While traditional computer based training applications can improve provide substantial benefits over classroom training, most are no more than a sophisticated flash-card systems. Even so, learning is estimated to be 38%-70% faster than through conventional instruction, while understanding is improved by about 50%, and retention by 25%-50%.

Intelligent tutors can outperform traditional applications by tailoring lessons to each student, and by tailoring the instruction to the way each student best learns. The system compares the student’s performance to expert performance, and actually changes its instructional content based on that evaluation. It does not teach what the student has already mastered; instead, it recognizes and employs the instructional methods most effective for each student.

Listed below is an example of an advanced training program being developed at the HITC:

Apricots: Advanced Produce Interactive Cashier Off-line Training System

APRICOTS is an adaptive, dynamic training system for recognizing non-barcodable items.   It trains cashiers to better recognize and remember the names and PLU numbers of produce items, which can not be easily scanned. Not only can it be deployed as a standalone, back-room training application, but also at the POS lane, to provide training at the front end.

Developed jointly with Balls Food to reduce shrink in their produce areas.  View a screen shot of the system.

Two articles describing "Apricots" were published in Supermarket Business in the 1997 September and October issues.