshold (e.g., > 0) . For example, if the candidate surge pricing ni is adopted, it may be referred to as the network value-based individual surge (NVIS) pricing.
[0130] The above example method will now be explained further according to various example embodiments of the present invention.
[0131] In Step 1, the BTC parameter provides the budget for the number of transportation bookings for a geographical region (e.g., geohash) over a predefined time period (e.g., 60 seconds) including the pick-up time and this parameter serves as an input for the demand pacing described hereinbefore according to various example embodiments. For example, the BTC parameter may be determined from an upstream module.
[0132] In Step 2, γ refers to a time discount factor or value which may be applied to a future network value (e.g., a default value may be 0.99) , such as applied to the network value obtained for the drop-off location doi and the drop-off time. In this regard, the time refers to the transit time taken to travel from the pickup location pui to the drop-off location doi. Therefore, the time discount factor γ seeks to discount the network value obtained with respect to the drop-off location doi at the drop-off time with respect to the pick-up time (or with respect to the transit time from the pickup location pui to the drop-off location doi) . In this regard, the time discount factor γ may be set as desired or as appropriate.
[0133] Step 4 corresponds to the implementation of the PI or PID controller for the rule-based online thresholding demand pacing described hereinbefore according to various example embodiments. In Equation (4) , αm denotes the initial booking conversion control parameter for the predefined time period (e.g., 60 seconds) , which is the rolling average of the past 30 minutes of the base network improvement measures Vi for previous transportation requests for the pick-up location pui. For example, the booking conversion control parameter αm may also be a historical feature based on offiine data that is averaged over similar time periods. In Step 4, the updating of the booking conversion control parameter αi is performed over the predefined time period (e.g., 60 seconds) . η denotes a tunable parameter for balancing between the budget for the target number of bookings-to-convert (BTC) used against the time τi passed within the predefined time period thus far. BCS denotes the transportation booking count signal which is a real time count of the transportation bookings that have been converted. BCS / BTC is the budget ratio that indicates how much of the budget afforded by the BTC parameter has been used so far. τi denotes the amount of time (e.g., number of seconds) passed so far in the predefined time period. Accordingly, by updating the booking conversion control parameter αi according to various example embodiments, for example, the more quickly the budget (BTC) is used, the higher the threshold (booking conversion control parameter αi) for converting transportation requests to actual bookings may be so as to cause less transportation requests to be converted into actual bookings over the remainder of the predefined time period, and vice versa. For example, in Equation (4) above, η (BCS / BTC) -τi / 60 may be operate as a control error under the PI or PID controller framework.
[0134] In Step 5, according to the manner in which the surge pricing adjustment δi is determined in Equation (5) , for example, if the base network improvement measure Vi of the transportation request exceeds the booking conversion control parameter αi, the transportation request may be favored by virtue of the surge pricing adjustment δi having a larger negative value (since the surge pricing adjustment δi is determined according to a function of the difference between Vi and αi) , thereby decreasing the original or base surge pricing si value by the surge pricing adjustment δi. In Equation (5) , δmax and δmin denote predefined parameters for limiting the maximum and minimum changes from the original surge pricing si, respectively.
[0135] In step 7, the net booking probability adjusted network improvement measure Mi is determined according to Equation (6) such that, for example, it has a positive value if the transportation request for using the candidate surge pricing ni contributes positively to the network revenue of the transportation service network compared with that of the transportation request for using the original surge pricing si, taking into account the booking probability b (ni, Ui) for the transportation request for using the candidate surge pricing.
[0136] In Step 8, the surge pricing (final surge pricing) for the transportation request is set as the candidate surge pricing ni if both surge adjustment conditions are satisfied, otherwise, the surge pricing (final surge pricing) for the transportation request is set as the original surge pricing si. Advantageously, the example method is simple to implement as well as being fast to serve online.
[0137] In various example embodiments, to evaluate the method 600 of controlling a surge pricing for a transportation request based on the network contribution measure determining module (or component) 608, the network revenue determining module (or component) 610 and the demand pacing module (or component) 612, a simulation (e.g., by a simulation module) of the method 600 is carried out. In particular, according to various example embodiments, a simulation module is built for selecting or optimizing (or tuning) various parameters (or hyperparameters) utilized in the method 600.
[0138] FIG. 9 depicts a schematic flow diagram for an example simulation of the method 600 for each predefined time period (e.g., for each 1-minute time period) . In the example simulation, the 1-minute segment simulations can be run in a historical consecutive manner (i.e., one-minute after another in historical order) or can be drawn at random which can be beneficial for agent learning in reinforcement learning (RL) . For example, the simulation module built may include the following components / operations:
[0139] ● the BTR model
[0140] ● inter-arrival time model, which can be fitted using the Generalized Linear Model (or more specifically, the exponential one)
[0141] ● gather surge data from either historical data or simulation results
[0142] ● network value results from machine learning flow (e.g., an example screenshot of network value results are shown in FIG. 10)
[0143] For the selection ofhyperparameters, for example, either a grid search or Bayesian optimisation may be conducted over a predefined search space for λ, η, γ and so on. A parameter set is selected which maximises the simulated objective, which may be maximizing the GMV or maximizing the number of transportation bookings (or rides) .
[0144] The simulation results show the total expected GMV and total expected rides across all the transportation requests for a given city and date. For example, for each transportation request, the expected revenue (or potential revenue) can be simulated by multiplying the probability of booking (given by BTR (p_t) in FIG. 9) with the price p_t. The expected number of rides is given by the BTR (p_t) . The expected revenue and the expected number of rides may then be summed together to obtain the total expected GMV and total expected number of rides, respectively.
[0145] While embodiments of the invention have been particularly shown and described with reference to specific embodiments, it should be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the scope of the invention as defined by the appended claims. The scope of the invention is thus indicated by the appended claims and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced.
Claims
1.A method of controlling a surge pricing for a transportation request to a transportation service network using at least one processor, the method comprising:receiving a transportation request having associated therewith a pick-up location, a drop-off location and a pick-up time for a transportation;obtaining a first network contribution measure for the transportation request with respect to the drop-off location, the first network contribution measure indicating a contribution measure of the transportation request to the transportation service network with respect to the drop-off location;determining a base network improvement measure for the transportation request for using a base surge pricing based on the base surge pricing and the first network contribution measure for the transportation request;setting a value of a booking conversion control parameter for the transportation request, the booking conversion control parameter being configured to control a threshold for implementing a surge pricing adjustment to the base surge pricing for the transportation request;determining a candidate surge pricing for the transportation request based on the base surge pricing and the surge pricing adjustment for the transportation request, the surge pricing adjustment being determined based on the base network improvement measure and the value of the booking conversion control parameter for the transportation request;determining a first booking probability for the transportation request for using the candidate surge pricing based on the pick-up location, the drop-off location and the pick-up time for the transportation and the candidate surge pricing for the transportation request;determining a first booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing based on the first booking probability, the first network contribution measure and the candidate surge pricing for the transportation request; anddetermining the surge pricing for the transportation request based on the value of the booking conversion control parameter and the first booking probability adjusted network improvement measure for the transportation request.2.The method according to claim 1, further comprising obtaining a second network contribution measure for the transportation request with respect to the pick-up location, the second network contribution measure indicating the contribution measure of the transportation request to the transportation service network with respect to the pick-up location, wherein the base network improvement measure for the transportation request for using the base surge pricing is determined further based on the second network contribution measure for the transportation request.3.The method according to claim 1, wherein the contribution measure of the transportation request to the transportation service network is with respect to an efficiency of the transportation service network.4.The method according to claim 1, wherein the value of the booking conversion control parameter for the transportation request is determined based on a target number of transportation bookings for the pick-up location in a predefined time period including the pick-up time.5.The method according to claim 4, wherein the value of the booking conversion control parameter for the transportation request is determined further based on a number of transportation bookings that has been made for the pick-up location in the predefined time period and based on an amount of time lapsed in the predefined time period.6.The method according to claim 4, wherein the value of the booking conversion control parameter for the transportation request is determined based on a difference between a ratio between the number of transportation bookings that has been made and the target number of transportation bookings for the pick-up location in the predefined time period and a ratio between the amount of time lapsed in the predefined time period and the predefined time period.7.The method according to claim 1, wherein the surge pricing adjustment for the transportation request is determined based on a difference between the base network improvement measure and the value of the booking conversion control parameter for the transportation request.8.The method according to claim 1, wherein the first booking probability for the transportation request for using the candidate surge pricing is determined based on the pick-up location, the drop-off location and the pick-up time for the transportation and the candidate surge pricing for the transportation request using on a trained booking probability machine learning model.9.The method according to claim 1, further comprising:determining a second booking probability for the transportation request for using the base surge pricing based on the pick-up location, the drop-off location and the pick-up time for the transportation and the base surge pricing for the transportation request;determining a second booking probability adjusted network improvement measure for the transportation request for using the base surge pricing based on the second booking probability, the first network contribution measure and the base surge pricing for the transportation request; anddetermining a net booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing based on the first and second booking probability adjusted network improvement measures for the transportation request,wherein the surge pricing for the transportation request is determined based on the value of the booking conversion control parameter and the net booking probability adjusted network improvement measure for the transportation request.10.The method according to claim 9, wherein said determining the surge pricing for the transportation request comprises determining whether to implement the candidate surge pricing for the transportation request based on a first surge adjustment condition configured based on the base network improvement measure and the booking conversion control parameter for the transportation request and a second surge adjustment condition configured based on the net booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing.11.The method according to claim 10, wherein said determining whether to implement the candidate surge pricing for the transportation request comprises determining to implement the candidate surge pricing for the transportation request based on determining that the first and second surge adjustment conditions are satisfied.12.The method according to claim 11, whereinthe first surge adjustment condition is satisfied if a magnitude of a difference between the base network improvement measure and the value of the booking conversion control parameter for the transportation request satisfies a first predetermined threshold; andthe second surge adjustment condition is satisfied if the net booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing satisfies a second predetermined threshold.13.The method according to claim 1, wherein the transportation request to the transportation service network is for a ride-hailing service or a delivery service.14.A system for controlling a surge pricing for a transportation request to a transportation service network, the system comprising:at least one memory; andat least one processor communicatively coupled to the at least one memory and configured to:receive a transportation request having associated therewith a pick-up location, a drop-off location and a pick-up time for a transportation;obtain a first network contribution measure for the transportation request with respect to the drop-off location, the first network contribution measure indicating a contribution measure of the transportation request to the transportation service network with respect to the drop-off location;determine a base network improvement measure for the transportation request for using a base surge pricing based on the base surge pricing and the first network contribution measure for the transportation request;set a value of a booking conversion control parameter for the transportation request, the booking conversion control parameter being configured to control a threshold for implementing a surge pricing adjustment to the base surge pricing for the transportation request;determine a candidate surge pricing for the transportation request based on the base surge pricing and the surge pricing adjustment for the transportation request, the surge pricing adjustment being determined based on the base network improvement measure and the value of the booking conversion control parameter for the transportation request;determine a first booking probability for the transportation request for using the candidate surge pricing based on the pick-up location, the drop-off location and the pick-up time for the transportation and the candidate surge pricing for the transportation request;determine a first booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing based on the first booking probability, the first network contribution measure and the candidate surge pricing for the transportation request; anddetermine the surge pricing for the transportation request based on the value of the booking conversion control parameter and the first booking probability adjusted network improvement measure for the transportation request.15.The system according to claim 14, wherein the at least one processor is further configured to obtain a second network contribution measure for the transportation request with respect to the pick-up location, the second network contribution measure indicating the contribution measure of the transportation request to the transportation service network with respect to the pick-up location, wherein the base network improvement measure for the transportation request for using the base surge pricing is determined further based on the second network contribution measure for the transportation request.16.The system according to claim 14, wherein the contribution measure of the transportation request to the transportation service network is with respect to an efficiency of the transportation service network.17.The system according to claim 14, wherein the value of the booking conversion control parameter for the transportation request is determined based on a target number of transportation bookings for the pick-up location in a predefined time period including the pick-up time.18.The system according to claim 17, wherein the value of the booking conversion control parameter for the transportation request is determined further based on a number of transportation bookings that has been made for the pick-up location in the predefined time period and based on an amount of time lapsed in the predefined time period.19.The system according to claim 17, wherein the value of the booking conversion control parameter for the transportation request is determined based on a difference between a ratio between the number of transportation bookings that has been made and the target number of transportation bookings for the pick-up location in the predefined time period and a ratio between the amount of time lapsed in the predefined time period and the predefined time period.20.The system according to claim 14, wherein the surge pricing adjustment for the transportation request is determined based on a difference between the base network improvement measure and the value of the booking conversion control parameter for the transportation request.21.The system according to claim 14, wherein the first booking probability for the transportation request for using the candidate surge pricing is determined based on the pick-up location, the drop-off location and the pick-up time for the transportation and the candidate surge pricing for the transportation request using on a trained booking probability machine learning model.22.The system according to claim 14, wherein the at least one processor is further configured to:determine a second booking probability for the transportation request for using the base surge pricing based on the pick-up location, the drop-off location and the pick-up time for the transportation and the base surge pricing for the transportation request;determine a second booking probability adjusted network improvement measure for the transportation request for using the base surge pricing based on the second booking probability, the first network contribution measure and the base surge pricing for the transportation request; anddetermine a net booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing based on the first and second booking probability adjusted network improvement measures for the transportation request,wherein the surge pricing for the transportation request is determined based on the value of the booking conversion control parameter and the net booking probability adjusted network improvement measure for the transportation request.23.The system according to claim 22, wherein said determine the surge pricing for the transportation request comprises determining whether to implement the candidate surge pricing for the transportation request based on a first surge adjustment condition configured based on the base network improvement measure and the booking conversion control parameter for the transportation request and a second surge adjustment condition configured based on the net booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing.24.The system according to claim 23, wherein said determining whether to implement the candidate surge pricing for the transportation request comprising determining to implement the candidate surge pricing for the transportation request based on determining that the first and second surge adjustment conditions are satisfied.25.The system according to claim 24, whereinthe first surge adjustment condition is satisfied if a magnitude of a difference between the base network improvement measure and the value of the booking conversion control parameter for the transportation request satisfies a first predetermined threshold; andthe second surge adjustment condition is satisfied if the net booking probability adjusted network improvement measure for the transportation request for using the candidate surge pricing satisfies a second predetermined threshold.26.The system according to claim 14, wherein the transportation request to the transportation service network is for a ride-hailing service or a delivery service.27.A computer program product, embodied in one or more non-transitory computer-readable storage mediums, comprising instructions executable by at least one processor to perform the method of controlling a surge pricing for a transportation request to a transportation service network according to claim 1.
Citation Information
Patent Citations
Auction mechanism-based online car-hailing real-time service vehicle resource distribution and pricing method
CN108564189A
System and method for geo-aware transportation billing verification
CN110800007A
Estimating taxi fare
US20140074757A1
Dynamic pricing systems and related methods
US20170154348A1
Self-adjusting machine learning system and methodology for prediction of auto shipping prices
US20240185172A1