Systems and methods for performing constraint space partitioning
The disclosed computer-implemented method may match transportation requestor devices to transportation provider devices pre-request by using the same matching process employed by a matching engine that is capable of predicting transportation swaps and walks. Multiple matches may be showcased on user devices, and offline transportation providers may see an exact match to a requestor, with an ability to go online and accept the pre-request match. Additional techniques disclosed include the curation and presentation of offers using constraint space partitioning and adjustment of price and/or the presentation of offers based on real-time information to improve the efficiency and/or utilization of transportation provider resources.
1. A system comprising:
a non-transitory memory; and
one or more hardware processors configured to execute instructions from the non-transitory memory to perform operations comprising:
determining that a transportation requestor device has instantiated a transportation application that provides access to a transportation network;
prior to receiving a request for transportation via the transportation application, generating, based on real-time information about a current state of the transportation network, a set of pre-request matches between a transportation requestor device and transportation provider devices;
determining, from the set of pre-request matches, a subset of pareto-optimal pre-request matches, wherein pareto-optimality of the subset of pareto-optimal pre-request matches is according to a set of attributes of the matches, the subset of pareto-optimal pre-request matches being defined by one or more user-related data that are associated with the transportation requestor device;
generating, based on the subset of pareto-optimal pre-request matches, a list of pre-request matches; and
presenting the list of pre-request matches for display to the transportation requestor device via the transportation application.
2. The system of claim 1 , wherein the list of pre-request matches for display comprises a first offering and a second offering, wherein the first offering is pareto optimal relative to the second offering based on a first attribute, and the second offering is pareto optimal relative to the first offering based on a second attribute.
3. The system of claim 1 , wherein determining the subset of pareto-optimal pre-request matches comprises:
analyzing, by a dynamic transportation matching system for at least one match of the set of pre-request matches, a tradeoff between at least one transportation provider resource availability metric of the match and at least one transportation requestor efficiency constraint of the match;
partitioning, by the dynamic transportation matching system based on the analyzing, a constraint space for the at least one transportation requestor efficiency constraint; and
determining, by the dynamic transportation matching system, a value or range of the at least one transportation provider availability metric corresponding to a value or range of the at least one transportation requestor efficiency constraint based at least in part on the partitioning.
4. The system of claim 3 , wherein:
the analyzing the tradeoff comprises determining, for a median value, a fundamental cost difference and a fundamental change in ratio between the at least one transportation provider resource availability metric and the at least one transportation requestor efficiency constraint; and
the partitioning comprises separating the pre-request matches at the median value in response to determining that there is both a fundamental cost difference and a fundamental change in ratio.
5. The system of claim 3 , wherein:
the analyzing the tradeoff comprises iteratively evaluating each pre-request match and finding points at which a fundamental change in the resource availability metric is observed; and
partitioning the constraint space is performed at one or more points at which the fundamental change in the resource availability metric is observed.
6. The system of claim 3 , wherein the list of pre-request matches includes one or more ranges corresponding to the one or more partitions.
7. The system of claim 6 , wherein the operations further comprise determining the one or more ranges according to a clustering technique that at least one of:
groups all pre-request matches having a predetermined positional relationship with respect to a partition;
groups a threshold number of pre-request matches having the predetermined positional relationship with respect to the partition; or
groups pre-request matches having the predetermined positional relationship with respect to the partition in a manner that ensures a threshold range.
8. A computer-implemented method comprising:
determining that a transportation requestor device has instantiated a transportation application that provides access to a transportation network;
prior to receiving a request for transportation via the transportation application, generating, based on real-time information about a current state of the transportation network, a set of pre-request matches between a transportation requestor device and transportation provider devices;
determining, from the set of pre-request matches, a subset of pareto-optimal pre-request matches, wherein pareto-optimality of the subset of pareto-optimal pre-request matches is according to a set of attributes of the matches, the subset of pareto-optimal pre-request matches being defined by one or more user-related data that are associated with the transportation requestor device;
generating, based on the subset of pareto-optimal pre-request matches, a list of pre-request matches; and
presenting the list of pre-request matches for display to the transportation requestor device via the transportation application.
9. The method of claim 8 , wherein the list of pre-request matches for display comprises a first offering and a second offering, wherein the first offering is pareto optimal relative to the second offering based on a first attribute, and the second offering is pareto optimal relative to the first offering based on a second attribute.
10. The method of claim 8 , wherein determining the subset of pareto-optimal pre-request matches comprises:
analyzing, by a dynamic transportation matching system for at least one match of the set of pre-request matches, a tradeoff between at least one transportation provider resource availability metric of the match and at least one transportation requestor efficiency constraint of the match;
partitioning, by the dynamic transportation matching system based on the analyzing, a constraint space for the at least one transportation requestor efficiency constraint; and
determining, by the dynamic transportation matching system, a value or range of the at least one transportation provider availability metric corresponding to a value or range of the at least one transportation requestor efficiency constraint based at least in part on the partitioning.
11. The method of claim 10 , wherein:
the analyzing the tradeoff comprises determining, for a median value, a fundamental cost difference and a fundamental change in ratio between the at least one transportation provider resource availability metric and the at least one transportation requestor efficiency constraint; and
the partitioning comprises separating the pre-request matches at the median value in response to determining that there is both a fundamental cost difference and a fundamental change in ratio.
12. The method of claim 10 , wherein:
the analyzing the tradeoff comprises iteratively evaluating each pre-request match and finding points at which a fundamental change in the resource availability metric is observed; and
partitioning the constraint space is performed at one or more points at which the fundamental change in the resource availability metric is observed.
13. The method of claim 10 , wherein the list of pre-request matches includes one or more ranges corresponding to the one or more partitions.
14. The method of claim 13 , further comprising determining the one or more ranges according to a clustering technique that at least one of:
groups all pre-request matches having a predetermined positional relationship with respect to a partition;
groups a threshold number of pre-request matches having the predetermined positional relationship with respect to the partition; or
groups pre-request matches having the predetermined positional relationship with respect to the partition in a manner that ensures a threshold range.
15. A non-transitory computer-readable medium comprising:
computer-readable instructions that, when executed by at least one processor of a computing device, cause the computing device to:
determine that a transportation requestor device has instantiated a transportation application that provides access to a transportation network;
prior to receiving a request for transportation via the transportation application, based on real-time information about a current state of a transportation network, a set of pre-request matches between a transportation requestor device and transportation provider devices;
determine, from the set of pre-request matches, a subset of pareto-optimal pre-request matches, wherein pareto-optimality of the subset of pareto-optimal pre-request matches is according to a set of attributes of the matches, the subset of pareto-optimal pre-request matches being defined by one or more user-related data that are associated with the transportation requestor device;
generate, based on the subset of pareto-optimal pre-request matches, a list of pre-request matches; and
present the list of pre-request matches for display to the transportation requestor device via the transportation application.
16. The non-transitory computer-readable medium of claim 15 , wherein the list of pre-request matches for display comprises a first offering and a second offering, wherein the first offering is pareto optimal relative to the second offering based on a first attribute, and the second offering is pareto optimal relative to the first offering based on a second attribute.
17. The non-transitory computer-readable medium of claim 15 , wherein determining the subset of pareto-optimal pre-request matches comprises:
analyzing, by a dynamic transportation matching system for at least one match of the set of pre-request matches, a tradeoff between at least one transportation provider resource availability metric of the match and at least one transportation requestor efficiency constraint of the match;
partitioning, by the dynamic transportation matching system based on the analyzing, a constraint space for the at least one transportation requestor efficiency constraint; and
determining, by the dynamic transportation matching system, a value or range of the at least one transportation provider availability metric corresponding to a value or range of the at least one transportation requestor efficiency constraint based at least in part on the partitioning.
18. The non-transitory computer-readable medium of claim 17 , wherein:
the analyzing the tradeoff comprises determining, for a median value, a fundamental cost difference and a fundamental change in ratio between the at least one transportation provider resource availability metric and the at least one transportation requestor efficiency constraint; and
the partitioning comprises separating the pre-request matches at the median value in response to determining that there is both a fundamental cost difference and a fundamental change in ratio.
19. The non-transitory computer-readable medium of claim 17 , wherein:
the analyzing the tradeoff comprises iteratively evaluating each pre-request match and finding points at which a fundamental change in the resource availability metric is observed; and
partitioning the constraint space is performed at one or more points at which the fundamental change in the resource availability metric is observed.
20. The non-transitory computer-readable medium of claim 17 , wherein the list of pre-request matches includes one or more ranges corresponding to the one or more partitions.