Design Science Research (DSR)

Activity 1: Problem Identification & Motivation

Purpose

Define the research problem and justify the value of a solution.

Questions to Answer

• What is the problem?
• Why is it important?
• Who is affected?
• What is the current state?
• What are the consequences of not solving it?

Template

<problem_identification>
  <problem_statement>
    [Clear, concise statement of the problem]
  </problem_statement>

  <importance>
    [Why does this matter? Quantify if possible]
  </importance>

  <stakeholders>
    [Who is affected by this problem?]
  </stakeholders>

  <current_state>
    [How is this currently handled? What are the limitations?]
  </current_state>

  <consequences>
    [What happens if we don't solve this?]
  </consequences>
</problem_identification>

Example

<problem_identification>
  <problem_statement>
    Human-AI collaboration lacks personalization, leading to repetitive
    context-setting and suboptimal interactions.
  </problem_statement>

  <importance>
    Developers spend 15-30% of AI interaction time re-establishing context.
    This reduces productivity and leads to frustration.
  </importance>

  <stakeholders>
    - Developers using AI assistants
    - Teams adopting AI-augmented workflows
    - Organizations investing in AI tools
  </stakeholders>

  <current_state>
    - Generic system prompts used for all users
    - No persistent memory of user preferences
    - Context lost between sessions
  </current_state>

  <consequences>
    - Continued inefficiency in human-AI collaboration
    - Slower AI adoption due to friction
    - Missed potential of personalized AI assistance
  </consequences>
</problem_identification>

Activity 2: Define Objectives of a Solution

Purpose

Infer objectives from the problem definition and knowledge of what is possible.

Questions to Answer

• What would an ideal solution do?
• What are the success criteria?
• How will we know if we've succeeded?
• What constraints must we respect?

Template

<objectives>
  <ideal_solution>
    [Description of what success looks like]
  </ideal_solution>

  <success_criteria>
    <criterion metric="[metric_name]">
      [Specific, measurable criterion]
    </criterion>
    <!-- Add more criteria -->
  </success_criteria>

  <constraints>
    <constraint type="technical">[constraint]</constraint>
    <constraint type="resource">[constraint]</constraint>
    <constraint type="time">[constraint]</constraint>
  </constraints>

  <non_goals>
    [What are we explicitly NOT trying to achieve?]
  </non_goals>
</objectives>

Objective Types

Activity 3: Design & Development

Purpose

Create the artifact - constructs, models, methods, or instantiations.

Artifact Types

Template

<design>
  <artifact_type>[construct|model|method|instantiation]</artifact_type>

  <architecture>
    [High-level design of the artifact]
  </architecture>

  <components>
    <component name="[name]">
      <purpose>[what it does]</purpose>
      <interface>[how to interact with it]</interface>
    </component>
  </components>

  <design_decisions>
    <decision>
      <choice>[what was decided]</choice>
      <rationale>[why this choice]</rationale>
      <alternatives>[what else was considered]</alternatives>
    </decision>
  </design_decisions>

  <theoretical_foundation>
    [What theories/prior work inform this design?]
  </theoretical_foundation>
</design>

Design Principles (Hevner's Guidelines)

1. Artifact Focus - Produce something concrete and usable
2. Problem Relevance - Solve a real problem that matters
3. Rigor - Apply appropriate theoretical foundations
4. Search Process - Iterate through solution space systematically

Activity 4: Demonstration

Purpose

Demonstrate use of the artifact to solve the problem.

Demonstration Methods

Template

<demonstration>
  <method>[proof_of_concept|case_study|simulation|experiment]</method>

  <context>
    [Where/how is the artifact being demonstrated?]
  </context>

  <scenario>
    <input>[what goes in]</input>
    <process>[what happens]</process>
    <output>[what comes out]</output>
  </scenario>

  <observations>
    [What did we observe during demonstration?]
  </observations>

  <artifacts_produced>
    [What tangible outputs were created?]
  </artifacts_produced>
</demonstration>

Activity 5: Evaluation

Purpose

Observe and measure how well the artifact supports a solution.

Evaluation Methods

Template

<evaluation>
  <method>[evaluation method]</method>

  <criteria>
    <criterion name="[name]" weight="[0-1]">
      <definition>[what does this measure?]</definition>
      <measurement>[how do we measure it?]</measurement>
      <threshold>[what counts as success?]</threshold>
    </criterion>
  </criteria>

  <results>
    <result criterion="[name]">
      <score>[actual measurement]</score>
      <evidence>[supporting data]</evidence>
    </result>
  </results>

  <analysis>
    [Interpretation of results]
  </analysis>

  <limitations>
    [What are the limitations of this evaluation?]
  </limitations>

  <improvements>
    [What could be improved based on evaluation?]
  </improvements>
</evaluation>

Evaluation Checklist

• Are criteria aligned with objectives?
• Is evidence quantifiable where possible?
• Are limitations acknowledged?
• Is comparison to alternatives included?
• Are threats to validity addressed?

Activity 6: Communication

Purpose

Communicate the problem, artifact, and results to relevant audiences.

Audiences

Template

<communication>
  <audience>[target audience]</audience>

  <key_messages>
    <message priority="1">[most important takeaway]</message>
    <message priority="2">[second takeaway]</message>
    <message priority="3">[third takeaway]</message>
  </key_messages>

  <contribution_type>
    [artifact|foundation|methodology]
  </contribution_type>

  <artifacts>
    <artifact type="[type]" location="[where]">
      [description]
    </artifact>
  </artifacts>

  <reproducibility>
    [How can others reproduce/build on this work?]
  </reproducibility>
</communication>

Hevner's 7 Guidelines Checklist

Use this to validate your DSR project:

Entry Points

You can enter the DSRM process at different points:

Iteration Patterns

Pattern 1: Linear

1 → 2 → 3 → 4 → 5 → 6

Use when: Problem is well-understood, single iteration sufficient.

Pattern 2: Evaluation Loop

1 → 2 → 3 → 4 → 5 → (improvements) → 3 → 4 → 5 → 6

Use when: Iterative refinement needed based on evaluation.

Pattern 3: Problem Refinement

1 → 2 → 3 → 4 → (new insights) → 1 → 2 → 3 → 4 → 5 → 6

Use when: Demonstration reveals problem was misunderstood.

Pattern 4: Continuous

1 → 2 → 3 → 4 → 5 → 6 → (feedback) → 1 ...

Use when: Ongoing research program with multiple cycles.

Integration with Cognitive Workflows

DSR activities can use cognitive workflow patterns:

References

• Hevner, A. R., March, S. T., Park, J., & Ram, S. (2004). Design Science in Information Systems Research. MIS Quarterly, 28(1), 75-105.
• Peffers, K., Tuunanen, T., Rothenberger, M. A., & Chatterjee, S. (2007). A Design Science Research Methodology. JMIS, 24(3), 45-77.

Evidence: evidence/patterns/004-design-science-research-methodology.md