Unit 4: Ideation and Creative Problem Solving

I. Orientation

Design thinking treats complex problems as human-centred, uncertain, and open to reframing. Ideation follows empathic problem definition and produces possible responses before solutions are assessed, prioritised, and selected. Its governing principle is diverge before converging: first expand the range of possibilities, then narrow it using explicit evidence and criteria.

• Human-centredness: Ideas respond to identified user needs, such as reducing waiting time for patients rather than merely adding technology.
• Divergent thinking: Participants deliberately seek variety, quantity, and unusual connections instead of judging ideas immediately.
• Convergent thinking: Alternatives are compared and narrowed using objectives, constraints, evidence, and stakeholder values.
• Deferred judgement: During early ideation, criticism is postponed so that partially formed ideas can develop.
• Ambiguity and iteration: A selected concept is provisional; prototyping and feedback may lead back to reframing or further ideation.
• Explicit criteria: Terms such as “affordable,” “safe,” or “inclusive” must be translated into observable indicators.
• Traceability: Decisions should show the link between the problem, idea, criterion, evidence, and final choice.

II. Creative Thinking Methods — Expanding the Solution Space

Creative thinking methods are structured ways of producing alternatives by changing perspectives, combining concepts, or challenging assumptions. They are not random inspiration: each method provides a deliberate stimulus for moving beyond the first obvious answer.

A. Exploration of creative thinking methods for generating solution alternatives

The purpose of exploration is to expose hidden assumptions and create several plausible directions before evaluation begins.

• Reframing: Restate the problem from a user-centred perspective. “How might we reduce queue frustration?” creates more possibilities than “How might we install more counters?”
• Analogy: Transfer a principle from another context. An airport’s staged boarding process may inspire orderly clinic check-in without copying the airport literally.
• SCAMPER: Modify an existing service through Substitute, Combine, Adapt, Modify, Put to another use, Eliminate, or Reverse.
  • Concrete prompt: For a library service, “What could be eliminated?” might produce self-checkout; “What could be reversed?” might produce book delivery before physical visits.
• Morphological analysis: Break a problem into dimensions and list alternatives for each. A delivery service may vary by vehicle, route, handover point, and booking method; combining one choice from each dimension creates concepts.
• Lateral thinking: Introduce a deliberate provocation, such as “Assume customers never visit the shop,” then investigate what useful principle follows.
• Design principles: Methods should increase variety, preserve a clear connection to the user need, and avoid treating the first feasible idea as the final answer.

B. Techniques for idea generation

Techniques convert creative methods into activities that individuals or groups can perform within a defined time.

• Brainstorming: Generate many ideas rapidly under four classic conditions: defer judgement, encourage unusual ideas, build on others’ ideas, and pursue quantity. A facilitator records every idea visibly.
• Brainwriting 6–3–5: Six participants write three ideas in five minutes, then pass their sheets for further development. It reduces dominance by confident speakers.
• Mind mapping: Place the central challenge in the middle and branch into users, causes, contexts, resources, and possible responses. Branches reveal relationships that a linear list can hide.
• Crazy 8s: Fold a sheet into eight panels and sketch one concept per panel in a short timed session. Speed discourages over-polishing and encourages visual alternatives.
• Worst possible idea: Deliberately generate harmful or absurd responses, such as “make the queue longer,” then reverse them into useful principles, such as “make waiting productive.”
• Role storming: Participants ideate as another stakeholder—such as a child, maintenance worker, regulator, or competitor—to expose overlooked requirements.
• Affinity clustering: Group similar ideas after generation, label each cluster, and identify recurring needs rather than counting repeated wording as separate innovation.
• Worked example: For “improve university meal access,” brainstorming may produce mobile ordering, extended hours, pre-packed meals, shared kitchens, and subsidised vending machines. Clustering separates access, speed, affordability, and dietary inclusion.

C. Evaluation of solutions based on defined criteria and objectives

Evaluation determines how well an alternative responds to the problem; it should follow ideation rather than interrupt it.

• Separate stages: During generation, ask “What else is possible?” During evaluation, ask “Which options best satisfy the agreed objectives?”
• Evidence quality: Distinguish assumptions from evidence. “Students will use an app” is an assumption until interviews, usage data, or a prototype support it.
• Balanced judgement: Consider desirability for users, feasibility for the organisation, and viability over time. A popular idea may still fail if it cannot be maintained.
• Risk visibility: Record safety, privacy, environmental, legal, and reputational risks instead of hiding them inside a single score.
• Avoid premature certainty: Early scores indicate current understanding; they do not prove that an idea will work in practice.

III. Idea Generation Techniques — Producing and Developing Alternatives

Idea generation is the active production of candidate responses. Strong practice creates enough breadth to compare genuinely different approaches, then develops promising ideas so that they are testable rather than vague slogans.

A. Exploration of creative thinking methods for generating solution alternatives

Methods can be combined when one technique produces too few or too similar alternatives.

• Sequence: Begin with open brainstorming, use SCAMPER or analogy to stretch familiar ideas, and use a constraint prompt to expose new possibilities.
• Constraint switching: Apply different conditions, such as “half the budget,” “no smartphone,” or “must serve users in five minutes.” Constraints can stimulate useful redesign rather than merely restrict it.
• Combination: Pair ideas from unrelated clusters. “Mobile ordering” plus “dietary inclusion” may become a menu system that flags allergens and suggests accessible alternatives.
• Levels of abstraction: Move from a concrete object to its function. Instead of asking for “more benches,” ask how to support “comfortable waiting,” which could produce seating, appointments, notifications, or entertainment.
• Stakeholder coverage: Generate at least one alternative from the viewpoint of each affected group, including non-users, staff, suppliers, and communities experiencing indirect effects.
• Facilitation conditions: Use a clear prompt, visible time limit, accessible materials, and a psychologically safe atmosphere. The facilitator protects contribution without evaluating content prematurely.

B. Techniques for idea generation

The quality of an idea depends partly on how it is recorded and developed.

• Idea statement: Express each concept as user, action, and benefit: “For commuters, provide real-time platform updates through low-cost displays to reduce uncertainty.”
• Sketching and storyboarding: Show the user’s sequence before, during, and after the proposed interaction. This reveals missing steps, such as payment, support, or data consent.
• Concept cards: Record the idea, target user, need addressed, main benefit, assumptions, and resources. This makes different ideas comparable.
• Build-on prompts: Ask “What would make this safer?”, “What would make it accessible?”, and “What would make it work offline?” to develop an initial concept.
• Individual-to-group balance: Allow silent individual generation before discussion, then collaborate. This combines independent originality with collective refinement.
• Avoiding common failures: Do not let hierarchy, loud voices, vague prompts, or early feasibility criticism determine the idea pool.
• Output measure: Count meaningful alternatives and the diversity of approaches, not simply the number of sticky notes.

IV. Evaluation of Solutions Based on Defined Criteria and Objectives — Judging Fit

Evaluation is a systematic comparison of alternatives against the problem’s objectives and the criteria derived from them. It should make trade-offs visible rather than pretending that one option is best in every dimension.

A. Evaluation of solutions based on defined criteria and objectives

Criteria must be defined before scoring to reduce bias and inconsistent standards.

• Objectives: State desired outcomes, such as reducing average service time from 20 minutes to 10 minutes or increasing access for wheelchair users.
• Criteria: Translate objectives into dimensions such as user value, accessibility, cost, technical feasibility, environmental impact, and risk.
• Indicators: Define what evidence counts. “Affordable” may mean implementation cost below £50,000 and operating cost below £5 per transaction.
• Scoring scale: Use a consistent scale, for example 1–5, where 1 means very poor fit and 5 means excellent fit. Define anchor descriptions rather than relying on intuition.
• Weighted decision matrix: Assign each criterion a weight (w_i), score each idea (s_i), and calculate:

Total score = Σ(w_i × s_i)

Here, (w_i) is the importance weight and (s_i) is the criterion score. If weights sum to 1, the total remains on the scoring scale.

• Worked example: If accessibility has weight 0.40 and an idea scores 5, its contribution is (0.40 × 5 = 2.0). Cost weighted 0.30 with score 3 contributes 0.9; feasibility weighted 0.30 with score 4 contributes 1.2. The total is 4.1 out of 5.
• Qualitative evidence: A matrix should be supported by user feedback, cost estimates, technical review, or prototype observations; numerical precision must not disguise weak evidence.

B. Applications and limitations

Evaluation tools support discussion, but they cannot replace judgement about uncertainty and values.

• Application: Use a quick feasibility screen to remove ideas that violate non-negotiable safety or legal constraints before detailed scoring.
• Application: Use desirability, feasibility, and viability as a balanced first comparison, then investigate high-potential concepts through prototyping.
• Limitation—false objectivity: A score of 4 rather than 3 may reflect opinion if evidence and scale anchors are unclear.
• Limitation—missing effects: Long-term environmental or social consequences may not appear in short-term financial calculations.
• Limitation—interdependence: Two moderate ideas may combine into a stronger system; evaluating them only as isolated options can miss that possibility.

V. Prioritisation of Solutions Based on Defined Criteria and Objectives — Ordering Action

Prioritisation orders alternatives according to importance, urgency, value, and effort. Unlike evaluation, which asks how well an idea performs, prioritisation asks which ideas deserve attention or resources first.

A. Prioritisation of solutions based on defined criteria and objectives

Prioritisation requires an explicit basis for deciding what proceeds, waits, combines, or stops.

• Value–effort matrix: Plot user or organisational value against implementation effort. High-value, low-effort ideas are early candidates; low-value, high-effort ideas are usually deferred.
• Impact–risk comparison: A high-impact idea with high risk may require a small experiment before full commitment, while a low-impact, high-risk idea may be rejected.
• MoSCoW logic: Classify requirements as Must have, Should have, Could have, or Won’t have for now. “Must” should describe a genuine condition of success, not every stakeholder preference.
• Weighted priority score: A practical formula is:

Priority score = (expected benefit × confidence) ÷ effort

Expected benefit and effort need defined scales; confidence represents evidence strength, such as 0.5 for a largely untested assumption.

• Dependencies: An apparently attractive feature may rank lower if it depends on infrastructure that does not yet exist.
• Resource limits: A portfolio should contain a mix of quick wins, strategic investments, and learning experiments rather than only the highest individual scores.
• Fairness: Prioritisation should consider who benefits and who bears costs; average impact can conceal harm to a smaller vulnerable group.

B. Applications and limitations

Prioritisation is most useful when teams face more promising ideas than available time, money, or expertise.

• Application: Rank concepts for prototyping, not necessarily for permanent implementation. A risky idea may deserve an early low-cost test because it has high learning value.
• Application: Revisit rankings when new user evidence changes confidence, cost, or expected impact.
• Limitation: A two-dimensional matrix oversimplifies uncertainty and may favour visible short-term benefits over structural change.
• Limitation: Stakeholders may manipulate weights to make a preferred option appear first; publish assumptions and conduct a sensitivity check.
• Sensitivity check: If small changes in weights reverse the ranking, the decision is fragile and should be tested further rather than presented as definitive.

VI. Selection of Solutions Based on Defined Criteria and Objectives — Committing to a Direction

Selection is the final reasoned choice among prioritised alternatives, usually accompanied by a prototype, pilot, or implementation plan. It is a commitment to the next justified action, not a claim that uncertainty has disappeared.

A. Selection of solutions based on defined criteria and objectives

A sound selection combines comparative evidence, stakeholder judgement, and a plan for managing remaining uncertainty.

• Decision rule: Choose the alternative that best satisfies the non-negotiable requirements and offers the strongest balance across weighted objectives.
• Thresholds: Reject an idea if it fails a critical condition, such as legal compliance, basic safety, or minimum accessibility, even if its total score is high.
• Stakeholder alignment: Explain how the choice affects users, implementers, funders, and communities. Agreement is strengthened when dissent and trade-offs are recorded.
• Assumption testing: Identify the riskiest assumption and design the smallest useful experiment. For a digital booking system, test whether users can complete booking before investing in full development.
• Pilot selection: Select a reversible, measurable first step where uncertainty is high. Define success measures, such as completion rate, error rate, cost per user, and satisfaction.
• Decision record: Document the chosen concept, rejected alternatives, criteria, scores, evidence, risks, owner, timeline, and conditions for revisiting the decision.
• Portfolio selection: Sometimes the best decision is a combination: implement a reliable low-risk improvement while prototyping a more radical alternative.

B. Significance and analytical dimension

Selection completes the movement from imagination to accountable action, but design thinking remains iterative.

• Learning orientation: A failed prototype can be valuable if it disproves a critical assumption cheaply and changes the next design decision.
• Reversibility: Prefer reversible commitments when evidence is weak; reserve irreversible investment for options supported by stronger validation.
• Outcome monitoring: Compare actual results with objectives, such as a measured reduction in waiting time rather than a belief that the service feels faster.
• Ethical responsibility: Assess unintended consequences, exclusion, privacy, accessibility, and environmental effects before scaling.
• Iteration: New evidence may require returning to ideation, changing criteria, reprioritising alternatives, or selecting a different solution. The process is therefore a learning cycle rather than a one-way funnel.