Authors: Maureen Fleming, Dan Vesset

Abstract

This IDC Perspective discusses how as machine learning adoption rapidly progresses, enterprises are beginning to realize that predicting problems and opportunities generates high volumes of decisions that workers must make to gain any kind of positive benefit from implementing the analytics solution. However, the decision volumes are so high that teams can only make the most urgent decisions.

This document introduces decision-centric computing, an emerging style of building applications that solves this problem by putting decision automation at the heart of the solution. Decision-centric solutions continuously receive and analyze data to predict when decisions need to be made, systematically learn how to automate those decisions, and act on each decision to improve performance.

As the results of manual and automated decisions and outcomes are fed back into the system, the accuracy of the predictions, algorithms, and decision models improves. As a result, a decision-oriented solution:

• Makes an ever-increasing percentage of decisions on its own with no manual assistance
• Passes increasingly smaller percentages of decision making to people
• Ignores an increasingly smaller percentage of conditions

"Without a way to incorporate decision automation to make repetitive decisions, enterprises will find it increasingly difficult to justify their investments in advanced analytics and risk failure to materialize the anticipated benefits," according to Maureen Fleming, program vice president for Integration and Process Automation Research.

Coverage