System and method for performing early delayed issue detection and providing alert notification

Patent metadata

Field	Value
Publication number	US-2025053414-A1
Application number	US-202318231019-A
Country	US
Kind code	A1
Filing date	Aug 7, 2023
Priority date	Aug 7, 2023
Publication date	Feb 13, 2025
Grant date	—

Abstract

A method and system for determining and mitigating a potential delay in task completion are disclosed. The method includes extracting previously resolved tasks and changes occurred to each of the previously resolved tasks, and calculating features for training a machine learning model. The method further includes dividing the tasks into different groups and training the machine learning model using the calculated features associated with the different groups. The method then starts a task and partially processes the respective task until reaching a cutoff time, and based on change history of the task up to the cutoff time and using the trained machine learning model, determines whether a delay is expected or not for the task.

First claim

What is claimed is:

1 . A method for determining and mitigating a potential delay in task completion, the method comprising:
extracting, by a processor, a plurality of previously resolved tasks and related information; analyzing, by the processor, the extracted plurality of previously resolved tasks and related information; extracting, by the processor, changes occurred to each of the plurality of previously resolved tasks; calculating, by the processor, features for training a machine learning model based on the extracted changes; dividing, by the processor, the plurality of previously resolved tasks into different groups; training, the machine learning model, using the calculated features associated with the different groups of the plurality of previously resolved tasks; starting, by the processor, a sprint including a target task; determining, by the processor, a cutoff time for determining whether the target task will be completed by an end date of the sprint; processing, by the processor, the target task until reaching the cutoff time; upon reaching the cutoff time, processing change history of the target task up to the cutoff time and before the sprint is completed, and extracting a sequence of changes included in the change history; and executing the machine learning model using the extracted sequence of changes for determining whether a delay is expected or not for the target task.

2 . The method according to claim 1 , further comprising:
when the delay is not expected, continue the processing of the target task without modification until completion of the target task.

3 . The method according to claim 1 , further comprising:
when the delay is expected, modifying the processing of the target task and continue the modified processing of the target task until completion of the target task.

4 . The method according to claim 3 , wherein the modifying includes a modification to allocated resources for the processing of the target task.

5 . The method according to claim 3 , wherein the modifying includes a modification to a sequence of the processing of the target task.

6 . The method according to claim 1 , wherein the different groups include a first group of tasks that were completed within their respective sprints, and a second group of tasks that were unable to be completed within their respective sprints.

7 . The method according to claim 3 , further comprising:
when the target task is completed, updating the machine learning model with processing information of the target task.

8 . The method according to claim 7 , further comprising:
executing the updated machine learning model on a subsequent task for determining whether the subsequent task will be completed prior to completion of its respective sprint.

9 . The method according to claim 1 , further comprising:
outputting a delay alert when the delay is expected for the target task.

10 . The method according to claim 9 , wherein the delay alert is a sound signal outputted by a speaker.

11 . The method according to claim 9 , wherein the delay alert is a flashing light.

12 . The method according to claim 9 , wherein the delay alert is an automated voice call.

13 . The method according to claim 9 , wherein the delay alert is a text transmission to a receiving device.

14 . The method according to claim 1 , wherein the machine learning model is a neural network model.

15 . The method according to claim 1 , wherein the cutoff time is at most six days after starting the sprint.

16 . The method according to claim 1 , wherein the cutoff time is determined by the machine learning model.

17 . The method according to claim 1 , wherein the changes occurred to each of the plurality of previously resolved tasks are extracted in sequence.

18 . The method according to claim 1 , wherein the plurality of previously resolved tasks is divided based on a task type, completion time period and resource allocation.

19 . A system for determining and mitigating a potential delay in task completion, the system comprising:
at least one memory; and at least one processor, wherein the system is configured to perform: extracting a plurality of previously resolved tasks and related information; analyzing the extracted plurality of previously resolved tasks and related information; extracting changes occurred to each of the plurality of previously resolved tasks; calculating features for training a machine learning model based on the extracted changes; dividing the plurality of previously resolved tasks into different groups, the different groups include a first group and a second group; training, the machine learning model, using the calculated features associated with the different groups of the plurality of previously resolved tasks; starting a sprint including a target task; determining a cutoff time for determining whether the target task will be completed by an end date of the sprint; processing the target task until reaching the cutoff time; upon reaching the cutoff time, processing change history of the target task up to the cutoff time and before the sprint is completed, and extracting a sequence of changes included in the change history; and executing the machine learning model using the extracted sequence of changes for determining whether a delay is expected or not for the target task.

20 . A non-transitory computer readable storage medium that stores a computer program for determining and mitigating a potential delay in task completion, the computer program, when executed by a processor, causing a system to perform a plurality of processes comprising:
extracting a plurality of previously resolved tasks and related information; analyzing the extracted plurality of previously resolved tasks and related information; extracting changes occurred to each of the plurality of previously resolved tasks; calculating features for training a machine learning model based on the extracted changes; dividing the plurality of previously resolved tasks into different groups, the different groups include a first group and a second group; training, the machine learning model, using the calculated features associated with the different groups of the plurality of previously resolved tasks; starting a sprint including a target task; determining a cutoff time for determining whether the target task will be completed by an end date of the sprint; processing the target task until reaching the cutoff time; upon reaching the cutoff time, processing change history of the target task up to the cutoff time and before the sprint is completed, and extracting a sequence of changes included in the change history; and executing the machine learning model using the extracted sequence of changes for determining whether a delay is expected or not for the target task.

Assignees

• Jpmorgan Chase Bank Na

Inventors

• Tawosi Vali
• Alamir Salwa Husam
• Tummalapenta Ravi Kumar
• Shah Sameena

Classifications

• G06F8/70 Primary

  Software maintenance or management · CPC title