Spatio-temporal calendar generation

Patent metadata

Field	Value
Publication number	US-12073369-B2
Application number	US-202117203585-A
Country	US
Kind code	B2
Filing date	Mar 16, 2021
Priority date	Mar 16, 2021
Publication date	Aug 27, 2024
Grant date	Aug 27, 2024

Abstract

A system, computer program product, and method are presented for forecasting a spatio-temporal calendar including predicted regions of interest based on time dependent factors such as long-term weather predictions, time-independent factors, and travel constraints. The method includes collecting information and constraints with respect to service visits. At least a portion of the collected information and constraints are directed toward weather and climate. The method also includes predicting weather and climate impacts on at least one geographical region of interest. The method further includes predicting, subject to the predictions of weather and climate impacts, one or more locations of interest within the at least one geographical region of interest that would be impacted by one or more service visits. The method also includes generating one or more spatio-temporal calendars that include the one or more locations of interest scheduled for the one or more service visits.

First claim

What is claimed is:

1. A computer system comprising:
one or more processing devices and at least one memory device operably coupled to the one or more processing devices, the one or more processing devices are configured to:
collect information and constraints with respect to service visits, wherein at least a portion of the collected information and constraints are directed toward weather and climate comprising:
generate, through a machine learning technology, questions and raise subjects of interest for the user;

predict, through the machine learning technology, weather and climate impacts on at least one geographical region of interest;
predict, automatically, through the machine learning technology, subject to the predictions of weather and climate impacts, one or more locations of interest within the at least one geographical region of interest that would be impacted by one or more service visits;
generate, through the machine learning technology, one or more spatio-temporal calendars that include the one or more locations of interest scheduled for the one or more service visits; and
ingest, by the machine learning technology, feedback from the users, wherein, subject to ingesting the feedback, the one or more processing devices are further configured to:
evaluate the effectiveness of the generated questions and subjects of interest raised;
update the machine learning technology with respect to predictions of the one or more locations of interest: and
improve an effectiveness of the questions generated and subjects of interest raised.

2. The system of claim 1 , wherein the one or more processing devices are further configured to:
identify, automatically, one or more locations of interest that have not been previously identified.

3. The system of claim 1 , wherein the one or more processing devices are further configured to:
generate a spatio-temporal calendar impact metric to measure an effectiveness of the one or more spatio-temporal calendars with respect to the impact of the one or more service visits to the one or more locations of interest.

4. The system of claim 1 , wherein the one or more processing devices are further configured to:
generate the one or more of questions and subjects of interest at least partially based on the predicting one or more locations of interest.

5. The system of claim 1 , wherein the one or more processing devices are further configured to:
group, automatically, the one or more locations of interest into one or more clusters;
select, automatically, at least one representative location from the one or more locations of interest within the respective clusters; and
schedule the at least one representative location for a service visit through the one or more spatio-temporal calendars.

6. The system of claim 1 , wherein the one or more processing devices are further configured to:
estimate, automatically, an importance of the one or more locations of interest comprising one or more of:
estimate the weather and climate impacts, based on weather and climate predictions, on the value of the service visit at any location of interest of the one or more locations of interest in a respective time period specified in the one or more spatio-temporal calendars;
estimate a current state of activity at the one or more locations of interest; and
analyze previous engagements with the one or more locations of interest.

7. The system of claim 1 , wherein the one or more processing devices are further configured to:
use multi-objective optimization to determine a solution to a multi-objective problem statement that includes optimizing two objectives including minimizing travel distance and time and generating an improved spatio-temporal calendar impact on the one or more locations of interest.

8. The system of claim 1 , wherein the one or more processing devices are further configured to:
generate an aggregated spatio-temporal calendar that includes a plurality of the one or more spatio-temporal calendars.

9. The system of claim 1 , wherein the one or more processing devices are further configured to:
use the machine learning technology to generate predictions of the impact to the one or more locations of interest from the weather and climate predictions.

10. A computer program product, the computer program product comprising:
one or more computer readable storage media; and
program instructions collectively stored on the one or more computer-readable storage media, the program instructions comprising:
program instructions to collect information and constraints with respect to service visits, wherein at least a portion of the collected information and constraints are directed toward weather and climate comprising:
program instructions to generate, through a machine learning technology, questions and raise subjects of interest for the user;

program instructions to predict, through the machine learning technology, weather and climate impacts on at least one geographical region of interest;
program instructions to predict, automatically, through the machine learning technology, subject to the predictions of weather and climate impacts, one or more locations of interest within the at least one geographical region of interest that would be impacted by one or more service visits;
program instructions to generate, through the machine learning technology, one or more spatio-temporal calendars that include the one or more locations of interest scheduled for the one or more service visits: and
program instructions to ingest, by the machine learning technology, feedback from the users comprising:
program instructions to evaluate, subject to ingesting the feedback, the effectiveness of the generated questions and subjects of interest raised;
program instructions to update, subject to ingesting of the feedback, the machine learning technology with respect to predictions of the one or more locations of interest: and
program instructions to improve, subject to ingesting of the feedback, an effectiveness of the questions generated and subjects of interest raised.

11. The computer program product of claim 10 , further comprising:
program instructions to identify, automatically, one or more locations of interest that have not been previously identified;
program instructions to generate a spatio-temporal calendar impact metric to measure an effectiveness of the one or more spatio-temporal calendars with respect to the impact of the one or more service visits to the one or more locations of interest;
program instructions to generate the one or more of questions and subjects of interest at least partially based on the prediction of one or more locations of interest; and
program instructions to use the machine learning technology to generate predictions of the impact to the one or more locations of interest from the weather and climate predictions.

12. A computer-implemented method comprising:
collecting information and constraints with respect to service visits, wherein at least a portion of the collected information and constraints are directed toward weather and climate comprising:
generating, through a machine learning technology, questions and raise subjects of interest for the user;

predicting, through the machine learning technology, weather and climate impacts on at least one geographical region of interest;
predicting, automatically, through the machine learning technology, subject to the predictions of weather and climate impacts, one or more locations of interest within the at least one geographical region of interest that would be impacted by one or more service visits;
generating, through the machine

Assignees

• IBM

Inventors

• Kinai Andrew
• Twarakavi Navin
• Otieno Fred Ochieng
• Das Kamal Chandra
• Godbole Shantanu R
• Weldemariam Komminist

Classifications

• G06Q10/1093 Primary

  Calendar-based scheduling for persons or groups · CPC title

• G06N5/04

  Inference or reasoning models · CPC title

• G06N20/00

  Machine learning · CPC title

• G06N5/01

  Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound · CPC title

• G06Q10/1095 Primary

  Physics · mapped topic