Computing estimated value of providing service among geographical regions

A non-transitory computer-readable storage medium storing computer program instructions executable by one or more processors of a system to perform steps comprising:

• selecting a sequence of time periods from a set of time periods, the sequence of time periods being later than a current time, each time period having an associated set of geographical regions;
• traversing the sequence of time periods in reverse chronological order to generate a value estimate, wherein a traversal from a later time period in the sequence of time periods to an adjacent earlier time period in the sequence of time periods comprises:
• for each geographical region in the set of geographical regions associated with the earlier time period:
• generating an accumulated value based on one or more of:
• an expected value of a transition from the geographical region associated with the earlier time period to the geographical region associated with the later time period, the expected value generated using a machine learning model;
• a probability of the transition from the geographical region associated with the earlier time period to the geographical region associated with the later time period, the probability generated using a machine learning model;
• and
• an accumulated value associated with the later time period;
• subsequent to the traversal of the sequence of time periods, selecting the value estimate from one of the generated accumulated values associated with an earliest time period in the sequence of time periods;
• and
• presenting the value estimate to a user of a user device.

A computer system comprising:

• one or more computer processors for executing computer program instructions;
• and
• a non-transitory computer-readable storage medium storing instructions executable by the one or more computer processors to perform steps comprising:
• selecting a sequence of time periods from a set of time periods, the sequence of time periods being later than a current time, each time period having an associated set of geographical regions;
• traversing the sequence of time periods in reverse chronological order to generate a value estimate, wherein a traversal from a later time period in the sequence of time periods to an adjacent earlier time period in the sequence of time periods comprises:
• for each geographical region in the set of geographical regions associated with the earlier time period:
• generating an accumulated value based on one or more of:
• an expected value of a transition from the geographical region associated with the earlier time period to the geographical region associated with the later time period, the expected value generated using a machine learning model;
• a probability of the transition from the geographical region associated with the earlier time period to the geographical region associated with the later time period, the probability generated using a machine learning model;
• and
• an accumulated value associated with the later time period;
• subsequent to the traversal of the sequence of time periods, selecting the value estimate from one of the generated accumulated values associated with an earliest time period in the sequence of time periods;
• and
• presenting the value estimate to a user of a user device.