22-Mec-B5 Product Design and Development · December 2017

Question 2 of 7: The Design Process, Iteration, and What Makes a Design Complete

Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)

Notes on this paper

Paper format. National Exams, December 2017. Three (3) hours. OPEN BOOK; an approved Casio or Sharp calculator is permitted. Question 1 is compulsory and carries 40 marks; four (4) of the remaining six (6) questions are chosen, each worth 15 marks, for 100 marks attempted out of 130 printed. Only the first five questions appearing in the answer book are marked. The marking scheme is printed on page 4 of the paper and is reproduced against each question below. Most answers are expected in essay form, supported by tables, figures and charts.

How to use this document. Every one of the seven printed questions is answered in full, not just the five a candidate would attempt, so that the set works as a study resource. This is a descriptive design-methodology paper: the marks are for method, structure and judgement rather than for arithmetic. Where a number genuinely sharpens an argument — a DFA index, a process break-even, a capability index, a material index — it is computed explicitly and framed with Given. and Find. so the reasoning can be checked. All monetary figures are Canadian dollars.

Reference texts for 16-Mec-B5

Ulrich, K. T. & Eppinger, S. D., Product Design and Development, 6th ed., McGraw-Hill — the framework text for this syllabus (concept development, specifications, DFM, industrial design, prototyping).
Boothroyd, G., Dewhurst, P. & Knight, W., Product Design for Manufacture and Assembly, 3rd ed., CRC Press — the DFA index, handling and insertion time standards, and process-cost models.
Dieter, G. E. & Schmidt, L. C., Engineering Design, 5th ed., McGraw-Hill — design process, decision methods, materials and process selection.
Pahl, G., Beitz, W., Feldhusen, J. & Grote, K.-H., Engineering Design: A Systematic Approach, 3rd ed., Springer — systematic embodiment and detail design.
Ashby, M. F., Materials Selection in Mechanical Design, 5th ed., Butterworth-Heinemann — material indices, screening and ranking (Question 7).
Kalpakjian, S. & Schmid, S. R., Manufacturing Engineering and Technology, 7th ed., Pearson — process capability, tolerance and process economics.
Montgomery, D. C., Introduction to Statistical Quality Control, 7th ed., Wiley — capability indices $C_p$, $C_{pk}$ (Question 3).
Canadian Intellectual Property Office (CIPO), A Guide to Patents and A Guide to Industrial Designs; Patent Act R.S.C. 1985 c. P-4 (Question 4).

Question 2: The Design Process, Iteration, and What Makes a Design Complete (15 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

Part A — Design as a verb and as a noun

Design as a verb is the activity of converting an incompletely stated need into a completely specified physical realisation, under constraints that are only partly known when the activity begins. Functionally, it is a search: the designer proposes candidate solutions, evaluates them against requirements, and uses the discrepancy to propose better candidates. Three properties distinguish it from analysis. It is divergent before it is convergent — the solution set must be widened before it is narrowed, or the chosen concept is merely the first one thought of. It is constrained but underdetermined — the requirements are necessary conditions, never sufficient ones, so there is no unique right answer and the designer must supply judgement the requirements do not contain. And it is decision-consuming — each commitment removes freedom from every later stage, which is why the cost of a change grows so steeply with time.

Design as a noun is the complete, unambiguous and self-consistent description of the artefact that results: the geometry, the materials, the tolerances, the processes, the interfaces and the operating and disposal conditions, together with the evidence that the description satisfies the requirements. The functional test of the noun is reproducibility — a competent third party, given the design and nothing else, must be able to make the product and get the same result. If they must telephone the designer to ask a question, the noun is not yet finished, however finished the verb feels.

The distinction is not academic. Contract disputes, professional-liability claims and technology-transfer failures usually arise because a client bought the noun and the engineer delivered the verb: a design direction, a set of sketches and a body of undocumented judgement that lived only in the design team. Under Canadian professional practice the deliverable is the noun, sealed, and the seal certifies that the description is complete enough to be built from.

Part B — Important steps in a typical design process

Every serious formulation — Ulrich and Eppinger, Pahl and Beitz, Dieter and Schmidt — contains the same six moves under different names.

Clarify the need and plan the project. Identify who the customer is, what problem they actually have as distinct from the solution they asked for, the business case, and the market and regulatory context. The output is a mission statement with a scope boundary.
Establish requirements and target specifications. Convert customer statements into measurable metrics with target and marginally acceptable values. This is the step that makes later evaluation possible; a process that skips it can generate concepts but cannot choose between them.
Generate concepts. Decompose the function, search externally (patents, competitors, analogous industries) and internally (structured ideation, morphological charts), and produce a genuinely diverse set of architectures rather than variants of one idea.
Select and embody a concept. Screen against a datum, then score the survivors against weighted criteria, then commit. Embodiment fixes the architecture, the layout, the major materials and the manufacturing routes, and it is where most of the eventual cost is determined.
Detail, prototype and validate. Complete every dimension, tolerance, material and process specification; build physical and analytical prototypes; verify against the requirement set and validate against the original need. Design reviews at defined gates are part of this step, not an interruption of it.
Release to production and support the life cycle. Transfer to manufacturing with a controlled data package, ramp up, then manage field feedback, service, obsolescence and end-of-life recovery. Design responsibility does not end at release.

Figure 2.1 — The six steps run forward, but the loops shown dashed are where most of the learning happens. The further right a loop starts, the more expensive it is.

Part C — The importance of iteration

Iteration is not a symptom of a badly run project; it is the mechanism by which design converges at all. Because the requirements are incomplete at the start and the designer’s knowledge of the solution is lowest exactly when the most consequential decisions are being made, the only way to reduce uncertainty is to build a candidate, test it against reality and use the error. Each pass closes some fraction of the remaining gap between the current design and a compliant one, so the gap decays geometrically rather than linearly.

Given. Each design pass closes 40 per cent of the remaining requirement gap, so the retention ratio is $\rho = 0.60$. A residual gap of 4 per cent is considered acceptable for release. Find. The number of passes required.

Model the residual gap. After $k$ passes the residual gap is $g_k=\rho^{k}$. Setting $g_k = 0.04$ and solving, $$k=\frac{\ln g_{\text{target}}}{\ln \rho}=\frac{\ln 0.04}{\ln 0.60}=6.30$$ so $\boxed{k = 7\ \text{iterations}}$ are needed, since a partial pass does not exist. The intermediate values are instructive: 60, 36, 21.6, 13.0, 7.8, 4.7 and 2.8 per cent, which is why a project that has budgeted three iterations is still 21.6 per cent away when it runs out of schedule.

Figure 2.2 — Geometric decay of the residual requirement gap at a 40 per cent closure rate per pass. Seven passes are needed to reach a 4 per cent residual; the first three passes do most of the visible work and the last four do most of the convincing.

Two consequences follow, and they are what the marks are for. First, iteration must be planned and budgeted, because seven passes at a realistic closure rate is a schedule item, not an accident. Second, and more important, iterations must be pushed early. The cost of a change rises by roughly an order of magnitude at each stage boundary — concept, detail, tooling, production, field — so a loop closed at the concept stage and the same loop closed after tooling differ in cost by a factor approaching ten thousand across four boundaries. This is the entire economic argument for cheap early prototypes, for simulation, and for set-based concurrent practice: they are ways of buying iterations at the price of the left-hand end of the process. The correct summary is that you do not get to choose whether to iterate, only where.

Part D — Key requirements that define a final, complete design

A design is complete when all of the following can be answered without consulting the designer.

Function is fully specified and demonstrated. Every required function is defined with a measurable performance level, and there is verification evidence that the design achieves it across the whole operating envelope, not at the nominal point only.
Form is completely defined. Geometry, dimensions, tolerances, surface finish and appearance requirements are stated unambiguously, and the tolerance stack-ups close at every interface.
Material is specified to a standard. Grade, condition, supplier qualification and acceptance criteria, together with any restriction on substitution.
Process is specified and shown to be capable. The manufacturing route is named and demonstrated capable of holding the specified tolerances at the required rate, with capability evidence for the critical characteristics.
Interfaces are controlled. Every mechanical, electrical, thermal, software and human interface is defined on a controlled document that both sides have signed.
Compliance is evidenced. Applicable codes and standards are identified and the certification test results exist — for a Canadian consumer appliance, the CSA/IEC series applicable to the product class.
The life cycle is addressed. Service, spare parts, expected life, maintenance intervals, packaging, transport and end-of-life recovery or disposal.
The data package is complete, controlled and traceable. Drawings, models, bills of material, specifications and test reports at a released revision, under change control, with a requirements trace showing that every requirement has an owner and a verification record.

The four-way test that Dieter and Schmidt use is a compact way to remember the first four: a complete design fixes function, form, material and process, and the marks are usually in noticing that these four are coupled rather than independent — changing the material changes the feasible processes, which changes the achievable tolerances, which changes the form, which can change the function.

Concept & Theory

Topic. The design process, iteration economics, and design completeness.

Key relations. Residual gap $g_k=\rho^{k}$; passes required $k=\ln g_{\text{target}}/\ln\rho$; the order-of-magnitude rule for the cost of change across stage boundaries.

Why this works. Iteration converges geometrically because each pass acts on what remains, not on the original gap — the same structure as any proportional feedback loop, and the reason convergence is asymptotic rather than finite. That is also why a fixed iteration budget is dangerous: the residual after a given number of passes depends entirely on the closure rate, which is a property of how good the feedback is. Better feedback — a physical prototype rather than an opinion, a validated simulation rather than a rule of thumb — raises the closure rate and cuts passes, which is the real argument for spending money on test capability.

Common pitfall. Students frequently describe the design process as strictly sequential and then treat any loop as rework to be eliminated. The examiner is looking for the opposite: loops are the value-adding part, and the objective is to move them left, not to remove them. A second common error is listing only function and form under Part D and omitting process capability and the controlled data package, which is where most real handover failures actually occur.

Source. Ulrich & Eppinger, Product Design and Development, 6th ed., ch. 2 (development processes) and ch. 5–6; Dieter & Schmidt, Engineering Design, 5th ed., ch. 1 and ch. 2 (the design process, function-form-material-process); Pahl & Beitz, ch. 4–5 (conceptual and embodiment design).

← Question 1 Next: Question 3 →