Release cycle optimization based on significant features values simulation

Patent metadata

Field	Value
Publication number	US-10657298-B2
Application number	US-201615183011-A
Country	US
Kind code	B2
Filing date	Jun 15, 2016
Priority date	Jun 15, 2016
Publication date	May 19, 2020
Grant date	May 19, 2020

Abstract

Embodiments include a system for release cycle optimization; the system includes a processor configured to perform a method. The method includes accessing, by a processor, historical data relating to a plurality of software version each having a plurality of attributes; selecting a subset of attributes from the plurality of attributes; receiving a set of data values for each of the subset of attributes from the plurality of attributes; performing one or more simulations of a software development cycle utilizing the set of data values; and obtaining a set of results from the one or more simulations comprising a plurality of predicted field defects values corresponding to each of the set of data values.

First claim

What is claimed is:

1. A computer-implemented method for release cycle optimization, the method comprising:
accessing, by a processor, historical data relating to a plurality of software versions each having a plurality of attributes;
determining a subset of attributes from the plurality of attributes based on a feature vector, generated by a machine learning model, comprising a plurality of features extracted from the historical data, wherein the subset of attributes are determined based at least in part on a desired software release performance associated with a software release;
receiving a set of data values for each attribute in the subset of attributes from the plurality of attributes;
performing one or more simulations of a software development cycle utilizing the set of data values; and
obtaining a set of optimized results from the one or more simulations comprising a plurality of predicted field defects values corresponding to each of the set of data values, wherein the set of results comprise a project cost associated with each result in the set of optimized results;
determining one or more constraints associated with a business organization for implementation of the software release;
discarding one or more optimized results in the set of optimized results that conflict with at least one constraint of the one or more constraints to determine a second set of optimized results;
identifying, by the processor, a resource need for each of the plurality of attributes based on the second set of optimized results, wherein the resource need comprises an individual outside the business organization having a skill set associated with the set of optimized results; and
generating, for a user, a workflow to obtain the resource.

2. The method according to claim 1 , further comprising:
receiving a threshold field defect value; and
identifying, by a user, a plurality of results from the set of optimized results with a predicted field defect value below the threshold field defect value.

3. The method according to claim 2 , further comprising:
accessing an availability of resources; and
selecting one or more of the plurality of optimized results based upon the predicted field defects values, the resource need, and the availability of resources.

4. The method according to claim 3 , wherein the availability of resources is determined by at least one of a software engineer, a hardware engineer, and a human resources individual.

5. The method according to claim 1 , wherein the machine learning model comprises a support vector machines algorithm.

6. The method according to claim 1 , wherein the plurality of attributes comprise at least one of a size of a development team, a development time period, a size of a test team, a number of executed test cases, and an iteration length.

7. A system for release cycle optimization, the system comprising:
a processor configured to:
access historical data, wherein the historical data relates to a plurality of software versions each having a plurality of attributes;
determine a subset of attributes from the plurality of attributes based on a feature vector, generated by a machine learning model, comprising a plurality of features extracted from the historical data, wherein the subset of attributes are determined based at least in part on a desired software release performance associated with a software release;
receive a set of data values for each of the subset of attributes from the plurality of attributes; and
perform one or more simulations of a software development cycle utilizing the set of data values; and
obtain a set of optimized results from the one or more simulations comprising a plurality of predicted field defects values corresponding to each of the set of data values, wherein the set of results comprise a project cost associated with each result in the set of results;
determine one or more constraints associated with a business organization for implementation of the software release;
discard one or more optimized results in the set of optimized results that conflict with at least one constraint of the one or more constraints to determine a second set of optimized results;
identify a resource need for each of the plurality of attributes based on the second set of results, wherein the resource need comprises an individual outside the business organization having a skill set associated with the set of optimized results; and
generate, for a user, a workflow to obtain the resource.

8. The system according to claim 7 , further comprising:
the processor configured to:
receive a threshold field defect value; and
identify, by a user of the processor, a plurality of results from the set of optimized results with a predicted field defect value below the threshold field defect value.

9. The system according to claim 8 , further comprising:
the processor configured to:
access an availability of resources; and
select one or more of the plurality of results based upon the predicted field defects values, the resource need, and the availability of resources.

10. The system according to claim 9 , wherein the availability of resources is determined by at least one of a software engineer, a hardware engineer, and a human resources individual.

11. The system according to claim 7 , wherein the machine learning model comprises a support vector machines algorithm.

12. The system according to claim 7 , wherein the plurality of attributes comprise at least one of a size of a development team, a development time period, a size of a test team, a number of executed test cases, and an iteration length.

13. A computer program product for release cycle optimization, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, wherein the computer readable storage medium is not a transitory signal per se, the program instructions executable by a processor to cause the processor to perform a method comprising:
accessing, by a processor, historical data, wherein the historical data relates to a plurality of software versions each having a plurality of attributes;
determining a subset of attributes from the plurality of attributes based on a feature vector, generated by a machine learning model, comprising a plurality of features extracted from the historical data, wherein the subset of attributes are determined based at least in part on a desired software release performance associated with a software release;
receiving a set of data values for each of the subset of attributes from the plurality of attributes; and
performing one or more simulations of a software development cycle utilizing the set of data values; and
obtaining a set of optimized results from the one or more simulations comprising a plurality of predicted field defects values corresponding to each of the set of data values, wherein the set of results comprise a project cost associated with each result in the set of results;
determining one or more constraints associated with a business organization for implementation of the software release;
discarding one or more optimized results in the set of optimized results that conflict with at least one constraint of the one or more constraints to determine a second set of optimized results;
identifying, by the processor, a resource need for each of the plurality of attributes based on the second set of results, wherein the resource need comprises an individual outside the business organization having a skill set associated with the set of optimized results; and
generating, for a user, a workflow to obtain the resource.

Assignees

• IBM

Inventors

• Cmielowski Lukasz G
• Franczyk Marek
• Gedliczka Tymoteusz
• Wrobel Andrzej J

Classifications

• G06F30/20 Primary

  Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F30/30) · CPC title

• G06F8/71

  Version control (security arrangements therefor G06F21/57); Configuration management · CPC title

• G06Q10/06

  Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling · CPC title

• G06N20/00

  Machine learning · CPC title

• G06N20/10 Primary

  using kernel methods, e.g. support vector machines [SVM] · CPC title