Skip Navigation
acfbanner  
ACF
Department of Health and Human Services Administration for Children and Families
          
ACF Home   |   Services   |   Working with ACF   |   Policy/Planning   |   About ACF   |   ACF News   |   HHS Home

  Questions?  |  Privacy  |  Site Index  |  Contact Us  |  Download Reader™Download Reader  |  Print Print      

National Human Services IT Resource Center

Checklist: Architectural Study

Synopsis

Checklist: Planning and Architectural Study
Checklist: Reviewing Study Results

Synopsis

An architectural study must provide reliable information upon which to make critical design decisions - as quickly as possible. These decisions carry considerable risk; therefore, the studies must be carefully planned and executed. The following extensible checklist can guide the Study Lead in planning the study and ensuring that it will provide a sound basis for decision making.

Planning an Architectural Study

Checkbox with check A lightweight management approach such as time-boxing ( McConnell 1996 ) is used as a basis for defining and organizing the study approach.
Checkbox with check The decision that the study is intended to support is described (one paragraph or less). This helps keep the study on track by helping determine use of the study results.
Checkbox with check The objective of the study is clearly stated, such as:
  • What is expected to be learned (to know something)
  • What is expected to be demonstrated (to show something)
  • Generating data to support a cost/technical risk analysis of one or more alternatives (develop alternatives).
Checkbox with check Studies should be scoped to address a single, well-defined objective. When multiple objectives are established, they may represent too broad a scope and possibly compromise the study. Objectives should not compete with one another (e.g., determine the fastest and cheapest method for doing something).
Checkbox with check The value of the information returned should be worth the study expense. This helps reduce studies to their essential objectives and determine the best study approach. Speed and accuracy is more important than precision in the results. Studies reduce risk, not eliminate it.
Checkbox with check The study approach is clearly articulated, such as:
  • Vendor survey
  • Literature search
  • Demonstration
  • Functional prototype
  • Analysis (modeling)
  • Simulation
  • Case study (review of existing implementations, subjects)
  • Benchmark characteristics of a (current) system
  • To eliminate options (rather that to select one)
  • Evaluate vendor products against standards
  • Evaluate adaptability
Checkbox with check The format and level of detail (summary/detailed) of the study results is clearly defined, such as:
  • Technical report (formal)
  • Engineering note (informal)
  • Prototype
  • Presentation
Checkbox with check Estimates for time, cost, and technical resources are written down and realistic. The basis for the estimate is provided.
  • Check against actual results of other studies that were performed
  • Sufficient time and materials are provided, such as computers or software tools
Checkbox with check Appropriate expertise is assigned to the study
  • Full time for the duration of the study.
  • Skills should be commensurate with the study outcome. Minimal learning, other than in the study area, should be needed (e.g., fluency in the programming language).
  • Agreements should be in place, as needed, when the study team is staffed with individuals outside the architecture group (i.e., those over which the Chief Architect has direct control). This ensures no contention for resources on tight study schedules, such as using programmers from the IT projects to implement a prototype.
  • The Study Team should have minimum distractions once they begin their work. Limit interruptions.
  • Consider just-in-time training, when needed.
  • Someone to support the Core Study Team should be available to remove any administrative overhead or handle interruptions, if needed. The Architecture Team Facilitator should arrange for this.
Checkbox with check Studies are experimental by nature. Criteria should be established to indicate when a study should be abandoned early. This keeps studies of little value from dragging on.
Checkbox with check Studies, once initiated, should be on the order of 1 week to no more than a calendar month. Planning should take only a few hours, although the need to coordinate with all study participants may require a calendar week or more (e.g., to get commitment from external stakeholders such as users if an interactive study is being performed). This keeps the time between establishing an information need and providing a study result short.
Checkbox with check Resources to conduct a studies should be readily available, such as access to computers, software, or facilities. Assess during non-prime time periods may be necessary, such as running performance studies when user activity is very low. Studies should be easy to set up, execute, and conclude.
Checkbox with check Reference implementations may be necessary, such as the need to demonstrate idiosyncrasies of a service interface. For other than simple implementations, these may need to be scheduled as separate IT projects

Reviewing Study Results

Checkbox with check The results are briefed by the Study Team (or just the lead) to the individuals needing the results. Results of the study can be documented in briefing charts.
Checkbox with check Detailed study data is saved for later analysis, if needed.
Checkbox with check Uncertainty in the result or assumptions should be noted.
Checkbox with check Study test harnesses may be useful for future studies and should be saved or easily reproduced, if necessary.
Checkbox with check The results of the study should be reproducible (within limits), if necessary.

Top



Last Updated: May 4, 2005