Browse Prior Art Database

Use of historical relevancy data to predict likely ad-hoc tasks in a human-driven case management solution

IP.com Disclosure Number: IPCOM000226010D
Publication Date: 2013-Mar-20
Document File: 2 page(s) / 38K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method by which systematic observations and predictions can be applied to advise and influence human operators. Applying those predictions to influence the relevancy ranking of various work objects pulls them gently into operator awareness in a manner similar to an advisor or mentor. The novel approach uses historical data to predict likely directions for in-process work. Predictions of future states based on current cases are used to influence the relevancy assigned to work objects of various types.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 2

Use of historical relevancy data to predict likely ad-hoc tasks in a human-driven case management solution

Rigid models of human-directed business processes lead to inflexible systems and significant investments in exception handling, allowances for alternative work paths, and other costs. Previous work indicates that relevancy weighting might allow a more flexible model, and human operators performing the work can influence which tasks, branches, etc., are the most important in achieving business goals.

The problem addressed here is how analytics might be applied to a strongly human-operator-driven process. An example of such a process is a complex case

management solution, such as might be used in a patent office, where submissions

are dealt with using a combination of predictable business processes and very flexible case management. Each case is potentially unique, and different aspects of internal processes might be needed. Some judgment can be modeled (e.g., a relevancy-weighting scheme). Other aspects of completing a case require human operators to intervene and make judgments based on experience .

To apply analytics to such a human-driven process, historical data about many completed cases could be used to compare similar conditions and to influence relevancy assigned to next or future tasks associated with the case.

Relevancy can be used as a property, and combining the concepts of urgency and priority can help direct the initiation and flow of various workflow and case management objects. This is described in a prescriptive fashion (i.e., system directed) or an ad-hoc fashion (i.e., human directed). It also includes concepts of visualizing work priority, managing parallel branches, supporting more human operator influence over modeled business processes, and other benefits.

The novel approach described herein uses historical data to predict likely directions for in-process work. Predictions of future states based on current cases are used to influence the relevancy assigned to work objects of various types. For example, if a combination of properties in a case was highly correlated with the initiation of a specific task, the relevancy of that task can be influenced to bring it more into the attention of caseworkers. This helps meet the success criteria for a case without completely taking over the human decision-making that is integral to such a system. Essentially, analytics applied to task relevancy in a case becomes something akin to a trusted advisor. Human caseworkers can consider that advice along with experience and judgment, and take steps to move cases forward.

A prescriptive workflow may be the best approach for business processes where human interactions are minimal and work follows a rigid path; many more complex business processes require the interaction of human operators to make risk

judgments and apply professional experience.

The present invention is a method by which systematic observations and predicti...