Engineering Design Process

Engineering is not invention by accident. It is the disciplined, iterative transformation of a human need into a working solution that satisfies constraints. The engineering design process is the backbone of every engineered artifact, from a bridge to a microprocessor. This skill covers the full design cycle with worked examples, decision tools, and integration with the college engineering concept graph.

Agent affinity: brunel (integrated design vision), polya-e (pedagogical scaffolding)

Concept IDs: engr-design-cycle, engr-problem-definition, engr-ideation-techniques, engr-design-communication

The Design Cycle at a Glance

The cycle is not linear. Phases 4 through 8 repeat until the design meets all requirements or the project constraints (budget, schedule) force a decision. The critical discipline is knowing when to iterate and when to converge.

Phase 1 -- Define the Problem

A well-defined problem is half-solved. Engineering problem definition translates a vague need ("we need a better bridge") into a precise engineering problem statement with measurable criteria.

Template. Design a [system/component] that [primary function] for [stakeholders] subject to [key constraints], measured by [success criteria].

Worked example. Design a pedestrian bridge that spans a 30-meter river crossing for a rural community subject to a $200K budget and 12-month timeline, measured by load capacity (minimum 500 kg/m), deflection limits (L/360), and 50-year design life.

Common mistake. Jumping to solutions before defining the problem. "We need a suspension bridge" is a solution, not a problem statement. The problem statement must be solution-neutral to preserve the design space.

Stakeholder Analysis

Every engineering project has stakeholders beyond the end user. A bridge serves pedestrians but must also satisfy regulators, maintenance crews, the funding authority, and adjacent property owners. Missing a stakeholder leads to requirements gaps that surface late, when they are expensive to fix.

Map stakeholders using: Who uses it? Who pays for it? Who maintains it? Who regulates it? Who is affected by it?

Phase 2 -- Research

Before generating new designs, understand what exists. Research covers:

• Prior art: What solutions exist for similar problems? What worked and what failed?
• Physics and materials: What physical principles govern the problem domain?
• Standards and codes: What regulatory requirements constrain the design?
• Failure history: What engineering failures in this domain provide lessons?

Research is not optional. Skipping it leads to reinventing known solutions or, worse, repeating known failures. The Hyatt Regency walkway collapse (1981) resulted from a design change that no one analyzed against structural principles -- a failure of research and review, not of engineering knowledge.

Phase 3 -- Specify Requirements

Requirements are the contract between the problem and the solution. They must be:

• Measurable: "Strong enough" is not a requirement. "Withstand 500 kg/m distributed load with deflection less than L/360" is.
• Testable: Every requirement must have a corresponding test. If you cannot test it, you cannot verify it.
• Traceable: Each requirement links back to a stakeholder need and forward to a design feature.
• Prioritized: Must-have (the design fails without it) vs. should-have (the design is degraded without it) vs. nice-to-have (improvement, not essential).

Functional vs. Non-Functional Requirements

Constraint Matrix

Constraints are non-negotiable boundaries. They differ from requirements in that they cannot be traded off -- they are binary pass/fail.

Phase 4 -- Ideate

Generate multiple design alternatives before committing to one. The goal is divergent thinking -- quantity of concepts, not quality.

Brainstorming Rules

1. No evaluation during ideation. Criticism kills creativity. Evaluate later.
2. Build on others' ideas. "Yes, and..." not "No, but..."
3. Encourage wild ideas. They often contain seeds of practical solutions.
4. Go for quantity. More concepts increase the probability of finding a good one.

Morphological Chart

A morphological chart decomposes the design into independent sub-functions and lists alternative solutions for each.

Each combination of options is a candidate design. For 4 sub-functions with 3 options each, there are 81 possible combinations. The morphological chart makes the design space explicit and prevents premature convergence on a single concept.

TRIZ

TRIZ (Theory of Inventive Problem Solving) is a systematic method for resolving design contradictions. When improving one parameter (strength) degrades another (weight), TRIZ provides 40 inventive principles for resolving the contradiction. Common principles include: segmentation, taking out, local quality, asymmetry, merging, universality, nesting, counterweight.

Phase 5 -- Analyze and Select

Evaluate candidate designs against requirements using structured methods.

Pugh Matrix (Controlled Convergence)

Select a baseline design (often the simplest or most conventional). Score each alternative relative to the baseline: better (+), same (S), or worse (-) on each criterion. Sum the scores. The design with the highest net score is the leading candidate.

Trade Study

For more rigorous evaluation, assign numerical scores (1-5 or 1-10) to each criterion for each alternative, multiply by weight, and sum. A trade study produces a defensible, documented selection rationale.

Feasibility Assessment

Before detailed design, verify that the selected concept is physically, technically, economically, and sched