Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

A Method and System for Dynamically Assigning Human Tasks in a Business Process

IP.com Disclosure Number: IPCOM000247270D
Publication Date: 2016-Aug-18
Document File: 4 page(s) / 56K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method and system for dynamically assigning human tasks in a business process is disclosed. The dynamic task assignment is done by examining natural language text in a business object and classifying the business object into a functional category and complexity category to assign the task to a participant or a group of participants in the business process based on their availability.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 46% of the total text.

Page 01 of 4

A Method and System for Dynamically Assigning Human Tasks in a Business Process

Disclosed is a method and system for dynamically assigning human tasks in a business process. The dynamic task assignment is done by examining natural language text in a business object and classifying the business object into a functional category and complexity category to assign the task to a participant or a group of participants in the business process also considering their availability.

The method and system on receiving the business object utilizes a machine learning classification algorithm to classify the business object into functional category and complexity category by examining the natural language text in one or more attributes of the business object and to associate process participant groups with each functional and complexity category combination. Subsequently, the mapping defined with the functional category and complexity category to the participant group is used to determine the group or groups of participants most suitable for the task. Further, the method and system on identifying a Light-weight Directory Access Protocol (LDAP) group based on the type of problem and complexity of the problem, the availability of each team member from the mapped LDAP group is calculated. Here, the availability of each team member is calculated by summing up the difference between the estimated completion time and actual percentage completion time of each task for each team member. Thus, the team member with smallest availability value is assigned the task by default from the filtered list of team members.

Figure illustrates a method and system for dynamically assigning human tasks in a business process.

1


Page 02 of 4

Figure

In order to be able to classify a business object into a functional class, a machine learning classification model needs to be developed. This is done by extracting natural language text from the attributes of interest from a known set of business objects that have functional classes associated with them. This is known as the training data set which can be used to train the classification model. This machine learning classification model is built with existing machine learning tools or services using the training data set. Once the classification model is built using the training data set it can be tested and refined using a labelled test set of business objects.

Similarly, the business object can be categorized into the complexity category by extracting natural language text in one or more attributes of the business objects participating in the business process. The categorized business object can be labeled by using a training data set based on the complexity with a finite set of enumerated values such a high, medium and low.

Once the classification models have been developed and tests, the functional and complexity classes identified are mapped to LDAP participant group. Here, the LDAP participant groups include team membe...