Using a predictive request model to optimize provider resources

A computing device comprising:

• at least one processor;
• and
• at least one non-transitory computer-readable storage medium storing instructions thereon that, when executed by the at least one processor, cause the computing device to:
• determine a predicted gap between predicted requestor time and predicted provider time in a geographic area during a future time period;
• generate a provider incentive metric for the geographic area during the future time period from the predicted gap utilizing a predictive request model, wherein the predictive request model is trained to generate provider incentive metrics indicating provider incentives to increase transportation provider activity in a given geographic area for a given future time period from gap metrics;
• and
• generate an interactive map associated with a customized schedule comprising:
• a geographic area overlay outlining the geographic area within the interactive map, and
• the generated provider incentive metric for the future time period overlaid on the geographic area overlay.

A non-transitory computer-readable medium storing instructions thereon that, when executed by at least one processor, cause a system to:

• determine a predicted gap between expected request activity and expected transportation provider activity for a geographic area during a future time period;
• utilize a predictive request model and the predicted gap to generate a provider incentive metric for the geographic area during the future time period, wherein the predictive request model is trained to generate provider incentive metrics indicating provider incentives to increase transportation provider activity in a given geographic area for a given future time period;
• generate an interactive map associated with a customized schedule for the geographic area and the future time period based on the generated provider incentive metric;
• detect a deficit in a training dataset for training the predictive request model;
• initiate a data collection experiment by:
• randomizing a plurality of provider incentives for the given geographic area and a given time period;
• and
• tracking results of the plurality of provider incentives;
• and
• including the results in the training dataset.