NSF PR 03-42 - April 18, 2003
NSF Researchers To Help Modern Organizations Adapt and Respond In The Information Age
ARLINGTON, Va. - The most critical networks in any organization are not necessarily the ones carrying Internet traffic, but the social networks among persons and groups that define an organization's process and knowledge flow. Kathleen Carley's goal is to estimate the size, shape and weaknesses of those social networks to help managers predict how an organization is likely to respond to anticipated and unanticipated changes. Ray Levitt wants to design, from the ground up, organizations without any weaknesses at all.
Carley, the director of Computational Analysis of Social and Organizational Systems at Carnegie Mellon University, and Levitt, academic director of Stanford University's Advanced Project Management Program, are leading two of the eight National Science Foundation (NSF) projects whose efforts are being supplemented by $4 million over two years as part of the Management of Knowledge Intensive Dynamic Systems (MKIDS) program. The program reflects an aspect of NSF's charter: to support science and engineering research related to national security.
The eight projects are researching how information technologies can help streamline processes for organizations that must respond rapidly to incoming knowledge, dynamic situations and uncertainty. Such organizations include news media, multinational research corporations, global finance institutions and the intelligence community.
The systems envisioned by the MKIDS program go beyond even today's leading-edge "data mining" systems, which attempt to monitor vast streams of data and pinpoint events of interest. Using input such as these events, an MKIDS system would employ scheduling tools to help decision-makers adjust organizational processes by dynamically allocating physical resources, technology services, and human resources. In addition, an MKIDS system would have controller functions that would monitor the organization's response to the scheduling decisions and provide ways to fine-tune the process and "remember" the best practices from ongoing experience.
Carley's team at Carnegie Mellon feeds an organization's external data sources-such as e-mail exchanges, phone calls and human resources databases-into the best available computational models from organizational science. These models then extrapolate an organization's internal structure and alert managers to likely organizational "failure points." By contrast, Levitt and Stanford colleague Stephen Barley are developing tools that will help managers design and "prototype" optimal organizational structures for complex, multinational projects.
"We want to develop computational tools to help managers design organizations the way engineers design bridges," Levitt said. "There is so little predictive ability for organizations in this area. It's all based on managers' experience and intuition." With the supplemental funding, Levitt's team is expanding successful tools for designing small-scale project teams to factor in the "organizational chemistry"-effects arising from human interactions and relationships-of global and multicultural activities such as World Bank-funded projects or production of timely newscasts by an international news organization.
The MKIDS projects are tackling various aspects of a wide-ranging and elusive goal. The success of many organizations is becoming more heavily, if not entirely, dependent on responding to changing situations that are reflected in accumulated knowledge. To ensure success, many organizations need to help their management teams apply global collections of knowledge in a complex and time-critical decision-making process.
"No one researcher or research team can solve this problem," said NSF program officer Suzanne Iacono. "These systems monitor and respond to events taking place throughout an organization's managerial levels and geographical locations. To achieve concerted performance, independent software 'agents' have to figure out how to work together and with human decision makers in the most effective way possible."
Many modern enterprises in today's information economy face the sorts of challenges being addressed by the MKIDS projects. A hypothetical example would be a global newsgathering organization, with multi-lingual television, radio, Internet, and news service reporting channels.
A shift in the world's political situation-such as a crisis suddenly propelling sub-Saharan Africa into the global spotlight-would involve rapid decisions on which reporters to move where to ensure the best coverage, editorial decisions on what news to cover when, and programming decisions on what programs to create, cancel, interrupt or reschedule.
Other research challenges within the scope of MKIDS include developing ways to represent, store and share knowledge derived by the systems, ensuring the systems can scale to global organizations, learning ill-defined workflow rules, allowing managers throughout an organization to participate in distributed decision-making, and developing innovative metrics and monitoring methods to evaluate performance.
NSF is an independent federal agency that supports fundamental research and education across all fields of science and engineering, with an annual budget of nearly $5 billion. NSF funds reach all 50 states through grants to nearly 2,000 universities and institutions. Each year, NSF receives about 30,000 competitive requests for funding, and makes about 10,000 new funding awards. NSF also awards over $200 million in professional and service contracts yearly.
Receive official NSF news electronically through the e-mail delivery system, NSFnews. To subscribe, send an e-mail message to firstname.lastname@example.org. In the body of the message, type "subscribe nsfnews" and then type your name. (Ex.: "subscribe nsfnews John Smith").
Useful Web Sites:
NSF Home Page: http://www.nsf.gov
News Highlights: http://www.nsf.gov/home/news.html
Science Statistics: http://www.nsf.gov/sbe/srs/stats.htm
Awards Searches: http://www.fastlane.nsf.gov/a6/A6Start.htm