System, method and article of manufacture for a goal based system utilizing a time based model

Abstract

A system is disclosed that provides a goal based learning system utilizing a rule based expert training system to provide a cognitive educational experience. The system provides the user with a simulated environment that presents a business opportunity to understand and solve optimally. Mistakes are noted and remedial educational material presented dynamically to build the necessary skills that a user requires for success in the business endeavor. The system utilizes an artificial intelligence engine driving individualized and dynamic feedback with synchronized video and graphics used to simulate real-world environment and interactions. Multiple "correct" answers are integrated into the learning system to allow individualized learning experiences in which navigation through the system is at a pace controlled by the learner. A robust business model provides support for realistic activities and allows a user to experience real world consequences for their actions and decisions and entails realtime decision-making and synthesis of the educational material. The system is architected around a time based model to manage and control the system.

Claims

What is claimed is:

1. A method for creating a presentation, comprising the steps of:

(a) receiving information indicative of a goal;

(b) integrating information that motivates accomplishment of the goal for use in the presentation;

(c) synchronizing events in the presentation utilizing a time based model; and

(d) evaluating progress toward the goal and providing feedback that further motivates accomplishment of the goal utilizing the time based model to control the presentation of information.

2. A method for creating a presentation as recited in claim 1, including the step of presenting an interactive session and querying the user for analysis of the interactive session including a decision.

3. A method for creating a presentation as recited in claim 2, including the step of altering the presentation based on the user's decision to further refine the accomplishment of the goal.

4. A method for creating a presentation as recited in claim 1, including the step of advancing time as the presentation proceeds.

5. A method for creating a presentation as recited in claim 4, including the step of presenting a new presentation and querying the user for a new decision after the time is advanced.

6. A method for creating a presentation as recited in claim 1, including the step of simulating the management of resources utilizing the presentation.

7. A method for creating a presentation as recited in claim 1, including the step of adjusting the feedback based on a current time.

8. A method for creating a presentation as recited in claim 1, including the step of passing information from the presentation to an expert system to analyze the information and formulate the appropriate feedback utilizing time as one of the variables for analysis.

9. A method for creating a presentation as recited in claim 1, including the step of utilizing an internal clock to synchronize the time.

10. An apparatus that creates a presentation, comprising;

(a) a processor;

(b) a memory that stores information under the control of the processor;

(c) logic that integrates information that motivates accomplishment of the goal for use in the presentation;

(d) logic that synchronizes events in the presentation utilizing a time based model; and

(e) logic that evaluates progress toward the goal and provides feedback that further motivates accomplishment of the goal utilizing the time based model to control the presentation of information.

11. A computer program embodied on a computer-readable medium that creates a presentation, comprising:

(a) a code segment that integrates information that motivates accomplishment of the goal for use in the presentation;

(b) a code segment that synchronizes events in the presentation utilizing a time based model; and

(c) a code segment that evaluates progress toward the goal and provides feedback that further motivates accomplishment of the goal utilizing the time based model to control the presentation of information.

12. A computer program embodied on a computer-readable medium as recited in claim 11, including a code segment that presents an interactive session and querying the user for analysis of the interactive session including a decision.

13. A computer program embodied on a computer-readable medium as recited in claim 11, including a code segment that alters the presentation based on the user's decision to further refine the accomplishment of the goal.

14. A computer program embodied on a computer-readable medium as recited in claim 11, including a code segment that advances time as the presentation proceeds.

15. A computer program embodied on a computer-readable medium as recited in claim 11, including a code segment that presents a new presentation and querying the user for a new decision after the time is advanced.

16. A computer program embodied on a computer-readable medium as recited in claim 11, including a code segment that simulates the management of resources utilizing the presentation.

17. A computer program embodied on a computer-readable medium as recited in claim 11, including a code segment that adjusts the feedback based on a current time.

18. A computer program embodied on a computer-readable medium as recited in claim 11, including a code segment that passes information from the presentation to an expert system to analyze the information and formulate the appropriate feedback utilizing time as one of the variables for analysis.

19. A computer program embodied on a computer-readable medium as recited in claim 11, including a code segment that utilizes an internal clock to synchronize the time.

Description

FIELD OF THE INVENTION

The present invention relates to education systems and more particularly to a rule based tutorial system that utilizes a time based model to control business simulations of actual environments to teach new skills.

BACKGROUND OF THE INVENTION

When building a knowledge based system or expert system, at least two disciplines are necessary to properly construct the rules that drive the knowledge base, the discipline of the knowledge engineer and the knowledge of the expert. The domain expert has knowledge of the domain or field of use of the expert system. For example, the domain expert of an expert for instructing students in an automotive manufacturing facility might be a process control engineer while the domain expert for a medical instruction system might be a doctor or a nurse. The knowledge engineer is a person that understands the expert system and utilizes the expert's knowledge to create an application for the system. In many instances, the knowledge engineer and domain expert are separate people who have to collaborate to construct the expert system.

Typically, this collaboration takes the form of the knowledge engineer asking questions of the domain expert and incorporating the answers to these questions into the design of the system. This approach is labor intensive, slow and error prone. The coordination of the two separate disciplines may lead to problems. Although the knowledge engineer can transcribe input from the expert utilizing videotape, audio tape, text and other sources, efforts from people of both disciplines have to be expended. Further, if the knowledge engineer does not ask the right questions or asks the questions in an incorrect way, the information utilized to design the knowledge base could be incorrect. Feedback to the knowledge engineer from the expert system is often not available in prior art system until the construction is completed. With conventional system, there is a time consuming feedback loop that ties together various processes from knowledge acquisition to validation.

Educational systems utilizing an expert system component often suffer from a lack of motivational aspects that result in a user becoming bored or ceasing to complete a training program. Current training programs utilize static, hard-coded feedback with some linear video and graphics used to add visual appeal and illustrate concepts. These systems typically support one "correct" answer and navigation through the system is only supported through a single defined path which results in a two-dimensional generic interaction, with no business model support and a single feedback to the learner of correct or incorrect based on the selected response. Current tutorial systems do not architect real business simulations into the rules to provide a creative learning environment to a user.

SUMMARY OF THE INVENTION

According to a broad aspect of a preferred embodiment of the invention, a goal based learning system utilizes a rule based expert training system to provide a cognitive educational experience. The system provides the user with a simulated environment that presents a business opportunity to understand and solve optimally. Mistakes are noted and remedial educational material presented dynamically to build the necessary skills that a user requires for success in the business endeavor. The system utilizes an artificial intelligence engine driving individualized and dynamic feedback with synchronized video and graphics used to simulate real-world environment and interactions. Multiple "correct" answers are integrated into the learning system to allow individualized learning experiences in which navigation through the system is at a pace controlled by the learner. A robust business model provides support for realistic activities and allows a user to experience real world consequences for their actions and decisions and entails realtime decision-making and synthesis of the educational material. The system is architected around a time based model to manage and control the system.

DESCRIPTION OF THE DRAWINGS

The foregoing and other objects, aspects and advantages are better understood from the following detailed description of a preferred embodiment of the invention with reference to the drawings, in which:

FIG. 1 is a block diagram of a representative hardware environment in accordance with a preferred embodiment;

FIG. 2 is a block diagram of a system architecture in accordance with a preferred embodiment;

FIG. 3 depicts the timeline and relative resource requirements for each phase of development for a typical application development in accordance with a preferred embodiment;

FIG. 4 depicts the potential savings in both functional and technical tasks in accordance with a preferred embodiment;

FIG. 5 illustrates commonalties in accordance with a preferred embodiment;

FIG. 6 illustrates a development architecture approach in accordance with a preferred embodiment;

FIG. 7 illustrates a small segment of a domain model for claims handlers in the auto insurance industry in accordance with a preferred embodiment;

FIG. 8 illustrates an instantiated domain model in accordance with a preferred embodiment;

FIG. 9 illustrates an insurance underwriting profile in accordance with a preferred embodiment;

FIG. 10 illustrates a transformation component in accordance with a preferred embodiment;

FIG. 11 illustrates the use of a toolbar to navigate and access application level features in accordance with a preferred embodiment;

FIG. 12 is a GBS display in accordance with a preferred embodiment;

FIG. 13 is a feedback display in accordance with a preferred embodiment;

FIG. 14 illustrates a display in which a student has made some mistakes in accordance with a preferred embodiment;

FIG. 15 illustrates a journal entry simulation in accordance with a preferred embodiment;

FIG. 16 illustrates a simulated Bell Phone Bill journal entry in accordance with a preferred embodiment;

FIG. 17 illustrates a feedback display in accordance with a preferred embodiment;

FIGS. 18 and 19 illustrate a feedback display in accordance with a preferred embodiment;

FIG. 20 illustrates a feedback display in accordance with a preferred embodiment;

FIG. 21 illustrates a simulation display in accordance with a preferred embodiment;

FIG. 22 illustrates the steps of the first scenario in accordance with a preferred embodiment;

FIG. 23 and 24 illustrate the steps associated with a build scenario in accordance with a preferred embodiment;

FIG. 25 illustrates how the tool suite supports student administration in accordance with a preferred embodiment;

FIG. 26 illustrates a suite to support a student interaction in accordance with a preferred embodiment;

FIG. 27 illustrates the remediation process in accordance with a preferred embodiment;

FIG. 28 illustrates a display of journalization transactions in accordance with a preferred embodiment;

FIG. 29 illustrates the objects for the journalization task in accordance with a preferred embodiment;

FIG. 30 illustrates the mapping of a source item to a target item in accordance with a preferred embodiment;

FIG. 31 illustrates target group bundles in accordance with a preferred embodiment;

FIG. 32 illustrates a TargetGroup Hierarchy in accordance with a preferred embodiment;

FIG. 33 illustrates a small section the amount of feedback in accordance with a preferred embodiment;

FIG. 34 illustrates an analysis of rules in accordance with a preferred embodiment;

FIG. 35 illustrates a feedback selection in accordance with a preferred embodiment;

FIG. 36 is a flowchart of the feedback logic in accordance with a preferred embodiment;

FIG. 37 illustrates an example of separating out some mistakes for the interface to catch and others for the ICAT to catch has positive and negative impacts in accordance with a preferred embodiment;

FIG. 38 is a block diagram of the hierarchical relationship of a transaction in accordance with a preferred embodiment;

FIG. 39 is a block diagram illustrating the feedback hierarchy in accordance with a preferred embodiment;

FIG. 40 is a block diagram illustrating how the simulation engine is architected into a preferred embodiment of the invention;

FIG. 41 is a block diagram setting forth the architecture of a simulation model in accordance with a preferred embodiment;

FIG. 42 illustrates the arithmetic steps in accordance with a preferred embodiment;

FIG. 43 illustrates a drag & drop input operation in accordance with a preferred embodiment;

FIG. 44 illustrates list object processing in accordance with a preferred embodiment;

FIG. 45 illustrates the steps for configuring a simulation in accordance with a preferred embodiment;

FIG. 46 illustrates a distinct output in accordance with a preferred embodiment;

FIG. 47 is a block diagram presenting the detailed architecture of a system dynamics model in accordance with a preferred embodiment;

FIG. 48 is graphical representation of the object model which is utilized to instantiate the system dynamic engine in accordance with a preferred embodiment.

FIG. 49 is a PInput Cell for a simulation model in accordance with a preferred embodiment;

FIG. 50 is a PInput backup cell in a simulation model in accordance with a preferred embodiment;

FIG. 51 is a display illustrating a POutput cell in accordance with a preferred embodiment. The steps required to configure the POutput are presented below;

FIG. 52 is an overview diagram of the logic utilized for initial configuration in accordance with a preferred embodiment;

FIG. 53 is a display of the source item and target configuration in accordance with a preferred embodiment;

FIG. 54 is a display of video information in accordance with a preferred embodiment;

FIG. 55 illustrates a display depicting configured rules in accordance with a preferred embodiment;

FIG. 56 illustrates feedback for configured rules in accordance with a preferred embodiment;

FIG. 57 illustrates a display with follow-up configuration questions in accordance with a preferred embodiment;

FIG. 58 illustrates configuration of aggregate rules in accordance with a preferred embodiment;

FIG. 59 illustrates a set of coach items in accordance with a preferred embodiment;

FIG. 60 is an ICA Meeting Configuration tool display in accordance with a preferred embodiment;

FIG. 61 illustrates an ICA utility in accordance with a preferred embodiment;

FIG. 62 illustrates a configuration utility display in accordance with a preferred embodiment;

FIG. 63 illustrates an object editor toolbar in accordance with a preferred embodiment;

FIG. 64 illustrates the seven areas that can be configured for a simulation in accordance with a preferred embodiment;

FIG. 65 illustrates a display that defines inputs in accordance with a preferred embodiment;

FIG. 66 illustrates a list editor in accordance with a preferred embodiment;

FIG. 67A illustrates a define student display in accordance with a preferred embodiment;

FIG. 67B illustrates a ControlSourceItem display in accordance with a preferred embodiment;

FIG. 68 illustrates a simulation workbench in accordance with a preferred embodiment;

FIG. 69 illustrates an object viewer in accordance with a preferred embodiment. As shown in

FIG. 70 illustrates an Object Viewer Configuration in an Utilities menu in accordance with a preferred embodiment;

FIG. 71 illustrates a log viewer in accordance with a preferred embodiment;

FIG. 72 illustrates a Doc Maker display in accordance with a preferred embodiment;

FIG. 73 illustrates a Feedback display in accordance with a preferred embodiment;

FIG. 74 is an object editor display that illustrates the use of references in accordance with a preferred embodiment; and

FIG. 75 presents the detailed design of smart spreadsheets in accordance with a preferred embodiment.

DETAILED DESCRIPTION

A preferred embodiment of a system in accordance with the present invention is preferably practiced in the context of a personal computer such as an IBM compatible personal computer, Apple Macintosh computer or UNIX based workstation. A representative hardware environment is depicted in FIG. 1, which illustrates a typical hardware configuration of a workstation in accordance with a preferred embodiment having a central processing unit 110, such as a microprocessor, and a number of other units interconnected via a system bus 112. The workstation shown in FIG. 1 includes a Random Access Memory (RAM) 114, Read Only Memory (ROM) 116, an I/O adapter 118 for connecting peripheral devices such as disk storage units 120 to the bus 112, a user interface adapter 122 for connecting a keyboard 124, a mouse 126, a speaker 128, a microphone 132, and/or other user interface devices such as a touch screen (not shown) to the bus 112, communication adapter 134 for connecting the workstation to a communication network (e.g., a data processing network) and a display adapter 136 for connecting the bus 112 to a display device 138. The workstation typically has resident thereon an operating system such as the Microsoft Windows NT or Windows/95 Operating System (OS), the IBM OS/2 operating system, the MAC OS, or UNIX operating system. Those skilled in the art will appreciate that the present invention may also be implemented on platforms and operating systems other than those mentioned.

A preferred embodiment is written using JAVA, C, and the C++ language and utilizes object oriented programming methodology. Object oriented programming (OOP) has become increasingly used to develop complex applications. As OOP moves toward the mainstream of software design and development, various software solutions require adaptation to make use of the benefits of OOP. A need exists for these principles of OOP to be applied to a messaging interface of an electronic messaging system such that a set of OOP classes and objects for the messaging interface can be provided.

OOP is a process of developing computer software using objects, including the steps of analyzing the problem, designing the system, and constructing the program. An object is a software package that contains both data and a collection of related structures and procedures. Since it contains both data and a collection of structures and procedures, it can be visualized as a self-sufficient component that does not require other additional structures, procedures or data to perform its specific task. OOP, therefore, views a computer program as a collection of largely autonomous components, called objects, each of which is responsible for a specific task. This concept of packaging data, structures, and procedures together in one component or module is called encapsulation.

In general, OOP components are reusable software modules which present an interface that conforms to an object model and which are accessed at run-time through a component integration architecture. A component integration architecture is a set of architecture mechanisms which allow software modules in different process spaces to utilize each others capabilities or functions. This is generally done by assuming a common component object model on which to build the architecture. It is worthwhile to differentiate between an object and a class of objects at this point. An object is a single instance of the class of objects, which is often just called a class. A class of objects can be viewed as a blueprint, from which many objects can be formed.

OOP allows the programmer to create an object that is a part of another object. For example, the object representing a piston engine is said to have a composition-relationship with the object representing a piston. In reality, a piston engine comprises a piston, valves and many other components; the fact that a piston is an element of a piston engine can be logically and semantically represented in OOP by two objects.

OOP also allows creation of an object that "depends from" another object. If there are two objects, one representing a piston engine and the other representing a piston engine wherein the piston is made of ceramic, then the relationship between the two objects is not that of composition. A ceramic piston engine does not make up a piston engine. Rather it is merely one kind of piston engine that has one more limitation than the piston engine; its piston is made of ceramic. In this case, the object representing the ceramic piston engine is called a derived object, and it inherits all of the aspects of the object representing the piston engine and adds further limitation or detail to it. The object representing the ceramic piston engine "depends from" the object representing the piston engine. The relationship between these objects is called inheritance.

When the object or class representing the ceramic piston engine inherits all of the aspects of the objects representing the piston engine, it inherits the thermal characteristics of a standard piston defined in the piston engine class. However, the ceramic piston engine object overrides these ceramic specific thermal characteristics, which are typically different from those associated with a metal piston. It skips over the original and uses new functions related to ceramic pistons. Different kinds of piston engines have different characteristics, but may have the same underlying functions associated with it (e.g., how many pistons in the engine, ignition sequences, lubrication, etc.). To access each of these functions in any piston engine object, a programmer would call the same functions with the same names, but each type of piston engine may have different/overriding implementations of functions behind the same name. This ability to hide different implementations of a function behind the same name is called polymorphism and it greatly simplifies communication among objects.

With the concepts of composition-relationship, encapsulation, inheritance and polymorphism, an object can represent just about anything in the real world. In fact, our logical perception of the reality is the only limit on determining the kinds of things that can become objects in object-oriented software. Some typical categories are as follows:

Objects can represent physical objects, such as automobiles in a traffic-flow simulation, electrical components in a circuit-design program, countries in an economics model, or aircraft in an air-traffic-control system.

Objects can represent elements of the computer-user environment such as windows, menus or graphics objects.

An object can represent an inventory, such as a personnel file or a table of the latitudes and longitudes of cities.

An object can represent user-defined data types such as time, angles, and complex numbers, or points on the plane.

With this enormous capability of an object to represent just about any logically separable matters, OOP allows the software developer to design and implement a computer program that is a model of some aspects of reality, whether that reality is a physical entity, a process, a system, or a composition of matter. Since the object can represent anything, the software developer can create an object which can be used as a component in a larger software project in the future.

If 90% of a new OOP software program consists of proven, existing components made from preexisting reusable objects, then only the remaining 10% of the new software project has to be written and tested from scratch. Since 90% already came from an inventory of extensively tested reusable objects, the potential domain from which an error could originate is 10% of the program. As a result, OOP enables software developers to build objects out of other, previously built objects.

This process closely resembles complex machinery being built out of assemblies and sub-assemblies. OOP technology, therefore, makes software engineering more like hardware engineering in that software is built from existing components, which are available to the developer as objects. All this adds up to an improved quality of the software as well as an increased speed of its development.

Programming languages are beginning to fully support the OOP principles, such as encapsulation, inheritance, polymorphism, and composition-relationship. With the advent of the C++ language, many commercial software developers have embraced OOP. C++ is an OOP language that offers a fast, machine-executable code. Furthermore, C++ is suitable for both commercial-application and systems-programming projects. For now, C++ appears to be the most popular choice among many OOP programmers, but there is a host of other OOP languages, such as Smalltalk, Common Lisp Object System (CLOS), and Eiffel. Additionally, OOP capabilities are being added to more traditional popular computer programming languages such as Pascal.

The benefits of object classes can be summarized, as follows:

Objects and their corresponding classes break down complex programming problems into many smaller, simpler problems.

Encapsulation enforces data abstraction through the organization of data into small, independent objects that can communicate with each other. Encapsulation protects the data in an object from accidental damage, but allows other objects to interact with that data by calling the object's member functions and structures.

Subclassing and inheritance make it possible to extend and modify objects through deriving new kinds of objects from the standard classes available in the system. Thus, new capabilities are created without having to start from scratch.

Polymorphism and multiple inheritance make it possible for different programmers to mix and match characteristics of many different classes and create specialized objects that can still work with related objects in predictable ways.

Class hierarchies and containment hierarchies provide a flexible mechanism for modeling real-world objects and the relationships among them.

Libraries of reusable classes are useful in many situations, but they also have some limitations. For example:

Complexity. In a complex system, the class hierarchies for related classes can become extremely confusing, with many dozens or even hundreds of classes.

Flow of control. A program written with the aid of class libraries is still responsible for the flow of control (i.e., it must control the interactions among all the objects created from a particular library). The programmer has to decide which functions to call at what times for which kinds of objects.

Duplication of effort. Although class libraries allow programmers to use and reuse many small pieces of code, each programmer puts those pieces together in a different way. Two different programmers can use the same set of class libraries to write two programs that do exactly the same thing but whose internal structure (i.e., design) may be quite different, depending on hundreds of small decisions each programmer makes along the way. Inevitably, similar pieces of code end up doing similar things in slightly different ways and do not work as well together as they should.

Class libraries are very flexible. As programs grow more complex, more programmers are forced to reinvent basic solutions to basic problems over and over again. A relatively new extension of the class library concept is to have a framework of class libraries. This framework is more complex and consists of significant collections of collaborating classes that capture both the small scale patterns and major mechanisms that implement the common requirements and design in a specific application domain. They were first developed to free application programmers from the chores involved in displaying menus, windows, dialog boxes, and other standard user interface elements for personal computers.

Frameworks also represent a change in the way programmers think about the interaction between the code they write and code written by others. In the early days of procedural programming, the programmer called libraries provided by the operating system to perform certain tasks, but basically the program executed down the page from start to finish, and the programmer was solely responsible for the flow of control. This was appropriate for printing out paychecks, calculating a mathematical table, or solving other problems with a program that executed in just one way.

The development of graphical user interfaces began to turn this procedural programming arrangement inside out. These interfaces allow the user, rather than program logic, to drive the program and decide when certain actions should be performed. Today, most personal computer software accomplishes this by means of an event loop which monitors the mouse, keyboard, and other sources of external events and calls the appropriate parts of the programmer's code according to actions that the user performs. The programmer no longer determines the order in which events occur. Instead, a program is divided into separate pieces that are called at unpredictable times and in an unpredictable order. By relinquishing control in this way to users, the developer creates a program that is much easier to use. Nevertheless, individual pieces of the program written by the developer still call libraries provided by the operating system to accomplish certain tasks, and the programmer must still determine the flow of control within each piece after it's called by the event loop. Application code still "sits on top of" the system.

Even event loop programs require programmers to write a lot of code that should not need to be written separately for every application. The concept of an application framework carries the event loop concept further. Instead of dealing with all the nuts and bolts of constructing basic menus, windows, and dialog boxes and then making these things all work together, programmers using application frameworks start with working application code and basic user interface elements in place. Subsequently, they build from there by replacing some of the generic capabilities of the framework with the specific capabilities of the intended application.

Application frameworks reduce the total amount of code that a programmer has to write from scratch. However, because the framework is really a generic application that displays windows, supports copy and paste, and so on, the programmer can also relinquish control to a greater degree than event loop programs permit. The framework code takes care of almost all event handling and flow of control, and the programmer's code is called only when the framework needs it (e.g., to create or manipulate a proprietary data structure).

A programmer writing a framework program not only relinquishes control to the user (as is also true for event loop programs), but also relinquishes the detailed flow of control within the program to the framework. This approach allows the creation of more complex systems that work together in interesting ways, as opposed to isolated programs, having custom code, being created over and over again for similar problems.

Thus, as is explained above, a framework basically is a collection of cooperating classes that make up a reusable design solution for a given problem domain. It typically includes objects that provide default behavior (e.g., for menus and windows), and programmers use it by inheriting some of that default behavior and overriding other behavior so that the framework calls application code at the appropriate times.

There are three main differences between frameworks and class libraries:

Behavior versus protocol. Class libraries are essentially collections of behaviors that you can call when you want those individual behaviors in your program. A framework, on the other hand, provides not only behavior but also the protocol or set of rules that govern the ways in which behaviors can be combined, including rules for what a programmer is supposed to provide versus what the framework provides.

Call versus override. With a class library, the code the programmer instantiates objects and calls their member functions. It's possible to instantiate and call objects in the same way with a framework (i.e., to treat the framework as a class library), but to take full advantage of a framework's reusable design, a programmer typically writes code that overrides and is called by the framework. The framework manages the flow of control among its objects. Writing a program involves dividing responsibilities among the various pieces of software that are called by the framework rather than specifying how the different pieces should work together.

Implementation versus design. With class libraries, programmers reuse only implementations, whereas with frameworks, they reuse design. A framework embodies the way a family of related programs or pieces of software work. It represents a generic design solution that can be adapted to a variety of specific problems in a given domain. For example, a single framework can embody the way a user interface works, even though two different user interfaces created with the same framework might solve quite different interface problems.

Thus, through the development of frameworks for solutions to various problems and programming tasks, significant reductions in the design and development effort for software can be achieved. A preferred embodiment of the invention utilizes HyperText Markup Language (HTML) to implement documents on the Internet together with a general-purpose secure communication protocol for a transport medium between the client and the Newco. HTTP or other protocols could be readily substituted for HTML without undue experimentation. Information on these products is available in T. Berners-Lee, D. Connoly, "RFC 1866:Hypertext Markup Language-2.0" (November 1995); and R. Fielding, H, Frystyk, T. Berners-Lee, J. Gettys and J. C. Mogul, "Hypertext Transfer Protocol--HTTP/1.1:HTTP Working Group Internet Draft" (May 2, 1996). HTML is a simple data format used to create hypertext documents that are portable from one platform to another. HTML documents are SGML documents with generic semantics that are appropriate for representing information from a wide range of domains. HTML has been in use by the World-Wide Web global information initiative since 1990. HTML is an application of ISO Standard 8879; 1986 Information Processing Text and Office Systems; Standard Generalized Markup Language (SGML).

To date, Web development tools have been limited in their ability to create dynamic Web applications which span from client to server and interoperate with existing computing resources. Until recently, HTML has been the dominant technology used in development of Web-based solutions. However, HTML has proven to be inadequate in the following areas:

Poor performance;

Restricted user interface capabilities;

Can only produce static Web pages;

Lack of interoperability with existing applications and data; and

Inability to scale.

Sun Microsystem's Java language solves many of the client-side problems by:

Improving performance on the client side;

Enabling the creation of dynamic, real-time Web applications; and

Providing the ability to create a wide variety of user interface components.

With Java, developers can create robust User Interface (UI) components. Custom "widgets" (e.g., real-time stock tickers, animated icons, etc.) can be created, and client-side performance is improved. Unlike HTML, Java supports the notion of client-side validation, offloading appropriate processing onto the client for improved performance. Dynamic, real-time Web pages can be created. Using the above-mentioned custom UI components, dynamic Web pages can also be created.

Sun's Java language has emerged as an industry-recognized language for "programming the Internet." Sun defines Java as: "a simple, object-oriented, distributed, interpreted, robust, secure, architecture-neutral, portable, high-performance, multithreaded, dynamic, buzzword-compliant, general-purpose programming language. Java supports programming for the Internet in the form of platform-independent Java applets." Java applets are small, specialized applications that comply with Sun's Java Application Programming Interface (API) allowing developers to add "interactive content" to Web documents (e.g., simple animations, page adornments, basic games, etc.). Applets execute within a Java-compatible browser (e.g., Netscape Navigator) by copying code from the server to client. From a language standpoint, Java's core feature set is based on C++. Sun's Java literature states that Java is basically, "C++ with extensions from Objective C for more dynamic method resolution."

Another technology that provides similar function to JAVA is provided by Microsoft and ActiveX Technologies, to give developers and Web designers wherewithal to build dynamic content for the Internet and personal computers. ActiveX includes tools for developing animation, 3-D virtual reality, video and other multimedia content. The tools use Internet standards, work on multiple platforms, and are being supported by over 100 companies. The group's building blocks are called ActiveX Controls, small, fast components that enable developers to embed parts of software in hypertext markup language (HTML) pages. ActiveX Controls work with a variety of programming languages including Microsoft Visual C++, Borland Delphi, Microsoft Visual Basic programming system and, in the future, Microsoft's development tool for Java, code named "Jakarta." ActiveX Technologies also includes ActiveX Server Framework, allowing developers to create server applications. One of ordinary skill in the art readily recognizes that ActiveX could be substituted for JAVA without undue experimentation to practice the invention.

A simulation engine in accordance with a preferred embodiment is based on a Microsoft Visual Basic component developed to help design and test feedback in relation to a Microsoft Excel spreadsheet. These spreadsheet models are what simulate actual business functions and become a task that will be performed by a student. The Simulation Engine accepts simulation inputs and calculates various outputs and notifies the system of the status of the simulation at a given time in order to obtain appropriate feedback.

Relationship of Components

The simulation model executes the business function that the student is learning and is therefore the center point of the application. An activity `layer` allows the user to visually guide the simulation by passing inputs into the simulation engine and receiving an output from the simulation model. For example, if the student was working on an income statement activity, the net sales and cost of goods sold calculations are passed as inputs to the simulation model and the net income value is calculated and retrieved as an output. As calculations are passed to and retrieved from the simulation model, they are also passed to the Intelligent Coaching Agent (ICA). The ICA analyzes the Inputs and Outputs to the simulation model and generates feedback based on a set of rules. This feedback is received and displayed through the Visual Basic Architecture.

FIG. 2 is a block diagram of a system architecture in accordance with a preferred embodiment. The Presentation `layer` 210 is separate from the activity `layer` 220 and communication is facilitated through a set of messages 230 that control the display specific content topics. A preferred embodiment enables knowledge workers 200 & 201 to acquire complex skills rapidly, reliably and consistently across an organization to deliver rapid acquisition of complex skills.

This result is achieved by placing individuals in a simulated business environment that "looks and feels" like real work, and challenging them to make decisions which support a business' strategic objectives utilizing highly effective learning theory (e.g., goal based learning, learn by doing, failure based learning, etc.), and the latest in multimedia user interfaces, coupled with three powerful, integrated software components. The first of these components is a software Solution Construction Aid (SCA) 230 consisting of a mathematical modeling tool 234 which simulates business outcomes of an individual's collective actions over a period of time. The second component is a knowledge system 250 consisting of an HTML content layer which organizes and presents packaged knowledge much like an online text book with practice exercises, video war stories, and a glossary. The third component is a software tutor 270 comprising an artificial intelligence engine 240 which generates individualized coaching messages based on decisions made by learner.

Feedback is unique for each individual completing the course and supports client cultural messages 242 "designed into" the course. A business simulation methodology that includes support for content acquisition, story line design, interaction design, feedback and coaching delivery, and content delivery is architected into the system in accordance with a preferred embodiment. A large number of "pre-designed" learning interactions such as drag and drop association of information 238, situation assessment/action planning, interviewing (one-on-one, one-to-many), presenting (to a group of experts/executives), metering of performance (handle now, handle later), "time jumping" for impact of decisions, competitive landscape shift (while "time jumping", competitors merge, customers are acquired, etc.) and video interviewing with automated note taking are also included in accordance with a preferred embodiment.

Business simulation in accordance with a preferred embodiment delivers training curricula in an optimal manner. This is because such applications provide effective training that mirrors a student's actual work environment. The application of skills "on the job" facilitates increased retention and higher overall job performance. While the results of such training applications are impressive, business simulations are very complex to design and build correctly. These simulations are characterized by a very open-ended environment, where students can go through the application along any number of paths, depending on their learning style and prior experiences/knowledge.

A category of learning approaches called Learn by Doing, is commonly used as a solution to support the first phase (Learn) of the Workforce Performance Cycle. However, it can also be a solution to support the second phase (Perform) of the cycle to enable point of need learning during job performance. By adopting the approach presented, some of the benefits of a technology based approach for building business simulation solutions which create more repeatable, predictable projects resulting in more perceived and actual user value at a lower cost and in less time are highlighted.

Most corporate training programs today are misdirected because they have failed to focus properly on the purpose of their training. These programs have confused the memorization of facts with the ability to perform tasks; the knowing of "that" with the knowing of "how". By adopting the methods of traditional schools, businesses are teaching a wide breadth of disconnected, decontextualized facts and figures, when they should be focused on improved performance. How do you teach performance, when lectures, books, and tests inherently are designed around facts and figures? Throw away the lectures, books, and tests. The best way to prepare for high performance is to perform; experience is the best teacher! Most business leaders agree that workers become more effective the more time they spend in their jobs. The best approach for training novice employees, therefore, would be letting them learn on the job, acquiring skills in their actual work environment. The idea of learning-by-doing is not revolutionary, yet it is resisted in business and academia. Why is this so, if higher competence is universally desired?

Learners are reluctant to adopt learning-by-doing because they are frightened of failure. People work hard to avoid making mistakes in front of others. Business leaders are hesitant to implement learning-by-doing because novice failure may have dramatic safety, legal and financial implications. Imagine a novice pilot learning-by-doing as he accelerates a large jet plane down a runway; likewise, consider a new financial analyst learning-by-doing as he structures a multi-million dollar financial loan. Few employers are willing to endure such failures to have a more competent workforce.

The concerns of employee and employer can be relieved if the training (and accompanying failure) didn't occur in front of co-workers and clients, if it didn't jeopardize a multi-million dollar aircraft or a multi-million dollar deal. What if the training was performed privately, in a richly modeled simulation of the workers actual job? In a simulated environment, failure would result in dedicated instruction instead of embarrassment, injury, or monetary losses. Simulated environments provide a sense of liberation for exploration that does not exist in the real world. Knowing that the consequences of experimentation will unlikely be dire, learners are able to explore the "what if . . . " inherent in us all. In this way, the best way to prepare for high performance is to simulate actual performance. New media technologies utilizing multimedia provide the possibility to create such business simulation experiences.

Even if companies didn't make the mistake of focusing on "what" learning vs. "how" learning, they still tend to have the overly narrow view of education/training as something that only occurs prior to workers being asked to actually perform their job. Learning is something that is constantly occurring, and doesn't cease once "real work" has begun. The closer new lessons occur in time with the application of those lessons, the greater the resultant learning. This fact helps to explain some of the reasoning behind the benefits of business simulation, described in the previous section. In those systems, all new lessons are taught in close relationship to their real world use; everything is in context and, most importantly, are presented at the appropriate time. But as the properly trained worker performs their job, the working environment changes. New demands occur, and new methods and rules are developed. As these events occur, there is a need for new support/training that, in most cases, must wait to be addressed until the next "pre-performance" training session.

In these cases, the need (or demand) for additional training doesn't match the supply. This lag between a training need and the fulfilling lesson has a dramatic negative impact on productivity, accuracy, and customer satisfaction. Workers need to have the opportunity to learn while they are performing. Just as during pre-performance training, one powerful mechanism for identifying and correcting (simulated) performance problems is to have an expert available at all time to watch your actions and remediate when appropriate. This, obviously, is too costly and time intensive of an approach to be practical with actual experts. But what if workers had access to a support system that provided the majority of the benefits of a real expert, transparently integrated into their work environment? Such a system would provide advice at key moments in the work flow for problem resolution and/or process improvement, tools to ease task completion, reference material of best practice knowledge, and point of need training courses. With a support system that proactively assists the worker in performance of their job tasks at a higher level of competency, productivity and customer satisfaction (both internal and external) would soar.

The key to such a support system is that it is seamlessly integrated into the business system that the knowledge worker uses to execute their job tasks. Workers don't need to go "off-line" or seek out cryptic information buried within paper manuals and binders for guidance or to find the answer to queries. All the support components are made available through the same applications the worker's use, at the point in which they need them, tailored to the individual to show "how", not just "what". Learning would be occurring all the time, with little distinction between performing and improving performance.

Establishing that training should focus on performance (how), rather than facts (what), and extending the model of learning to include assistance while performing, rather than only before performance, still leaves us dangerously exposed in preparing to compete in the new, chaotic economy. As was mentioned in the opening of this paper, the pace of change in business today is whiplash fast. Not only are new methods of doing business evolving every 18-24 months, new competitors emerge, dominate, and fade in time periods businesses used to take to perform demographic studies. Now more than ever, those who do not reinvent themselves on a regular basis will be fossilized by the pace of change.

Even the best pre-performance training and the most advanced performance support tools are destined to be outdated if there isn't a fresh supply of real-world requirements and lessons learned being fed back as inputs for the next go 'round. Innovation is a key step in the Workforce Performance Cycle. This step requires business to employ Stephen Covey's famous habit of "sharpening the saw", or "take time to be fast".

There is an untold wealth of information buried within the heads of business users responsible for implementing the steps outlined in their pre-performance training and performance support tools. No other group within an organization can more accurately assess the effectiveness of current methods, or project needed changes that will have dramatic future impact. This step of reflecting on the current and past state of affairs, uncovering new approaches by identifying what is working and what is not, and adapting accordingly for the future is the last phase of the learning/performance cycle.

Innovation to fuel future training and performance support comes directly from the community most closely tied to task performance. Effective businesses need to develop and cultivate a mechanism for communication and collaboration among the experts in these communities to more efficiently benefit from their collective wisdom. In today's business, that which is evident to your business is evident to nearly all your competitors as well. The competitive advantage lies in uncovering that which is unexpected or not immediately apparent, adapting your business processes to exploit the discovery, and quickly, but effectively, educating your workforce on the new policies and procedures, all before the competition catches on or the market changes again.

This innovation process is the critical final step in continuous education of the most effective and up-to-date policies, procedures, and information. Without formalized innovation, companies not only risk being a step behind the ever advancing competition, but compound the problem by continuing to train their personnel with outdated strategies and information. One way to formalize this vital step in the Workforce Performance Cycle is to construct Virtual Learning Communities, where many `experts` can share experiences, submit ideas for improvements, play out "what-if" scenarios, and contribute on complex problems that may be insurmountable without significant collaboration with others. Such Learning Communities could nurture up-to-date discussion of what is actually happening within a business, eliminating the traditional information-passing lag that plagues many business as new data travels through corporate hierarchies. This increased nimbleness would help organizations quickly address new competitive trends and outdated strategies, identify potential solutions, and implement improved processes in the form of updated training and performance support reference materials.

Currently, a typical BusSim engagement takes between one and two years to complete and requires a variety of both functional and technical skills. FIG. 3 depicts the timeline and relative resource requirements for each phase of development for a typical application development in accordance with a preferred embodiment. The chart clearly depicts the relationship between the large number of technical resources required for both the build and test phases of development. This is because the traditional development process used to build BusSim solutions reflects more of a "one off" philosophy, where development is done from scratch in a monolithic fashion, with little or no reuse from one application to the next. This lack of reuse makes this approach prohibitively expensive, as well as lengthy, for future BusSim projects.

The solution to this problem is to put tools in the hands of instructional designers that allows them to create their BusSim designs and implement them without the need for programmers to write code. And to put application architectures that integrate with the tools in the hands of developers, providing them with the ability to quickly deliver solutions for a number of different platforms. The reuse, then, comes in using the tools and architectures from one engagement to another. Both functional and technical resources carry with them the knowledge of how to use the technology, which also has an associated benefit of establishing a best-practice development methodology for BusSim engagements.

The tools and architectures described herein are part of a next-generation Business Simulation delivery framework that will reduce the total effort necessary to build solutions in accordance with a preferred embodiment. FIG. 4 depicts the potential savings in both functional and technical tasks in accordance with a preferred embodiment.

Development Cycle Activities

Design Phase

In the Design Phase, instructional designers become oriented to the content area and begin to conceptualize an instructional approach. They familiarize themselves with the subject matter through reading materials and interviews with Subject Matter Experts (SMEs). They also identify learning objectives from key client contacts. Conceptual designs for student interactions and interface layouts also begin to emerge. After the conceptual designs have taken shape, Low-Fi user testing (a.k.a Conference Room Piloting) is performed. Students interact with interface mock-ups while facilitators observe and record any issues. Finally, detailed designs are created that incorporate findings. These detailed designs are handed off to the development team for implementation.

The design phase has traditionally been fraught with several problems. Unlike a traditional business system, BusSim solutions are not rooted in tangible business processes, so requirements are difficult to identify in a concrete way. This leaves instructional designers with a `blue sky` design problem. With few business-driven constraints on the solution, shallow expertise in the content area, and limited technical skills, instructional designers have little help in beginning a design. Typically, only experienced designers have been able to conjure interface, analysis, and feedback designs that meet the learning objectives yet remain technically feasible to implement. To compound the problem, BusSim solutions are very open ended in nature. The designer must anticipate a huge combination of student behavior to design feedback that is helpful and realistic.

Build Phase

During the build phase, the application development team uses the detailed designs to code the application. Coding tasks include the interfaces and widgets that the student interacts with. The interfaces can be made up of buttons, grids, check boxes, or any other screen controls that allow the student to view and manipulate his deliverables. The developer must also code logic that analyzes the student's work and provides feedback interactions. These interactions may take the form of text and/or multimedia feedback from simulated team members, conversations with simulated team members, or direct manipulations of the student's work by simulated team members. In parallel with these coding efforts, graphics, videos, and audio are being created for use in the application. Managing the development of these assets have their own complications.

Risks in the build phase include misinterpretation of the designs. If the developer does not accurately understand the designer's intentions, the application will not function as desired. Also, coding these applications requires very skilled developers because the logic that analyzes the student's work and composes feedback is very complex.

Test Phase

The Test Phase, as the name implies, is for testing the application. Testing is performed to verify the application in three ways: first that the application functions properly (functional testing), second that the students understand the interface and can navigate effectively (usability testing), and third that the learning objectives are met (cognition testing). Functional testing of the application can be carried out by the development team or by a dedicated test team. If the application fails to function properly, it is debugged, fixed, recompiled and retested until its operation is satisfactory. Usability and cognition testing can only be carried out by test students who are unfamiliar with the application. If usability is unsatisfactory, parts of the interface and or feedback logic may need to be redesigned, recoded, and retested. If the learning objectives are not met, large parts of the application may need to be removed and completely redeveloped from a different perspective.

The test phase is typically where most of the difficulties in the BusSim development cycle are encountered. The process of discovering and fixing functional, usability, and cognition problems is a difficult process and not an exact science.

For functional testing, testers operate the application, either by following a test script or by acting spontaneously and documenting their actions as they go. When a problem or unexpected result is encountered, it too is documented. The application developer responsible for that part of the application then receives the documentation and attempts to duplicate the problem by repeating the tester's actions. When the problem is duplicated, the developer investigates further to find the cause and implement a fix. The developer once again repeats the tester's actions to verify that the fix solved the problem. Finally, all other test scripts must be rerun to verify that the fix did not have unintended consequences elsewhere in the application.

This process has inherent difficulty. If the tester is inaccurate in recording his actions, or the developer is inaccurate in repeating them, then the problem may never be duplicated and the defect never found. Furthermore, the problem may be dependent on some condition in the tester's environment that is not readily observable or is not even related to the application. This process has proven to be tedious, time-consuming, and of limited reliability.

For usability testing, test students operate the application as it will be operated in production. Ideally, their progress is only impeded by their lack of mastery of the content. As they gain proficiency, they are able to complete the activities and move on. As is often the case, however, they are impeded by unclear instructions, a non-intuitive interface, or other usability shortcomings. When these difficulties are encountered, the facilitators document the problems and student comments on what is needed to improve usability.

There are two major risks associated with usability testing. First, student action recording is not as rigorous because actual students are performing the testing, so functional problems that don't appear until now are difficult to reproduce. Second, resolutions to usability problems often involve significant modification to the application code and interface which requires repeating of portions of design, build, and test.

For cognition testing, students are surveyed and/or tested to determine their level of understanding of the material. If results indicate that the learning objectives are not being adequately met, the design is reevaluated. Changes proposed to improve the cognition may include massive redesign and rebuilding.

Execution Phase

The Execution Phase refers to the steady state operation of the completed application in its production environment. For some clients, this involves phone support for students. Clients may also want the ability to track students' progress and control their progression through the course. Lastly, clients may want the ability to track issues so they may be considered for inclusion in course maintenance releases.

One of the key values of on-line courses is that they can be taken at a time, location, and pace that is convenient for the individual student. However, because students are not centrally located, support is not always readily available. For this reason it is often desirable to have phone support for students.

Clients may also desire to track students' progress, or control their advancement through the course. Under this strategy, after a student completes a section of the course, he will transfer his progress data to a processing center either electronically or by physically mailing a disk. There it can be analyzed to verify that he completed all required work satisfactorily. One difficulty commonly associated with student tracking is isolating the student data for analysis. It can be unwieldy to transmit all the course data, so it is often imperative to isolate the minimum data required to perform the necessary analysis of the student's progress.

A Delivery Framework for Business Simulation

As discussed earlier, the traditional development process used to build BusSim solutions reflects more of a "one off" philosophy, where development is done from scratch in a monolithic fashion, with little or no reuse from one application to the next. A better approach would be to focus on reducing the total effort required for development through reuse, which, in turn would decrease cost and development time.

The first step in considering reuse as an option is the identification of common aspects of the different BusSim applications that can be generalized to be useful in future applications. In examination of the elements that make up these applications, three common aspects emerge as integral parts of each:

Interface

Analysis

Interpretation

Interface

Every BusSim application must have a mechanism for interaction with the student. The degree of complexity of each interface may vary, from the high interactivity of a high-fidelity real-time simulation task, to the less complex information delivery requirements of a business case background information task. Regardless of how sophisticated the User Interface (UI), it is a vital piece of making the underlying simulation and feedback logic useful to the end user.

Analysis

Every BusSim application does analysis on the data that defines the current state of the simulation many times throughout the execution of the application. This analysis is done either to determine what is happening in the simulation, or to perform additional calculations on the data which are then fed back into the simulation. For example, the analysis may be the recognition of any actions the student has taken on artifacts within the simulated environment (notebooks, number values, interviews conducted, etc.), or it may be the calculation of an ROI based on numbers the student has supplied.

Interpretation

Substantive, useful feedback is a critical piece of any BusSim application. It is the main mechanism to communicate if actions taken by the student are helping or hurting them meet their performance objectives. The interpretation piece of the set of proposed commonalties takes the results of any analysis performed and makes sense of it. It takes the non-biased view of the world that the Analysis portion delivers (i.e., "Demand is up 3%") and places some evaluative context around it (i.e., "Demand is below the expected 7%; you're in trouble!", or "Demand has exceeded projections of 1.5%; Great job!"). FIG. 5 illustrates commonalties in accordance with a preferred embodiment.

Common Features of Business Simulation Applications

There are several approaches to capturing commonalties for reuse. Two of the more common approaches are framework-based and component-based. To help illustrate the differences between the two approaches, we will draw an analogy between building an application and building a house. One can construct a house from scratch, using the raw materials, 2.times.4s, nails, paint, concrete, etc. One can also construct an application from scratch, using the raw materials of new designs and new code. The effort involved in both undertakings can be reduced through framework-based and/or component-based reuse.

Framework-Based Reuse

Within the paradigm of framework-based reuse, a generic framework or architecture is constructed that contains commonalties. In the house analogy, one could purchase a prefabricated house framework consisting of floors, outside walls, bearing walls and a roof. The house can be customized by adding partition walls, wall-paper, woodwork, carpeting etc. Similarly, prefabricated application frameworks are available that contain baseline application structure and functionality. Individual applications are completed by adding specific functionality and customizing the look-and-feel. An example of a commonly used application framework is Microsoft Foundation Classes. It is a framework for developing Windows applications using C++. MFC supplies the base functionality of a windowing application and the developer completes the application by adding functionality within the framework. Framework-based reuse is best suited for capturing template-like features, for example user interface management, procedural object behaviors, and any other features that may require specialization.

Some benefits of using a framework include:

Extensive Functionality Can Be Incorporated into a Framework

In the house analogy, if I know I am going to build a whole neighborhood of three bedroom ranches, I can build the plumbing, wiring, and partition walls right into the framework, reducing the incremental effort required for each house. If I know I am going to build a large number of very similar applications, they will have more commonalties that can be included in the framework rather than built individually.

Applications Can Override the Framework-supplied Functionality wherever Appropriate

If a house framework came with pre-painted walls, the builder could just paint over them with preferred colors. Similarly, the object oriented principle of inheritance allows an application developer to override the behavior of the framework.

Component-Based Reuse

In the paradigm of component-based reuse, key functionality is encapsulated in a component. The component can then be reused in multiple applications. In the house analogy, components correspond to appliances such as dishwashers, refrigerators, microwaves, etc. Similarly, many application components with pre-packaged functionality are available from a variety of vendors. An example of a popular component is a Data Grid. It is a component that can be integrated into an application to deliver the capability of viewing columnar data in a spreadsheet-like grid. Component-based reuse is best suited for capturing black-box-like features, for example text processing, data manipulation, or any other features that do not require specialization.

Some benefits of using components include:

Several Applications on the Same Computer Can Share a Single Component

This is not such a good fit with the analogy, but imagine if all the houses in a neighborhood could share the same dishwasher simultaneously. Each home would have to supply its own dishes, detergent, and water, but they could all wash dishes in parallel. In the application component world, this type of sharing is easily accomplished and results in reduced disk and memory requirements.

Components Tend to Be Less Platform and Tool Dependent.

A microwave can be used in virtually any house, whether it's framework is steel or wood, and regardless of whether it was customized for building mansions or shacks. You can put a high-end microwave in a low-end house and vice-versa. You can even have multiple different microwaves in your house. component Technologies such as CORBA, COM, and Java Beans make this kind of flexibility commonplace in application development.

The Solution: A Combined Approach

Often, the best answer to achieving reuse is through a combination of framework-based and component-based techniques. A framework-based approach for building BusSim applications is appropriate for developing the user interface, handling user and system events, starting and stopping the application, and other application-specific and delivery platform-specific functions. A component-based approach is appropriate for black-box functionality. That is, functionality that can be used as-is with no specialization required.

In creating architectures to support BusSim application development, it is imperative that any assets remain as flexible and extensible as possible or reusability may be diminished. Therefore, we chose to implement the unique aspects of BusSim applications using a component approach rather than a framework approach. This decision is further supported by the following observations.

An Application Can Only Be Based on One Framework

Using the house analogy, if you like the first floor of one framework and the second floor of another, it is difficult or impossible to integrate the features of the two. Or, it is so costly as to erase the benefit of using a framework in the first place. Likewise with application frameworks. You can only use one framework when building an application. You can't mix and match features from multiple frameworks, so any framework that we developed would have to compete against existing and future frameworks. With components, however, you can mix and match from multiple vendors.

Components Are Less Platform and Development Tool Dependent, Leaving More Options Open for Development Teams

An appliance like a dishwasher is not restricted for use in a particular type of house. Similarly, component technologies exist that are independent of platform and development tool. For example ActiveX can be used in almost every development environment for Windows and Java Beans components can be used on a wide variety of platforms.

Frameworks Become Obsolete More Quickly

Rapid emergence and evolution of technology has introduced a wealth of new feature requirements into application development. Frameworks that do not include the most current features become obsolete quickly. Components typically address a more focused feature set and are not as impacted by technology advances outside their core functionality areas.

Based on these observations, we believe a combined framework/component approach is optimal for achieving reuse. FIG. 6 illustrates a development architecture approach in accordance with a preferred embodiment.

Delivery Framework for Business Simulation

This diagram illustrates an ideal solution where components are combined with an Application Framework and an Application Architecture to achieve maximum reuse and minimum custom development effort. The Application Architecture is added to provide communication support between the application interface and the components, and between the components. This solution has the following features:

The components (identified by the icons) encapsulate key BusSim functionality.

The Application Architecture provides the glue that allows application-to-component and component-to-component communication.

The Application Framework provides structure and base functionality that can be customized for different interaction styles.

Only the application interface must be custom developed.

The next section discusses each of these components in further detail.

The Business Simulation Toolset

We have clearly defined why a combined component/framework approach is the best solution for delivering high-quality BusSim solutions at a lower cost. Given that there are a number of third party frameworks already on the market that provide delivery capability for a wide variety of platforms, the TEL project is focused on defining and developing a set of components that provide unique services for the development and delivery of BusSim solutions. These components along with a set of design and test workbenches are the tools used by instructional designers to support activities in the four phases of BusSim development. We call this suite of tools the Business Simulation Toolset. Following is a description of each of the components and workbenches of the toolset.

Components

A Component can be thought of as a black box that encapsulates the behavior and data necessary to support a related set of services. It exposes these services to the outside world through published interfaces. The published interface of a component allows you to understand what it does through the services it offers, but not how it does it. The complexity of its implementation is hidden from the user. The following are the key components of the BusSim Toolset.

Domain Component--provides services for modeling the state of a simulation

Profiling Component--provides services for rule-based evaluating the state of a simulation

Transformation Component--provides services for manipulating the state of a simulation

Remediation Component--provides services for the rule-based delivering of feedback to the student

Domain Component

The Domain Model component is the central component of the suite that facilitates communication of context data across the application and the other components. It is a modeling tool that can use industry-standard database such as Informix, Oracle, or Sybase to store its data. A domain model is a representation of the objects in a simulation. The objects are such pseudo tangible things as a lever the student can pull, a form or notepad the student fills out, a character the student interacts with in a simulated meeting, etc. They can also be abstract objects such as the ROI for a particular investment, the number of times the student asked a particular question, etc. These objects are called entities. Some example entities include:

Vehicles, operators and incidents in an insurance domain

Journal entries, cash flow statements and balance sheets in a financial accounting domain

Consumers and purchases in a marketing domain

An entity can also contain other entities. For example, a personal bank account entity might contain an entity that represents a savings account. Every entity has a set of properties where each property in some way describes the entity. The set of properties owned by an entity, in essence, define the entity. Some example properties include:

An incident entity on an insurance application owns properties such as "Occurrence Date", "Incident Type Code", etc.

A journal entry owns properties such as "Credit Account", "Debit Account", and "Amount"

A revolving credit account entity on a mortgage application owns properties such as "Outstanding Balance", "Available Limit", etc. FIG. 7 illustrates a small segment of a domain model for claims handlers in the auto insurance industry in accordance with a preferred embodiment.

Example Domain Model

The domain model is created by the instructional designer in a visual editing design tool called the Knowledge Workbench. The designer creates the objects of the domain model using generic entities and properties; that is, not having specific values associated with the entities and properties.

At runtime, an application's domain model is instantiated so that every entity and property is given a particular value that makes it unique. The result of a domain model instantiation is called the domain. The values of a domain's entities and properties can change throughout the course of the simulation based on student actions and updates from other components. FIG. 8 illustrates an instantiated domain model in accordance with a preferred embodiment.

Example Domain

Creating a domain model in data rather than in code facilitates reuse of the components in multiple applications in multiple domains without code changes. For example, a typical application in the Financial Services domain would have to define classes in code such as `Customer`, `Account`, etc. An Insurance Domain application might have classes such as `Customer`, `Incident`, `Prior Policy`, etc. To be able to perform analysis on any of these classes, the analysis logic must be coded to recognize the classes. This requires all logic to be custom-coded for every application; an effort-intensive undertaking that demands a high degree of technical skill.

By modeling the domain in data using generic objects, we can build standard generic analysis capability that can be applied to the domain. This allows implementation of analysis logic with much less effort by people with a low degree of technical skill. Functional experts can create the objects of the domain and apply various types of analysis from a pallet. All of this is accomplished in a visual development environment that supports the designer with visual feedback and only allows valid designs to be created.

Profiling Component

In the simplest terms, the purpose of the Profiling Component is to analyze the current state of a domain and identify specific things that are true about that domain. This information is then passed to the Remediation Component which provides feedback to the student. The Profiling Component analyzes the domain by asking questions about the domain's state, akin to an investigator asking questions about a case. The questions that the Profiler asks are called profiles. For example, suppose there is a task about building a campfire and the student has just thrown a match on a pile of wood, but the fire didn't start. In order to give useful feedback to the student, a tutor would need to know things like: was the match lit?, was the wood wet?, was there kindling in the pile?, etc. These questions would be among the profiles that the Profiling Component would use to analyze the domain. The results of the analysis would then be passed off to the Remediation Component which would use this information to provide specific feedback to the student.

Specifically, a profile is a set of criteria that is matched against the domain. The purpose of a profile is to check whether the criteria defined by the profile is met in the domain. Using a visual editing tool, instructional designers create profiles to identify those things that are important to know about the domain for a given task. During execution of a BusSim application at the point that feedback is requested either by the student or pro-actively by the application, the set of profiles associated with the current task are evaluated to determine which ones are true. Example profiles include:

Good productions strategy but wrong Break-Even Formula

Good driving record and low claims history

Correct Cash Flow Analysis but poor Return on Investment (ROI)

A profile is composed of two types of structures: characteristics and collective characteristics. A characteristic is a conditional (the if half of a rule) that identifies a subset of the domain that is important for determining what feedback to deliver to the student. Example characteristics include:

Wrong debit account in transaction 1

Perfect cost classification

At Least 1 DUI in the last 3 years

More than $4000 in claims in the last 2 years

More than two at-fault accidents in 5 years

A characteristic's conditional uses one or more atomics as the operands to identify the subset of the domain that defines the characteristic. An atomic only makes reference to a single property of a single entity in the domain; thus the term atomic. Example atomics include:

The number of DUI's>=1

ROI>10%

Income between $75,000 and $110,000

A collective characteristic is a conditional that uses multiple characteristics and/or other collective characteristics as its operands. Collective characteristics allow instructional designers to build richer expressions (i.e., ask more complex questions). Example collective characteristics include:

Bad Household driving record

Good Credit Rating

Marginal Credit Rating

Problems with Cash for Expense transactions

Problems with Sources and uses of cash

Once created, designers are able to reuse these elements within multiple expressions, which significantly eases the burden of creating additional profiles. When building a profile from its elements, atomics can be used by multiple characteristics, characteristics can be used by multiple collective characteristics and profiles, and collective characteristics can be used by multiple collective characteristics and profiles.

FIG. 9 illustrates an insurance underwriting profile in accordance with a preferred embodiment.

Example Profile for Insurance Underwriting

Transformation Component

Whereas the Profiling Component asks questions about the domain, the Transformation Component performs calculations on the domain and feeds the results back into the domain for further analysis by the Profiling Component. This facilitates the modeling of complex business systems that would otherwise be very difficult to implement as part of the application. Within the Analysis phase of the Interface/Analysis/Interpretation execution flow, the Transformation Component actually acts on the domain before the Profiling Component does its analysis. The Transformation Component acts as a shell that wraps one or more data modeling components for the purpose of integrating these components into a BusSim application. The Transformation Component facilitates the transfer of specific data from the domain to the data modeling component (inputs) for calculations to be performed on the data, as well as the transfer of the results of the calculations from the data modeling component back to the domain (outputs). FIG. 10 illustrates a transformation component in accordance with a preferred embodiment.

The data modeling components could be third party modeling environments such as spreadsheet-based modeling (e.g., Excel, Formula1) or discrete time-based simulation modeling (e.g., PowerSim, VenSim). The components could also be custom built in C++, VB, Access, or any tool that is ODBC compliant to provide unique modeling environments. Using the Transformation Component to wrap a third party spreadsheet component provides an easy way of integrating into an application spreadsheet-based data analysis, created by such tools as Excel. The Transformation Component provides a shell for the spreadsheet so that it can look into the domain, pull out values needed as inputs, performs its calculations, and post outputs back to the domain.

For example, if the financial statements of a company are stored in the domain, the domain would hold the baseline data like how much cash the company has, what its assets and liabilities are, etc. The Transformation Component would be able to look at the data and calculate additional values like cash flow ratios, ROI or NPV of investments, or any other calculations to quantitatively analyze the financial health of the company. Depending on their complexity, these calculations could be performed by pre-existing spreadsheets that a client has already spent considerable time developing.

Remediation Component

The Remediation Component is an expert system that facilitates integration of intelligent feedback into BusSim applications. It has the following features:

Ability to compose high quality text feedback.

Ability to compose multimedia feedback that includes video and/or audio.

Ability to include reference material in feedback such as Authorware pages or Web Pages.

Ability to actively manipulate the users deliverables to highlight or even fix users' errors.

A proven remediation theory embedded in its feedback composition algorithm.

Allows integration of digital assets into the Remediation of a training or IPS application.

The Remediation model consists of three primary objects:

Concepts

Coach Topics

Coach Items

Concepts are objects that represent real-world concepts that the user will be faced with in the interface. Concepts can be broken into sub-concepts, creating a hierarchical tree of concepts.

This tree can be arbitrarily deep and wide to support rich concept modeling. Concepts can also own an arbitrary number of Coach Topics.

Coach Topics are objects that represent a discussion topic that may be appropriate for a concept. Coach Topics can own an arbitrary number of Coach Items.

Coach Items are items of feedback that may include text, audio, video, URL's, or updates to the Domain Model. Coach Items are owned by Coach Topics and are assembled by the Remediation Component algorithm.

Details of the Remediation Component algorithm for feedback is discussed in The Intelligent Coaching Agent Tool whitepaper and can be found on the Knowledge Exchange at the Technology Enabled Learning ETA Home Page.

Workbenches

The BusSim Toolset also includes a set of workbenches that are used by instructional designers to design and build BusSim applications. A workbench is a tool that facilitates visual editing or testing of the data that the BusSim Components use for determining an application's run-time behavior. The BusSim Toolset includes the following workbenches:

Knowledge Workbench

The Knowledge Workbench is a tool for the creation of domain, analysis and feedback data that is used by the BusSim Components. It has the following features:

Allows the designer to `paint` knowledge in a drag-and-drop interface.

Knowledge is represented visually for easy communication among designers.

The interface is intelligent, allowing designers to only paint valid interactions.

Designer's Task creations are stored in a central repository.

The workbench supports check-in/check-out for exclusive editing of a task.

Supports LAN-based or untethered editing.

Automatically generates documentation of the designs.

Generates the data files that drive the behavior of the components.

Simulated Student Test Workbench

The Simulated Student Test Workbench is a tool for the creation of data that simulates student's actions for testing BusSim Component behaviors. It has the following features:

The Test Bench generates a simulated application interface based on the Domain Model.

The designer manipulates the objects in the Domain Model to simulate student activity.

The designer can invoke the components to experience the interactions the student will experience in production.

The designer can fully test the interaction behavior prior to development of the application interface.

Regression Test Workbench

The Regression Test Workbench is a tool for replaying and testing of student sessions to aid debugging. It has the following features:

Each student submission can be individually replayed through the components.

An arbitrary number of student submissions from the same session can be replayed in succession.

Entire student sessions can be replayed in batch instantly.

The interaction results of the student are juxtaposed with the results of the regression test for comparison.

Development Cycle Activities

Design Phase

The design phase of a BusSim application is streamlined by the use of the Knowledge Workbench. The Knowledge Workbench is a visual editor for configuring the objects of the component engines to control their runtime behavior. The components are based on proven algorithms that capture and implement best practices and provide a conceptual framework and methodology for instructional design.

In conceptual design, the workbench allows the designer to paint a model of the hierarchy of Concepts that the student will need to master in the activity. This helps the designer organize the content in a logical way. The visual representation of the Concepts helps to communicate ideas to other designers for review. The consistent look and feel of the workbench also contributes to a streamlined Quality Assurance process. In addition, standard documentation can be automatically generated for the entire design.

As the design phase progresses, the designer adds more detail to the design of the Concept hierarchy by painting in Coach Topics that the student may need feedback on. The designer can associate multiple feedback topics with each Concept. The designer also characterizes each topic as being Praise, Polish, Focus, Redirect or one of several other types of feedback that are consistent with a proven remediation methodology. The designer can then fill each topic with text, video war stories, Web page links, Authorware links, or any other media object that can be delivered to the student as part of the feedback topic.

As the designer's thoughts for the interface become clearer, she can begin to model the domain objects in the Knowledge Workbench. The student's world is constructed using objects in the Domain Model.

The designer again uses the Knowledge Workbench to configure objects in the Transformation Component. The Transformation Component is used to perform calculations or other analysis of the student's domain. Lastly, the designer uses the workbench to configure objects in the Profiling Component. The Profiling Component examines the student's domain, looking for conditions that indicate what feedback topics are appropriate for delivery. More importantly, the Student Simulator Test Workbench allows the designer to exercise the designs. It allows the designer to manipulate the domain as if she were a student. The designer can interact with the simulated interface and invoke the component engines to see the feedback that the student would receive. This capability can also be utilized in a usability test such as a Conference Room Pilot. As the test student interacts with screen mock-ups, a facilitator can mimic his actions in the interface simulator and tell the student what the actual feedback will be. This results in much more rigorous testing prior to application construction. A big payoff is realized downstream in the form of reduced redesign after usability and cognition testing.

Throughout all these steps in the initial design, the workbench supports the design process by allowing the designer great flexibility within the framework of a proven methodology. This allows experienced users to design rich, realistic interactions, and inexperienced users to become competent in a shorter time by learning from the best practices embedded in the workbench. This greatly diminishes the `blue sky` design problem. Also, since the designs can be tested prior to application construction, there is reduced rework after testing. Lastly, the visual knowledge representation enhances communication within the design team and greatly streamlines the QA process.

Build Phase

It is very clear how the tools support the Build Phase. The designs that the designer painted in the Knowledge Workbench drive the components at runtime. The application developer no longer has to write code that analyzes the student's work and provides feedback. The developer only has to build the interface and logic to report any student actions to the domain model. The components do the rest. What used to be the most difficult part of the build phase has been eliminated!

There is no chance for a developer to misinterpret the feedback designs because she never has to interpret them. In fact, the developer doesn't even have to know anything about the feedback behavior as long as she is familiar with the domain model. This also means the skill level required to develop the application can be greatly reduced. It's not hard to teach someone how to paint a screen in Visual Basic or Delphi and call API functions to notify the Domain Model of student actions.

In addition to the economies gained by the components, it is possible to use templates to further streamline design and development of commonly used interactions. We have created templates for several common interactions. For example, Journalizing of Transactions is an interaction that has appeared in several applications. We have built application and Knowledge Workbench templates for Journalization. All one must do to create a new Journalize task is to add graphics for new Transactions and fill in new data into placeholders in the Knowledge Workbench.

Test Phase

The toolset greatly reduces effort during functionality testing. The key driver of the effort reduction is that the components can automatically track the actions of the tester without the need to add code support in the application. Whenever the tester takes an action in the interface, it is reported to the domain model. From there it can be tracked in a database. Testers no longer need to write down their actions for use in debugging; they are automatically written to disk. There is also a feature for attaching comments to a tester's actions. When unexpected behavior is encountered, the tester can hit a control key sequence that pops up a dialog to record a description of the errant behavior.

Of far greater impact is the ability to replay the tester's actions automatically through the Regression Test Workbench. The designer does not need to spend hours trying to duplicate the error. She simply loads the tester's session into the Regression Test Workbench and replays it. In seconds the error is replicated and can be located and fixed using a variety of debugging utilities. After changes have been made, one more pass through the Regression Test Workbench verifies the fix.

The major difficulties of usability and cognition testing are also addressed by the toolset. First, since student tracking is no longer a manual activity, the precision of functional testing can also be applied to usability and cognition testing. Second, because of the increased rigor in the Conference Room Pilot, the risk of significant rework is greatly reduced.

Execution Phase

During the Execution Phase, the components are deployed to the student's platform. They provide simulated team member and feedback functionality with sub-second response time and error-free operation. If the client desires it, student tracking mechanisms can be deployed at runtime for evaluation and administration of students. This also enables the isolation of any defects that may have made it to production.

Scenarios for Using the Business Simulation Toolset

A good way to gain a better appreciation for how the BusSim Toolset can vastly improve the BusSim development effort is to walk through scenarios of how the tools would be used throughout the development lifecycle of a particular task in a BusSim application. For this purpose, we'll assume that the goal of the student in a specific task is to journalize invoice transactions, and that this task is within the broader context of learning the fundamentals of financial accounting. A cursory description of the task from the student's perspective will help set the context for the scenarios. Following the description are five scenarios which describe various activities in the development of this task. The figure below shows a screen shot of the task interface.

FIG. 11 illustrates the use of a toolbar to navigate and access application level features in accordance with a preferred embodiment. A student uses a toolbar to navigate and also to access some of the application-level features of the application. The toolbar is the inverted L-shaped object across the top and left of the interface. The top section of the toolbar allows the user to navigate to tasks within the current activity. The left section of the toolbar allows the student to access other features of the application, including feedback. The student can have his deliverables analyzed and receive feedback by clicking on the Team button. In this task, the student must journalize twenty-two invoices and other source documents to record the flow of budget dollars between internal accounts. (Note: "Journalizing", or "Journalization", is the process of recording journal entries in a general ledger from invoices or other source documents during an accounting period. The process entails creating debit and balancing credit entries for each document. At the completion of this process, the general ledger records are used to create a trial balance and subsequent financial reports.)

In accordance with a preferred embodiment, an Intelligent Coaching Agent Tool (ICAT) was developed to standardize and simplify the creation and delivery of feedback in a highly complex and open-ended environment. Feedback from a coach or tutor is instrumental in guiding the learner through an application. Moreover, by diagnosing trouble areas and recommending specific actions based on predicted student understanding of the domain student comprehension of key concepts is increased. By writing rules and feedback that correspond to a proven feedback strategy, consistent feedback is delivered throughout the application, regardless of the interaction type or of the specific designer/developer creating the feedback. The ICAT is packaged with a user-friendly workbench, so that it may be reused to increase productivity on projects requiring a similar rule-based data engine and repository.

Definition of ICAT In Accordance with a Preferred Embodiment

The Intelligent Coaching Agent Tool (ICAT) is a suite of tools--a database and a Dynamic Link Library (DLL) run-time engine--used by designers to create and execute just-in-time feedback of Goal Based training. Designers write feedback and rules in the development tools. Once the feedback is set, the run-time engine monitors user actions, fires rules and composes feedback which describes the business deliverable.

I. The ICAT Remediation Model

The remediation model used within ICAT dynamically composes the most appropriate feedback to deliver to a student based on student's previous responses. The ICAT model is based on a theory of feedback which has been proven effective by pilot results and informal interviews. The model is embodied in the object model and algorithms of the ICAT. Because the model is built into the tools, all feedback created with the tool will conform to the model.

II. The Role of ICAT in Student Training

The ICAT plays two roles in student training. First, the ICAT is a teaching system, helping students to fully comprehend and apply information. Second, ICAT is a gatekeeper, ensuring that each student has mastered the material before moving on to additional information.

III. The Functional Definition of the ICAT

The ICAT is a self contained module, separate from the application. Separating the ICAT from the application allows other projects to use the ICAT and allows designers to test feedback before the application is complete. The ICAT Module is built on six processes which allow a student to interact effectively with the interface to compose and deliver the appropriate feedback for a student's mistakes.

IV. The ICAT Development Methodology for Creating Feedback

The ICAT development methodology is a seven step methodology for creating feedback. The methodology contains specific steps, general guidelines and lessons learned from the field. Using the methodology increases the effectiveness of the feedback to meet the educational requirements of the course.

V. Components

The processes each contain a knowledge model and some contain algorithms. Each process has specific knowledge architected into its design to enhance remediation and teaching.

VI. Testing Utilities, Reports and Methodology

There is a suite of testing tools for the ICAT. These tools allow designers and developers test all of their feedback and rules. In addition, the utilities let designers capture real time activities of students as they go through the course.

Expert Remediation Model within the Tools

The tools and run-time engine in accordance with a preferred embodiment include expert knowledge of remediation. These objects include logic that analyzes a student's work to identify problem areas and deliver focused feedback. The designers need only instantiate the objects to put the tools to work. Embodying expert knowledge in the tools and engine ensures that each section of a course has the same effective feedback structure in place.

Any project which is creating a Goal-Based Scenario (GBS) business simulation or an Integrated Performance Support (IPS) system to help users understand and create business deliverables can profit from technology in accordance with a preferred embodiment. A GBS allows students to learn in a comprehensive simulated environment. Students work in a simulated environment to accomplish real world tasks, and when they make mistakes, remediation is provided to help identify and correct the mistakes. The hands-on experience of the simulated environment and the timely remediation account for the high retention rate from subjects presented utilizing a system in accordance with a preferred embodiment. A system in accordance with a preferred embodiment can be used in conjunction with an IPS to help users develop deliverables. If a customer service representative (CSR) is completing a form while conducting a phone conversation, the ICAT can be used to observe how the task is completed to provide a live analysis of mistakes.

A file structure in accordance with a preferred embodiment provides a standard system environment for all applications in accordance with a preferred embodiment. A development directory holds a plurality of sub-directories. The content in the documentation directory is part of a separate installation from the architecture. This is due to the size of the documentation directory. It does not require any support files, thus it may be placed on a LAN or on individual computers.

When the architecture is installed in accordance with a preferred embodiment, the development directory has an _Arch, _Tools, _Utilities, Documentation, QED, and XDefault development directory. Each folder has its own directory structure that is inter-linked with the other directories. This structure must be maintained to assure consistency and compatibility between projects to clarify project differences, and architecture updates.

The _Arch directory stores many of the most common parts of the system architecture. These files generally do not change and can be reused in any area of the project. If there is common visual basic code for applications that will continuously be used in other applications, the files will be housed in a folder in this directory.

The sub-directories in the _Arch directory are broken into certain objects of the main project. Object in this case refers to parts of a project that are commonly referred to within the project. For example, modules and classes are defined here, and the directory is analogous to a library of functions, APIs, etc . . . that do not change. For example the IcaObj directory stores code for the Intelligent Coaching Agent (ICA). The InBoxObj directory stores code for the InBox part of the project and so on. The file structure uses some primary object references as file directories. For example, the IcaObj directory is a component that contains primary objects for the ICA such as functional forms, modules and classes.

The BrowserObj directory contains modules, classes and forms related to the browser functionality in the architecture.

The HTMLGlossary directory contains code that is used for the HTML reference and glossary component of the architecture.

The IcaObj directory contains ICA functional code to be used in an application. This code is instantiated and enhanced in accordance with a preferred embodiment.

The InBoxObj directory contains code pertaining to the inbox functionality used within the architecture. Specifically, there are two major components in this architecture directory. There is a new .ocx control that was created to provide functionality for an inbox in the application. There is also code that provides support for a legacy inbox application. The PracticeObj directory contains code for the topics component of the architecture. The topics component can be implemented with the HTMLGlossary component as well.

The QmediaObj directory contains the components that are media related. An example is the QVIDctrl.cls. The QVIDctrl is the code that creates the links between QVID files in an application and the system in accordance with a preferred embodiment.

The SimObj directory contains the Simulation Engine, a component of the application that notifies the tutor of inputs and outputs using a spreadsheet to facilitate communication.

The StaticObj directory holds any component that the application will use statically from the rest of the application. For example, the login form is kept in this folder and is used as a static object in accordance with a preferred embodiment.

The SysDynObj directory contains the code that allows the Systems Dynamics Engine (Powersim) to pass values to the Simulation Engine and return the values to the tutor.

The VBObj directory contains common Visual Basic objects used in applications. For example the NowWhat, Visual Basic Reference forms, and specific message box components are stored in this folder.

The _Tools directory contains two main directories. They represent the two most used tools in accordance with a preferred embodiment. The two directories provide the code for the tools themselves. The reason for providing the code for these tools is to allow a developer to enhance certain parts of the tools to extend their ability. This is important for the current project development and also for the growth of the tools.

The Icautils directory contains a data, database, default, graphics, icadoc, and testdata directory. The purpose of all of these directories is to provide a secondary working directory for a developer to keep their testing environment of enhanced Icautils applications separate from the project application. It is built as a testbed for the tool only. No application specific work should be done here. The purpose of each of these directories will be explained in more depth in the project directory section. The TestData folder is unique to the _Tools/ICAUtils directory. It contains test data for the regression bench among others components in ICAUtils.

Utilities

The Utilities directory holds the available utilities that a Business Simulation project requires for optimal results. This is a repository for code and executable utilities that developers and designers may utilize and enhance in accordance with a preferred embodiment. Most of the utilities are small applications or tools that can be used in the production of simulations which comprise an executable and code to go with it for any enhancements or changes to the utility. If new utilities are created on a project or existing utilities are enhanced, it is important to notify the managers or developers in charge of keeping track of the Business Simulation assets. Any enhancements, changes or additions to the Business Simulation technology assets are important for future and existing projects.

Documentation

A Documentation directory is used to store pertinent documentation. The documentation directory is structured as follows. Most of the directories are labeled after the specific information held within them. The following is a list of all the documentation directories and a description of what is contain in each.

Ref Website--This directory contains The Business Simulation Reference website, which is a general reference for many things. If the website has not been set up for users on a LAN or website, all you need to do is go into the root directory of website and double click on index.htm. This is the main page for the site.

Components--This directory contains any documentation on classes and modules that are used in the archtecture. For example there are documents here on the ICAMeeting class, the Inbox class etc.

Database--This directory contains any documents describing the databases that are included and used in the Architecture. For example the ICAObj overview doc contains a description of the model and each element in the database.

HTML Component--This directory contains relevant documentation about the HTML part of the architecture.

Process Models--This directory should contain the documents that describe the process of the application or related information.

ReferenceApp--This directory contains documents with descriptions and views of the reference app. (QED) for explanation and documentation. Testing conditions are stored in the Testing directory.

Standards&Templates--This directory contains any type of architecture relevant coding standard documents or templates that a developer is required to follow.

UserGuides--This directory has 6 sub-directories. Each one of these sub-directories contains user guides for a given tool or component in accordance with a preferred embodiment which include user guides for the architecture, the Tutor Suite, ICA Utilities, the simulation Engine and the System Dynamics Engine. There is also a directory for other documentation that contains user guides for any other tools or code like third party controls etc.

WorkFlows--This directory contains the WF_Develop.doc which includes the workflow documentation for an application.

Project Directory

The sample project directory, QED has the same structure that a real project would be designed after. The QED directory has all custom architecture code, databases, spreadsheets, and any other application specific files stored in it. The QED project directory stores a Design and SrcVB directory. The Design directory contains all relevant files for a designer. The SrcVB directory is used for a developer.

The root directory of the Design and SrcVB directory contain a few important files to note. Both have two .rtf files, a few log files and an .ini file. The .rtf files are the feedback that is output from the tutor, the logs are also output from the tutor and the .ini file is for ICAUtils initialization. The design directory has three subdirectories that contain a data directory, which stores .xls files, sim models, and any other important data like html and video. It also has a database directory that holds any relevant databases for development and application use. The last directory is the icadoc directory which includes all .tut files or .ica files, which are both created with the tutor.

The SrcVB directory stores all of the directories previously described. The reason for duplicating the data and database directories is to assure that a developer does not interfere with the designer's files. The developer tends to not do as much design work and can easily corrupt files. This duplication of directories provides a safer environment for the developer to test in. As was mentioned above, the developer tends to have a lot more to do with the application build than the design so there needs to be more content in the SrcVB directory. The SrcVB directory also contains an .exe and .vbp file which are created in a developers visual basic application.

The following are directories found in the SrcVB directory that are not found in the Design directory followed by a short definition:

The _CustomArch directory contains any application specific architecture. Look in the QED folder for an example.

The _CustomDistribution directory contains any files that need to be distributed with the application.

The Default directory contains any backup files that might need to be copied and reused later. Some files occasionally are corrupted and need to be replaced.

The Fonts directory contains application specific font libraries.

The Graphics directory contains any relevant graphics for the application.

The Help directory contains all files for a help reference layer in the application. This can be implemented in many ways but most commonly in an HTML form.

The Saved directory is for saved information that is produced by the application. This can be used for saving student information or saving temporary information for the application steps.

The StudentData directory is for storing any relevant student data, lists of students, their personal information or any relevant student data that needs to be saved.

XDefault Development

The XDefault Development environment is used to provide a shell for any new project. A developer would rename this directory as an acronym of the project. QED is the default for the installation sample application. The XDefault development directory is a shell and serves as a building block for a startup project. A good idea is to use the QED sample application and build the XDefault Development project with the sample code in QED.

Shared Development

The last directory to be mentioned is the shared development directory which is placed on a LAN or central network area and is shared by all designers and developers of a project to assure that files in the project are up to date, managed properly and appropriately synchronized. There are many databases and files that will be shared in accordance with a preferred embodiment. These files need to be shared and have a location that someone can edit without having to worry about merging files later. A source control program is used to restrict access to a file to one application at a time.

The ICAT Model of Remediation

The ICAT has a model of remediation architected into the system in accordance with a preferred embodiment. Feedback should help students complete tasks and learn the underlying concepts. To achieve this goal, the ICAT reviews student's work with the following objectives in mind.

Identify Student Misconceptions

Identifying that a student does not understand a topic and then clearly explaining it is the goal of human and computer tutors alike. Human tutors, however, have many more clues--including facial expressions and body language--to help them identify student misconceptions. The computer tutor is much more limited and can only view the outputs--such as documents and reports--the student produces. If a computer tutor is looking for a misunderstanding about debits and credits, the computer analyzes all the mistakes a student made concerning debits and credits and tries to identify what misunderstanding would account for this pattern of mistakes.

Identify what Students Should Fix

If the coach cannot diagnose a student's misconception, or cannot do it with 100% accuracy, the coach must at least tell the student what he did wrong so that he can correct it. If at all possible, the coach should identify groups or types of problems the student is making so that the student can generalize the solution and answer.

Prompt Students to Reflect on Mistakes

When identifying problems, the tutor needs to prompt the student to reflect on a problem and start to point the student towards the answer. The tutor should not tell the student the answer, but instead should attempt to provide an appropriate answer or give the student a question to think about.

Reinforce Correct Concepts and Ideas

Once a student has gotten the correct answer, it is important to reinforce the learning. Students may feel uncertain about their understanding even after he has gotten the answer correct. To reinforce the student's understanding of the concept and provide a confidence boost, the tutor should walk the student through the answer so that it is completely understood. These goals are not just the goals of a computer tutor, but they are the goals of a human tutor as well. All tutors must look at a student's work to help identify and correct errors as well as learn the material. One of the most difficult tasks facing a tutor is the difficult task of balancing the appropriate amount of assistance provided the student to complete the task with the requirement to help the student learn the material.

Model of Feedback

A preferred embodiment utilizes feedback to address the balancing task. The theory is centered on the idea of severity. Severe errors require severe feedback while mild errors require mild feedback. If a student writes a paper on the wrong subject, a human tutor will spend little time reviewing the paper, but instead, identify it as a serious mistake and ask the student to rewrite the paper. If the student simply misses one paragraph of the argument, then the tutor will focus the student on that paragraph. Finally, if the paper is correct except for a couple of spelling mistakes, the tutor will point out the specific mistakes and ask the student to correct them. The point is that because a tutor and a student do not want to waste each others' time, they will match the severity of the error with the severity of the feedback.

In the ICAT model of feedback, there are four levels of severity of error and four corresponding levels of feedback. The tutor goes through the student's work, identifies the severity of the error and then provides the corresponding level of feedback.

                Educational Categories of Feedback
             ERROR                        FEEDBACK
                              Feedback
    Error Type  Description   Type      Description
    1. None     No errors exist. 1. Praise Confirmation that the student
                The student's           completed the task correctly.
                work is perfect.           Example:
                                        Great. You have journalized
                                        all accounts correctly. I am
                                        happy to see you recognized
                                        we are paying for most of our
                                        bills "on account".
    2. Syntactic There may be  2. Polish Tells the student the specific
                spelling                actions he did incorrectly,
                mistakes or             and possibly correct them for
                other syntactic           him.
                errors. As a            Example:
                designer, you           There are one or two errors
                should be               in your work. It looks like
                confident that           you misclassified the
                the student will           purchase of the fax as a cash
                have mastered           purchase when it is really a
                the material at           purchase on account.
                this point.
    3. Local    A paragraph of 3. Focus  Focus the student on this area
                a paper is              of his work. Point out that
                missing or the           he does not understand at
                student has             least one major concept.
                made a number           Example:
                of mistakes all           Looking over your work, I
                in one area.            see that you do not
                The student             understand the concept of
                clearly does not           "on account". Why don't
                understand this           you review that concept and
                area.                   review your work for errors.
    4. Global   The student has 4. Re-    Restate the goal of the
                written on the direct    activity and tell the student to
                wrong subject           review main concepts and
                or there are            retry the activity.
                mistakes all            Example:
                over the                There are lots of mistakes
                student's work           throughout your work. You
                which indicates           need to think about what type
                he does not             of transaction each source
                understand              document represents before
                most of the             journalizing it.
                concepts in the
                activity.

              Administrative Categories of Feedback
                                Feed-
    Error     Description       back  Description
    No work   The student has   Mas-  Tell the student that he has
    done since made no changes   ter-  done no work and that a
    last review since the last time mind  substantial amount of work
              he asked for the tu-       needs to be completed before
              tor to review his       review.
              work.                   Example:
                                      You have done no work since I
                                      last reviewed your work.
                                      Please try and correct at least
                                      three journal entries before
                                      asking me to review your work
                                      again.
    All work  If a designer wants In-   State that the student has not
    is not    to give an interim com-  completed all of the work
    complete  report of how the plete- required, but you will review
    but a     student is doing be- con-  what the student has done so
    substantial fore everything is tinue far.
    amount of done, they he would       Example:
    work has  use incomplete--        It looks like you have not
    been done continue.               finished journalizing, but I will
                                      review what you have done up
                                      to this point. The first three
                                      entries are correct.
    All