Job value model generation method and system

A system, comprising:

• one or more processors;
• and
• a non-transitory computer-readable storage medium containing instructions which, when executed on the one or more processors, cause the one or more processors to perform operations including:
• receiving, based on the tenant receiving a first job request from a first customer, live data associated with the first job request, wherein the first job request comprises a first request that at least one service technician associated with the tenant perform the job at a job location associated with the first customer;
• preprocessing the live data to obtain a set of features associated with the first job request, the set of features consumable by the trained machine learning model;
• applying the trained machine learning model to the set of features to generate a first job value prediction associated with the first job request;
• sending the first generated job value prediction to the tenant;
• updating the trained machine learning model based at least in part on the first generated job value prediction and an actual job value determined based on the at least one service technician completing the job at the job location associated with the first customer;
• receiving, based on the tenant receiving a second job request from a second customer, live data associated with the second request, wherein the second job request comprises a second request that the at least one service technician associated with the tenant perform the job at a job location associated with the second customer;
• and
• based on the trained machine learning model being unable to generate a second job value prediction associated with the second job request using the live data associated with the second request within a predetermined time window, generating the second job value prediction associated with the second job request based on a statistics associated with job values of other job requests.

A non-transitory machine-readable storage medium, including instructions configured to cause a data processing apparatus to perform operations including:

• training a machine learning model to generate job value predictions associated with a tenant;
• receiving, based on the tenant receiving a first job request from a first customer, live data associated with the first job request, wherein the first job request comprises a first request that at least one service technician associated with the tenant perform the job at a job location associated with the first customer;
• preprocessing the live data to obtain a set of features associated with the first job request, the set of features consumable by the trained machine learning model;
• applying the trained machine learning model to the set of features to generate a first job value prediction associated with the first job request;
• sending the first generated job value prediction to the tenant;
• updating the trained machine learning model based at least in part on the first generated job value prediction and an actual job value determined based on the at least one service technician completing the job at the job location associated with the first customer;
• receiving, based on the tenant receiving a second job request from a second customer, live data associated with the second request, wherein the second job request comprises a second request that the at least one service technician associated with the tenant perform the job at a job location associated with the second customer;
• and
• based on the trained machine learning model being unable to generate a second job value prediction associated with the second job request using the live data associated with the second request within a predetermined time window, generating the second job value prediction associated with the second job request based on a statistics associated with job values of other job requests.