Granted Patent B1

US 12,737,734 · App. 18/974,600 · Granted Sep 15, 2026

Inventors: Charles W. Fastner (Pewaukee, WI); Timothy L. Ernst (Grafton, WI)

Assignee: Direct Supply, Inc.

G06Q10/20G06Q30/0645

Abstract

Claims (69)

1 . A computer-implemented method for predicting an occupancy schedule for a unit in a facility, comprising:

subsequent to the unit in the facility being vacated by a resident, receiving an input from a first user via a first computing device, the input detailing a condition of the unit in the facility;

generating a current grade for the unit in the facility based on the input detailing the condition of the unit in the facility;

identifying a target grade and a target time-to-occupancy for the unit in the facility;

determining, using one or more predictive machine learning (ML) models, a maintenance task for completion to modify the unit in the facility such that the unit in the facility can be associated with the target grade instead of the current grade, based on the one or more predictive ML models generating an output in response to receiving an input that includes the maintenance task, the output including a predicted time-to-occupancy resulting from the completion of the maintenance task or one or more predicted likelihoods that the completion of the maintenance task will result in one or more target time-to-occupancies,

wherein the one or more predictive ML models are trained using a dataset that includes outcome data including actual time-to-occupancies associated with historical maintenance tasks; and

in response to the output of the one or more predictive ML models satisfying the target time-to-occupancy, providing the maintenance task to a second user via a second computing device such that the second user can complete the maintenance task to modify the unit in the facility such that the unit in the facility can be associated with the target grade instead of the current grade and can satisfy the target time-to-occupancy.

2 . The method of claim 1 , wherein receiving the input from the first user detailing the condition of the unit in the facility comprises receiving images of the unit in the facility.

3 . The method of claim 1 , wherein generating the current grade for the unit in the facility based on the input detailing the condition of the unit in the facility comprises generating the current grade for the unit in the facility using the one or more predictive ML models.

4 . The method of claim 1 , comprising:

generating a first unit turn plan comprising a first target grade for the unit in the facility and a first maintenance task for completion to modify the unit in the facility such that the unit in the facility can be associated with the first target grade instead of the current grade;

generating a second unit turn plan comprising a second target grade for the unit in the facility and a second maintenance task for completion to modify the unit in the facility such that the unit in the facility can be associated with the second target grade instead of the current grade;

providing the first unit turn plan and the second unit turn plan to a third user via a third user device; and

receiving a selection of the first unit turn plan from the third user via the third user device;

wherein identifying the target grade for the unit in the facility comprises identifying the target grade as the first target grade responsive to receiving the selection of the first unit turn plan from the third user via the third user device; and

wherein determining the maintenance task for completion comprises determining the maintenance task to be the first maintenance task responsive to receiving the selection of the first unit turn plan from the third user via the third user device.

5 . The method of claim 1 , comprising:

applying the current grade for the unit in the facility and the economic data associated with the location of the facility as input to the one or more predictive ML models; and

determining a recommended rent level based on an output of the one or more predictive ML models;

wherein the target rent for the unit desired by the stakeholder is the recommended rent level.

6 . The method of claim 1 , wherein providing the maintenance task to the second user via the second computing device comprises providing the maintenance task to an internal staff member associated with the facility.

7 . The method of claim 1 , wherein the first user is the same as the second user.

8 . The method of claim 1 , wherein the predicted time-to-occupancy includes a predicted time until the unit is ready for a new resident.

9 . The method of claim 1 , comprising:

providing an indication of the target grade for the unit in the facility and the maintenance task for completion to modify the unit in the facility to a stakeholder associated with the facility for approval via a third computing device; and

receiving a second input from the stakeholder associated with the facility from the third computing device, the second input comprising an approval of the target grade for the unit in the facility and the maintenance task for completion to modify the unit in the facility;

wherein providing the maintenance task to the second user via the second computing device comprises providing the maintenance task to the second user via the second computing device responsive to receiving the approval.

10 . The method of claim 9 , comprising:

determining that the predicted time-to-occupancy included in the output generated by the one or more predictive ML models associated with the maintenance task exceeds a threshold amount of time;

wherein providing the indication of the target grade and the maintenance task to the stakeholder for approval via the third computing device comprises providing the indication of the target grade and the maintenance task to the stakeholder for approval responsive to determining that the predicted time-to-occupancy included in the output generated by the one or more predictive ML models associated with the maintenance task exceeds the threshold amount of time.

11 . The method of claim 1 , wherein determining the maintenance task for completion to modify the unit in the facility comprises:

applying the current grade for the unit in the facility and the target grade for the unit in the facility as input to the one or more predictive ML models;

determining a unit turn plan for the unit in the facility based on an output of the one or more predictive ML models; and

identifying the maintenance task as being part of the unit turn plan.

12 . The method of claim 11 , comprising:

receiving an updated grade for the unit in the facility after completion of the maintenance task; and

training the one or more predictive ML models based on the updated grade.

13 . The method of claim 1 , wherein:

receiving the target time-to-occupancy for the unit desired by a stakeholder; and

identifying the target grade for the unit in the facility comprises applying the preference data and the economic data as input to the one or more predictive ML models.

14 . The method of claim 13 , comprising:

receiving indication of an actual time-to-occupancy associated with a resident that moves into the unit in the facility; and

training the one or more predictive ML models based on the actual time-to-occupancy.

15 . The method of claim 1 , comprising:

prompting a third user to provide a human grade for the unit in the facility based on the input detailing a condition of the unit in the facility; and

receiving a second input from the third user via a third computing device, the second input comprising the human grade for the unit in the facility;

wherein generating the current grade for the unit in the facility comprises generating the current grade for the unit in the facility based on the human grade for the unit in the facility.

16 . The method of claim 15 , wherein generating the current grade for the unit in the facility comprises generating the current grade for the unit in the facility by applying the input detailing the condition of the unit in the facility and the human grade for the unit in the facility as input to the one or more predictive ML models.

17 . The method of claim 16 , comprising:

receiving an indication of an actual rent received from a new resident that moves into the unit in the facility;

receiving an indication of an actual time-to-occupancy associated with the new resident that moves into the unit in the facility; and

training the one or more predictive ML models based on the actual time-to-occupancy.

18 . One or more non-transitory computer-readable storage medium having instructions stored thereon that, when executed by processing circuitry, cause the processing circuitry to:

subsequent to a unit in a facility being vacated by a resident, receive an input from a first user via a first computing device, the input detailing a condition of the unit in the facility;

generate a current grade for the unit in the facility based on the input detailing the condition of the unit in the facility;

identify a target grade and a target time-to-occupancy for the unit in the facility;

determine, using one or more predictive ML models, a maintenance task for completion to modify the unit in the facility such that the unit in the facility can be associated with the target grade instead of the current grade, based on the one or more predictive ML models generating an output in response to receiving an input that includes the maintenance task, the output including a predicted time-to-occupancy resulting from the completion of the maintenance task or one or more predicted likelihoods that the completion of the maintenance task will result in one or more target time-to-occupancies,

wherein the one or more predictive ML models are trained using a dataset that includes outcome data including actual time-to-occupancies associated with historical maintenance tasks; and

in response to the output of the one or more predictive ML models satisfying the target time-to-occupancy, provide the maintenance task to a second user via a second computing device such that the second user can complete the maintenance task to modify the unit in the facility such that the unit in the facility can be associated with the target grade instead of the current grade and can satisfy the target time-to-occupancy.

19 . A system for predicting an occupancy schedule for a unit in a facility, comprising:

memory comprising machine-readable instructions; and

processing circuitry to execute the machine-readable instructions to:

subsequent to the unit in the facility being vacated by a resident, receive an input from a first user via a first computing device, the input detailing a condition of the unit in the facility;

generate a current grade for the unit in the facility based on the input detailing the condition of the unit in the facility;

identify a target grade and a target time-to-occupancy for the unit in the facility;

determine, using one or more predictive machine learning (ML) models, a maintenance task for completion to modify the unit in the facility such that the unit in the facility can be associated with the target grade instead of the current grade, based on the one or more predictive ML models generating an output in response to receiving an input that includes the maintenance task, the output including a predicted time-to-occupancy resulting from the completion of the maintenance task or one or more predicted likelihoods that the completion of the maintenance task will result in one or more target time-to-occupancies,

wherein the one or more predictive ML models are trained using a dataset that includes outcome data including actual time-to-occupancies associated with historical maintenance tasks; and

in response to the output of the one or more predictive ML models satisfying the target time-to-occupancy, provide the maintenance task to a second user via a second computing device such that the second user can complete the maintenance task to modify the unit in the facility such that the unit in the facility can be associated with the target grade instead of the current grade and can satisfy the target time-to-occupancy.

20 . The system of claim 19 , wherein the predicted time-to-occupancy includes a predicted time until the unit is ready for a new resident.

Assignments (1)

Continuity (1)

References Cited (51)