*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Complex domains, or complex-dynamic systems, permeate daily life. Moreover, they are often safety-critical. In these systems, practitioners are both professional and expert in their roles. Historically, change in complex systems was slow. Both system and operational knowledge were embedded in the organization. On-the-job training, with one generation of operations personnel training the next, was common. Today, and for the foreseeable future, rapid change is a nearly ubiquitous characteristic of complex-dynamic systems. Widespread use of digital technology greatly increases the rate at which systems change and the complexity of systems. Changes in the work environment, however, can degrade even the most skilled practitioners expertise. Whether change is due to the addition of a new system feature, introduction of a modified procedure, replacement of a familiar system interface, or rotation of personnel to a similar system, change can create gaps or misunderstandings in practitioner knowledge. Moreover, such gaps or misunderstanding can significantly affect performance. Under such circumstances, these practitioners, although highly skilled, can sometimes be thought of as trained novices.
To maintain expertise in the face of rapid change, ongoing training of expert personnel is a necessity. It is difficult, however, with current training methods, e.g., classroom training, to ensure that practitioners keep abreast of new knowledge and procedures. Training expense and logistics are major obstacles.
The goals of this research are to address the growing training demands of maintaining practitioner expertise by using computer-based training that merges ITS and case-based teaching. A further goal is to implement this new approach in such a way that facilitates the ease and decreases the cost of incorporating new cases as training needs evolve. To address these goals, this research proposes a theory and architecture for case-based computer-implemented training, Case-Based Intelligent Tutoring System (CBITS). CBITS builds upon the experience and research in both ITS and case-based teaching. The ITS provides a control structure for monitoring the individual student and addressing their individual needs. Within that structure, cases provide a method of teaching, using memorable experiences to create focused instruction. Cases also allow tutor content to evolve as the operational environment evolves.
Implementations of CBITS are relevant in a range of domains in which practitioners interact with a technological and evolving system. Examples include airline pilots and maintenance, electronic manufacturing, and telecommunications. Currently, CBITS is implemented in proof-of-concept for MD-11 pilots, teaching a newly licensed capability of the aircraft. This capability introduces a new technique that improves safety but is unfamiliar to experienced pilots. An evaluation of the system with active airline pilots showed the system and training to be highly effective.
Committee
----------------------------------------------------------------------
Prof. Christine M. Mitchell (Advisor)
Prof. T. Govindaraj
Prof. Dave Goldsman
Prof. William B. Rouse
Dr. Everett A. Palmer, III (NASA Ames Research Center)
