Predicting likelihood of on-time product delivery, diagnosing issues that threaten delivery, and exploration of likely outcome of different solutions

• US 9,251,484 B2
• Filed: 05/31/2013
• Issued: 02/02/2016
• Est. Priority Date: 06/01/2012
• Status: Active Grant

• Current Assignee: Green Market Square Limited
• Original Assignee: International Business Machines Corporation
• Inventors: Cantor, Murray R., Duesterwald, Evelyn, Klinger, Tamir, Malkin, Peter K., Matchen, Paul M., Subramanian, Dharmashankar, Sutton, Stanley M., Tarr, Peri L., Wegman, Mark N.
• Primary Examiner(s): Gaffin, Jeffrey A
• Assistant Examiner(s): Misir, Dave
• Application Number:
  US13/906,490
• Publication Number:
  US 20130325763A1
• Time in Patent Office: 977 Days
• Field of Search: 706/12, 705/7.27, 705 7- 9, 717/101
• US Class Current: 1/1
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/063114 Status monitoring or status determination for a person or group
• CPC Class Codes: G06Q 10/06312 Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
• CPC Class Codes: G06Q 10/06313 Resource planning in a project environment
• CPC Class Codes: G06Q 10/0633 Workflow analysis
• CPC Class Codes: G06Q 10/0635 Risk analysis of enterprise or organisation activities

Abstract

A task effort estimator may determine a probability distribution of an estimated effort needed to complete unfinished tasks in a project based on one or more of a set of completed tasks belonging to a project and attributes associated with the completed tasks belonging to the project, a set of completed tasks not belonging to the project and attributes associated with the completed tasks not belonging to the project, or the combination of both. A project completion predictor may determine a probability distribution of completion time for the project based on the probability distribution of an estimated effort needed to complete the unfinished tasks in the project, and one or more resource and scheduling constraints associated with the project.

Claims

1. A method for predicting a probability distribution of completion times for a project, comprising:

receiving a set of unfinished tasks belonging to the project and attributes associated with the unfinished tasks;

obtaining a probability distribution of an estimated effort needed to complete each of the unfinished tasks by a determination based on one or more of a set of completed tasks belonging to the project and attributes associated with the completed tasks belonging to the project, a set of completed tasks not belonging to the project and attributes associated with the completed tasks not belonging to the project, or combination of both;

determining the probability distribution of completion time for the project based on the probability distribution of an estimated effort needed to complete each of the unfinished tasks;

receiving one or more resource and scheduling constraints associated with the project, wherein the determining comprises determining the probability distribution of completion time for the project based on the probability distribution of an estimated effort needed to complete each of the unfinished tasks and on the one or more resource and scheduling constraints associated with the project; and

building a task effort estimation model based on one or more of the set of completed tasks belonging to the project and attributes associated with the completed tasks belonging to the project, or the set of completed tasks not belonging to the project and attributes associated with the completed tasks not belonging to the project, or combination of both,

wherein the determining of the probability distribution of an estimated effort needed to complete each of the unfinished tasks comprises applying the set of unfinished tasks belonging to the project and attributes associated with the unfinished tasks to the task effort estimation model.

2. The method of claim 1, wherein the task effort estimation model is built by a machine learning algorithm, the machine learning algorithm discovering information from the attributes associated the completed tasks belonging to the project if available and the attributes associated with the completed tasks not belonging to the project if available, and applying the discovered information to determine the probability distribution of an estimated effort needed to complete each of the unfinished tasks.

3. The method of claim 2, wherein the discovered information comprises which of the attributes associated with the completed tasks belonging to the project if available, and which of the attributes associated with the completed tasks not belonging to the project if available, are significant in determining the estimated effort, and relationships of the attributes associated the completed tasks belonging to the project if available to a task effort, and relationships of the attributes associated with the completed tasks not belonging to the project if available to the task effort.

4. The method of claim 3, wherein the machine learning algorithm further produces a categorization comprising categories of tasks based on similarity of effort needed to complete the tasks, and wherein the machine learning algorithm further produces a set of attributes for each of the categories of tasks that is predictive of the effort needed to complete a task in said each category.

5. The method of claim 2, wherein an input to the machine learning algorithm comprises one or more attributes associated with a task comprising:

corrected estimate of time needed to complete the task;

creation date of the task;

user who created the task;

due date of the task;

estimate of the task;

links associated with the task;

number of comments recorded associated with the task;

number of teams;

owner name;

planned-for-date;

planned-for iteration;

planned-for start date;

priority;

project name;

resolution;

resolution date;

severity;

state;

state modification date;

story points;

summary;

team name;

time spent;

type;

or

work item identifier;

or

combinations thereof.

6. The method of claim 2, wherein an input to the machine learning algorithm further comprises information derived from the attributes associated with the completed tasks belonging to the project if available and the attributes associated with the completed tasks not belonging to the project if available,

wherein an input to the machine learning algorithm further comprises history of one or more tasks comprising history of changes to values of task attributes, absolute and relative sequence, timing, frequency, properties, and patterns of the history of changes, said one or more tasks being the completed tasks belonging to the project if available and the completed tasks not belonging to the project if available,

wherein the history of changes to values of task attributes comprises one or more of a number of times a task has been rescheduled, a number of times a task has been considered completed and then considered not completed, a number of times an owner of a task has changed, a number of times a task attribute has changed, and a number of times a task relationship has changed, or combinations thereof.

7. The method of claim 1, wherein the determining the probability distribution of completion time for the project comprises repeatedly estimating the completion time for the project based on the probability distribution of the estimated effort needed to complete each of the unfinished tasks, by running a random sampling technique repeatedly.

8. The method of claim 1, further comprising outputting the probability distribution of an estimated effort needed to complete each of the unfinished tasks.

9. The method of claim 1, further comprising producing successive estimates of the probability distribution of an estimated effort needed to complete each of the unfinished tasks by repeating the steps of receiving a set of unfinished tasks belonging to the project and attributes associated with the unfinished tasks, obtaining a probability distribution of an estimated effort needed to complete each of the unfinished tasks, and determining the probability distribution of completion time for the project, as the project progresses toward completion.

10. The method of claim 1, wherein the attributes associated with the unfinished tasks comprise at least initial input estimates of effort needed to complete the unfinished tasks or estimates of a probability distribution of the effort needed to complete the unfinished tasks.

11. A system for predicting a probability distribution of completion times for a project, comprising:

a processor;

a task estimator operable to execute on the processor and further operable to receive a set of unfinished tasks belonging to the project and attributes associated with the unfinished tasks, the task estimator further operable to receive one or more of, a set of completed tasks belonging to the project and attributes associated with the completed tasks belonging to the project, a set of completed tasks not belonging to the project and attributes associated with the completed tasks not belonging to the project, or combinations thereof,

the task estimator further operable to determine a probability distribution of an estimated effort needed to complete each of the unfinished tasks based on one or more of the set of completed tasks belonging to the project and attributes associated with the completed tasks belonging to the project, the set of completed tasks not belonging to the project and attributes associated with the completed tasks not belonging to the project, or the combination of both;

a project completion predictor operable to execute on the processor and further operable to receive one or more resource and scheduling constraints associated with the project, the project completion predictor further operable to determine the probability distribution of completion time for the project based on the probability distribution of an estimated effort needed to complete each of the unfinished tasks and the one or more resource and scheduling constraints associated with the project,

wherein the task estimator comprises a task estimation model built based on one or more of the set of completed tasks belonging to the project and attributes associated with the completed tasks belonging to the project, or the set of completed tasks not belonging to the project and attributes associated with the completed tasks not belonging to the project, or the combination of both,

wherein the task estimator determines the probability distribution of an estimated effort needed to complete each of the unfinished tasks by applying the set of unfinished tasks belonging to the project and attributes associated with the unfinished tasks to the task estimation model.

12. The system of claim 11, wherein the task estimator and the project completion predictor are provided in an integration tool, wherein the integration tool further comprises a problem diagnoses module operable to diagnose one or more project risks and determine one or more remediation actions, a what-if analysis module operable to provide an effect of changing a predefined set of parameters on the probability distribution of complete time for the project, an expert assessment module operable to incorporate addition user supplied information in predicting the probability distribution of complete time for the project, and a graphical user interface operable to provide graphical presentations associated with the probability distribution of an estimated effort needed to complete each of the unfinished tasks and the probability distribution of complete time for the project, the graphical user interface further operable to allow a user to interact with the problem diagnoses module, the what-if analysis module, and the expert assessment module.

13. A non-transitory computer readable storage medium storing a program of instructions executable by a machine to perform a method of predicting a probability distribution of completion times for a project, the method comprising:

receiving a set of unfinished tasks belonging to the project and attributes associated with the unfinished tasks;

obtaining a probability distribution of an estimated effort needed to complete each of the unfinished tasks;

determining the probability distribution of completion time for the project based on the probability distribution of an estimated effort needed to complete each of the unfinished tasks;

receiving one or more resource and scheduling constraints associated with the project, wherein the determining comprises determining the probability distribution of completion time for the project based on the probability distribution of an estimated effort needed to complete each of the unfinished tasks and on the one or more resource and scheduling constraints associated with the project; and

building a task estimation model based on one or more of a set of completed tasks belonging to the project and attributes associated with the completed tasks belonging to the project, or a set of completed tasks not belonging to the project and attributes associated with the completed tasks not belonging to the project, or a user estimates of effort involved in finishing one or more of the unfinished tasks belonging to the project, or combination of thereof; and

applying the set of unfinished tasks belonging to the project to the task estimation model to determine a probability distribution of an estimated effort needed to complete each of the unfinished tasks.

14. The non-transitory computer readable storage medium of claim 13, wherein the obtaining comprises obtaining the probability distribution of an estimated effort needed to complete each of the unfinished tasks by a determination based on one or more of a set of completed tasks belonging to the project and attributes associated with the completed tasks belonging to the project, a set of completed tasks not belonging to the project and attributes associated with the completed tasks not belonging to the project, or combination of both.