Prototyping

Contents 1 Applications 2 Procedures 3 Instructions 4 References

A technique used to experiment with new ideas or concepts at relatively low cost. Prototyping is the process of creating a simple and experimental model of a digital product to visualize the concept and get feedback from teammates and stakeholders. Prototyping is used to model and create a functional form of a newly redesigned or reengineered component of a value stream. It is utilized to test alternative approaches and to validate that they represent real improvements. The greatest value is to study changes that could be very beneficial if implemented correctly, but could be extremely disruptive to the enterprise if implemented incorrectly.

Prototyping is related to, but should not be confused with Simulation. Prototyping, in the context of business reengineering, is used to physically test components of a value stream in a laboratory environment or control group setting. Simulation is used to model a complete value stream and to test the effects of changes in variables associated with the value stream. Prototyping often follows simulation to further validate the value of process changes.

Prototyping, in the context of information systems development, involves the development of a component of an automated system (e.g. a horizontal prototype is paper-based; a vertical prototype is an automated version). Generally, prototypes are used to demonstrate a model of the human interface (i.e., screen design), although they may be developed for other uses, as well such as value engineering and marketing purposes.

Applications

• To determine how a solution will function before it is put in place in a department or enterprise-wide.
• To determine the effect of change and verify the correctness and efficiency of a redesigned component of a value stream.
• To capture information about functional requirements to enable more robust and complete solutions.

Procedures

• Choose a component of the reengineered value stream to prototype, avoiding elaborate or complex requirements.
• Prototype manual processes initially. A second level of prototype may follow in which a computerized prototype is developed, if warranted.
• Define the prototype "ground rules," including the duration and the criteria and/or metrics/statistics/measurement categories by which its performance will be judged, as well as the method and frequency of metrics/statistics/measurement collection.
• Establish a control group of selected individuals to participate in the prototype activities, or conduct the prototype in a laboratory environment.
• Develop interim procedures and train the control group in the process.
• Execute the prototype, collecting the required metrics/statistics/measurements at the appropriate intervals.
• Monitor the prototype process to ensure adherence and to analyze interim results.
• At the completion of the prototype period, analyze and evaluate the performance against current activity performance to determine the results. Feed the results back into the process to further refine or change it.

Instructions

Use prototyping techniques to physically test a component of the newly engineered or redesigned value stream. Design and create procedures representing the candidate process or activity. Avoid elaborate requirements, ensuring that the prototype is practical and executable in the environment in which it is being conducted. Elaborate prototypes can become disruptive and costly, and there is a diminishing return of value to the organization. Poorly designed prototypes result in a waste of corporate resources and, potentially, poorly designed processes.

A quick and effective method of prototyping is to create a laboratory environment for testing key activities. The laboratory performance can be compared to current activity performance. This is sometimes referred to as the Design by experiment technique. Design by Experiment can be particularly effective in determining social system requirements where qualitative impacts must be assessed. Select a control group to conduct the prototype. Occasionally, external customers and suppliers may need to be included in the control group or in the laboratory environment. For example, a new travel order activity may require that an external travel agent participate. Ensure that the role and responsibilities of all participants are clearly defined and that they understand the objectives, particularly in the case of external customers and suppliers.

Define the prototype "ground rules," including the duration and the criteria and/or metrics/statistics/measurement categories by which its performance will be judged, as well as the methods and frequency for collecting metrics/statistics/measurements. Develop interim procedures, as required, and train the participants in the process, including the ground rules.

Execute the prototype, collecting the appropriate metrics, statistics and measurements, as required. Monitor the process to ensure adherence to the new process and to analyze interim results. It is critical that the information gathered and intelligence gained through the prototype be fed back into the process design to fine tune the process. Analyze the final results to determine the strengths and weaknesses and where the process needs further refinement or change.

If the value stream activity includes a technology solution or if new or enhanced automation is recommended, the prototype may be carried to a second level. In this level, a computerized prototype is designed and developed which supports the business process. System prototypes are classified as horizontal prototypes and vertical prototypes. Horizontal prototypes are generally paper-based and demonstrate the human interface (screen design) only. Users of the value stream participate in the process of requirements definition and system design and are able to anticipate the look and feel of a future system based upon the screen layouts. Vertical prototypes are computerized models, again generally demonstrating screen design, although they can incorporate some greater depth of functionality. They provide a very real sense of the future system, its look and feel, and the flow of information and processing through the system. System prototypes should be developed using advanced prototyping or Rapid Application Development (RAD) tools to expedite the development cycle.

References

1. Walter J. Utz. Software Technology Transitions. Prentice-Hall, Inc. 1992.
2. William E. Perry. Quality Assurance For Information Systems. QED Technical Publishing Group, 1991.