• All ‘running’ events and data
• The data and history of all students
The engine
• The engine program is continuously creating,
scheduling and evaluating all events for all the
students for all the simulations
• 24 hours a day, 365 days a year
The client
• Each client has his ‘simbrowser’, a rich internet
browser with all the necessary objects and
resources to present the working environment
• The simbrowser communicates continuously
with the engine program and the database on the
server
With this 3rd generation engine it is possible to
build a worldwide learning system. Currently we are
exploring the possibility to build a 4th generation
completely based on web technology (HTML5 and
Javascript).
8 THE INTELLIGENT TUTOR
One of the main plans for this moment is to add a
intelligent virtual tutor to the system, who guides the
learners based on their behaviour and their results.
The Dutch vocational education is becoming
more and more individual: each student has his/her
own learning trajectory. Also, our simulations are
more and more used for training a specific learning
deficit. Therefore the need arises for a system that
guides the student to certain learning goals, using a
learning route that is ideal for that student.
To achieve this goal, an intelligent virtual tutor is
being developed, that can monitor the student within
the virtual learning environment, and present
him/her an ideal learning route.
This tutor can:
• Diagnose where mistakes are being made
• Analyze the learning deficits of the student
• Present new assignments specific for those
deficits
• Give advice and tips for the work involved
• Give the impression that someone is watching
over his/her shoulder and helping
Apart from that, the tutor is capable to adjust the
learning tempo and style to the capabilities of the
student.
Originally, the simulation engine was ment to
simulate ‘real practice’ so students could experience
the real world and could ‘learn by doing’. Now, the
possibility arises to use this virtual environment for
efficient training and learning routes. Every time the
student logs in, the system knows everything the
student has done in the past, and can pick up where
the student left off. Unnecessary steps are avoided.
Figure 4: Schematic overview learning route.
The contents of the tutor system is composed
with a tool that enables presenting learning material
in different ways and styles. It allows specifying
schematically different learning routes and maps.
The problem is mainly filling the contents of this
tutor system. The system can contain many
decisions and rules for following learning routes, but
this content will have to be specified by experts,
requiring analysis of learning processes based on
different learning principles.
Since every followed learning route and
corresponding results are stored in the database, the
engine becomes a powerful platform for scientific
analysis of learning in simulated practical
environments.
9 DISCUSSION
The virtual tutor system enables monitoring every
learning step and individual students following their
learning routes. A more scientific analysis of
achieved results is obvious, the test groups are
already stored in our database.
As the intelligent tutor is just as intelligent as the
rules it uses, it becomes more important to use
didactic analysis on learning material and to use
multiple learning principles within the system.
The authors would like to discuss with
colleagues in this working area about the problems
and challenges this brings forth.
CSEDU2012-4thInternationalConferenceonComputerSupportedEducation
194