Generating and communicating device balance graphical representations for a dynamic transportation system

Abstract

The present disclosure relates to systems, non-transitory computer-readable media, and methods for generating and utilizing provider device balance graphs reflecting device utilization ratio between provider devices and requester devices. In particular, in one or more embodiments, the disclosed systems determine and utilize probabilities to generate and provide provider device balance graphs based on probability distributions for device utilization ratios. The disclosed systems can apply various scaling models to address seasonality, special events, and/or differences between regions. The disclosed systems can also apply thresholds to generate device incentive graphs that more efficiently deploy provider devices across a transportation matching system.

What is claimed is: 
     
         1 . A computer-implemented method comprising:
 monitoring, by one or more servers of a transportation matching system, historical metrics from a plurality of requester computing devices and a plurality of provider computing devices via a communications network;   generating, by the one or more servers and based on the historical metrics:
 forecasted numbers of requester computing devices across time periods, and 
 projected earnings metrics for provider computing devices across the time periods; 
   generating, by the one or more servers and based on the forecasted numbers of requester computing devices and the projected earnings metrics, combined earnings-ridership metrics for the time periods; and   providing, for display to a user interface of a provider computing device, a provider device incentive graph comprising the combined earnings-ridership metrics for the time periods.   
     
     
         2 . The computer-implemented method of  claim 1 , wherein generating the forecasted numbers of requester computing devices and the projected earnings metrics comprises utilizing one or more machine learning models to generate the forecasted numbers of requester computing devices and the projected earnings metrics based on the historical metrics. 
     
     
         3 . The computer-implemented method of  claim 1 , further comprising generating the forecasted numbers of requester computing devices across the time periods and the projected earnings metrics across the time periods by generating a first forecasted number of requester computing devices for a first time period and generating a first projected earnings metrics for the first time period. 
     
     
         4 . The computer-implemented method of  claim 3 , wherein generating the combined earnings-ridership metrics comprises, for a first time period of the time periods combining the first projected earnings metrics with a radical of the first forecasted number of requester computing devices. 
     
     
         5 . The computer-implemented method of  claim 1 , further comprising:
 generating a probability distribution of the combined earnings-ridership metrics;   mapping the combined earnings-ridership metrics to probability ranges based on the probability distribution; and   generating the provider device incentive graph from the probability ranges.   
     
     
         6 . The computer-implemented method of  claim 5 , further comprising generating the provider device incentive graph by applying a minimum threshold or a maximum threshold to the probability ranges. 
     
     
         7 . The computer-implemented method of  claim 1 , further comprising generating a suggested provider schedule for the provider computing device based on the combined earnings-ridership metrics. 
     
     
         8 . The computer-implemented method of  claim 7 , further comprising providing, for display via the user interface of the provider computing device, a scheduling indicator via the provider device incentive graph indicating the suggested provider schedule. 
     
     
         9 . A non-transitory computer-readable medium storing instructions that, when executed by at least one processor, cause a computer system to:
 monitor historical metrics from a plurality of requester computing devices and a plurality of provider computing devices via a communications network;   generate, based on the historical metrics:
 forecasted numbers of requester computing devices across time periods, and 
 projected earnings metrics for provider computing devices across the time periods; 
   generate, based on the forecasted numbers of requester computing devices and the projected earnings metrics, combined earnings-ridership metrics for the time periods; and   provide, for display to a user interface of a provider computing device, a provider device incentive graph comprising the combined earnings-ridership metrics for the time periods.   
     
     
         10 . The non-transitory computer-readable medium of  claim 9 , further comprising instructions that, when executed by at least one processor, cause the computer system to generate the forecasted numbers of requester computing devices and the projected earnings metrics by utilizing one or more machine learning models to generate the forecasted numbers of requester computing devices and the projected earnings metrics based on the historical metrics. 
     
     
         11 . The non-transitory computer-readable medium of  claim 9 , further comprising instructions that, when executed by at least one processor, cause the computer system to generate the combined earnings-ridership metrics by combining the projected earnings metrics with radicals of the forecasted numbers of requester computing devices. 
     
     
         12 . The non-transitory computer-readable medium of  claim 9 , further comprising instructions that, when executed by at least one processor, cause the computer system to:
 generate a probability distribution of the combined earnings-ridership metrics;   map the combined earnings-ridership metrics to probability ranges based on the probability distribution; and   generate the provider device incentive graph from the probability ranges.   
     
     
         13 . The non-transitory computer-readable medium of  claim 9 , further comprising instructions that, when executed by at least one processor, cause the computer system to generate a suggested provider schedule for the provider computing device based on the combined earnings-ridership metrics. 
     
     
         14 . The non-transitory computer-readable medium of  claim 13 , further comprising instructions that, when executed by at least one processor, cause the computer system to provide, for display via the user interface of the provider computing device, a scheduling indicator via the provider device incentive graph indicating the suggested provider schedule. 
     
     
         15 . A system comprising:
 at least one processor; and   at least one non-transitory computer-readable storage medium storing instructions that, when executed by the at least one processor, cause the system to:
 monitor historical metrics from a plurality of requester computing devices and a plurality of provider computing devices via a communications network; 
 generate, based on the historical metrics:
 forecasted numbers of requester computing devices across time periods, and 
 projected earnings metrics for provider computing devices across the time periods; 
 
 generate, based on the forecasted numbers of requester computing devices and the projected earnings metrics, combined earnings-ridership metrics for the time periods; and 
 provide, for display to a user interface of a provider computing device, a provider device incentive graph comprising the combined earnings-ridership metrics for the time periods. 
   
     
     
         16 . The system of  claim 15 , further comprising instructions that, when executed by the at least one processor, cause the system to generate the forecasted numbers of requester computing devices and the projected earnings metrics by utilizing one or more machine learning models to generate the forecasted numbers of requester computing devices and the projected earnings metrics based on the historical metrics. 
     
     
         17 . The system of  claim 15 , further comprising instructions that, when executed by the at least one processor, cause the system to generate the combined earnings-ridership metrics by combining the projected earnings metrics with radicals of the forecasted numbers of requester computing devices. 
     
     
         18 . The system of  claim 15 , further comprising instructions that, when executed by the at least one processor, cause the system to:
 generate a probability distribution of the combined earnings-ridership metrics;   map the combined earnings-ridership metrics to probability ranges based on the probability distribution; and   generate the provider device incentive graph from the probability ranges.   
     
     
         19 . The system of  claim 15 , further comprising instructions that, when executed by the at least one processor, cause the system to generate a suggested provider schedule for the provider computing device based on the combined earnings-ridership metrics. 
     
     
         20 . The system of  claim 19 , further comprising instructions that, when executed by the at least one processor, cause the system to provide, for display via the user interface of the provider computing device, a scheduling indicator via the provider device incentive graph indicating the suggested provider schedule.