 |
 |
 |
 |
| |
|||||
|
|
|
Home > News & Events > MITRE Publications > The MITRE Digest > | |||||||||||||||||||
| Helping Organizations Manage Transformations May 2008
Mathematician James A. Yorke once commented, "The most successful people are those who are good at Plan B." Anyone who's weathered a workplace reorganization will recognize the truth of this sentiment. Whether your company recently merged with a former competitor or the IT department replaced familiar systems with radically new ones, most of us have experienced the discomfort that accompanies change. Research now underway at MITRE aims to help individuals and organizations deal successfully with dramatic organizational change. Through this research, MITRE is developing tools to help civil and military organizations predict how populations such as government employees, healthcare providers, and migrant populations will respond to events like business restructurings, procedure changes, and government policy overhauls. An analytical model developed by MITRE will help our sponsors take advantage of these change management insights as they seek more effective ways to communicate critical information about new government initiatives, procedures, and emergency response plans. "So many of our sponsors confront environments in which people must react to change, and we're trying to help them anticipate how people will react," says Charles Worrell, a MITRE principal information systems engineer who leads the behavioral modeling research project. It has long been possible to develop simulations of systems to predict how well they will perform, he says, but efforts to develop models for predicting human behavior have been less successful. MITRE's research now shows that this doesn't have to be the case. By gathering information about the reactions of disparate groups of people, breaking down those reactions into what Worrell calls "layers of perception," and then analyzing these data with proven psychological and statistical models, it's possible to determine the best medium and timing for a given message. Armed with this knowledge, the messenger (for example, an employer announcing significant changes in workers' job responsibilities) has a better chance of producing the desired reaction (such as compliance with new workplace policies), he says. Building More Effective Communications Plans "Often when you try to make a simulation of the operations of a business enterprise, the results miss the mark because of human factors," Worrell says. "People exhibit variable behaviors that can be very hard for traditional simulation models to represent—and people are a vital part of any enterprise." Quantifying these "human factors" becomes critical in analyzing a significant business overhaul, such as the post-Sept. 11 formation of the Department of Homeland Security (DHS), which was created out of nearly two dozen previously-independent government agencies. It is possible to gain a quantitative understanding of these dynamic situations by leveraging tools from multiple disciplines. Such insights are critical to any organization attempting to effect significant, enterprise-wide change, says Craig Petrun, a senior principal information systems engineer and head of MITRE's Organizational Strategy and Change department, which helps sponsors deal with such transformations. "If you look at where we're going with complex systems engineering, you see the need to manage change across people, organizations, and cultures," he says. "Just as you have to manage the architecture, you have to manage the people element of organizations." MITRE's model, known as "DAP-E," for its four-part approach—"Decompose,
Aggregate, Propagate, Evaluate"—allows users to assess the effectiveness
of their communications plans both prospectively and as the plans are
being carried out, says Worrell. The model, for which MITRE has submitted
a patent application, represents the people in an organization by using
existing psychological models to predict how they would answer a series
of questions about the factors that drive their reactions, he explains.
For example, under the business overhaul scenario, in the "Decompose"
phase, workers' reactions would be based on their answers to the following
four questions:
In the case of a public health organization attempting to gauge the effectiveness
of campaigns for smoking cessation, or to increase participation in cancer
screenings like colonoscopies or mammograms, the four "Decompose" questions
for health care consumers would be:
The likely answers different groups would give are estimated using a number of established psychological models in the Decompose phase (see sidebar, "Leveraging Old Models to Gain New Insights.") During the "Aggregate" phase, the model estimates the statistical likelihood of, for example, an employee's compliance with proposed rule changes on the basis of the combined answers from the Decompose phase. "This method allows us to incorporate the uncertainty people have as they go through a thought process," Worrell explains. This "uncertainty" is represented through something called a Bayesian inference network, which is a statistical model that represents a set of variables and their independencies. Once the likely conclusions of various groups of employees are quantified, the model simulates how these opinions are likely to travel from one group to another. The frequency of contact between and among these groups, and the nature of the relationships (manager to manager, supervisor to lower-level worker, and so on) determines the amount of influence the communicated opinions will have. This is the "Propagate" phase of the model. During the "Evaluate" phase, researchers parse this data to assess the impact of the various groups' opinions on the organization or program's performance. Applying Research Findings to Sponsor Problems Worrell's team is using the insights gained from this work to enhance a decision support model for DHS's Secure Border Initiative. "This will help decision makers think through the ramifications of different immigration policies," he says. "It gives DHS the ability to represent the proportions of people in different demographic groups who are likely to alter their migration choices as a result of changes in U.S. policy." This same model will also be tailored for the public health community, for use in judging the effectiveness of hand-washing promotional campaigns within hospitals and other health centers, Worrell says. Plans are underway for a related internal MITRE research project to customize the model for the Environmental Protection Agency, so that the EPA can examine the effectiveness of messages regarding enhanced waste management procedures, he says. Petrun sees almost limitless opportunities for applying the model to real-world sponsor problems. "Take the area of homeland security—there is a critical need to share information more quickly across the Department of Homeland Security, the Department of Defense, and the Director of National Intelligence," he says. "We've had the technology and tools to share data and assess the reliability of data, but with a new model like this one, we can also simulate the human aspects of what would facilitate increased communication and information sharing across boundaries."
—by Maria S. Lee Related Information Articles and News
Technical Papers and Presentations
Websites |
||||||||||||||||||||
| Page last updated: May 22, 2008 | Top of page |
Solutions That Make a Difference.® |
|
|
