Question 2 of 7: Diversity and the Design Process

Question 2: Diversity and the Design Process (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 — how diversity enhances the design process

Diversity acts on the design process in three distinct places, and it is worth separating them because they are strengthened by different actions.

It widens the set of needs the team can see. A design team discovers needs by recognising them, and a person recognises most easily the needs that resemble their own. A team drawn from one age band, one physical ability, one first language and one region will produce a specification calibrated to that band, and will be genuinely surprised by field failures that were predictable all along — the control that cannot be operated with arthritic hands, the instruction sheet that only works in English, the drainage detail that assumes rain rather than freeze–thaw. Canada's population diversity is directly useful here: designing for a user population that spans several first languages, a wide anthropometric range and a large proportion of older adults produces specifications that are more complete, and the resulting product is more robust in export markets as well.

It improves the quality of concept generation. Concept generation is an analogy-making activity: engineers solve a new problem by importing a mechanism from a problem they already know. The stock of available analogies is the union of the team's experience, so a team whose members come from mining, aerospace, food processing and medical devices — which Canadian industrial diversity makes easy to assemble — will generate a wider concept set than a team of specialists from one sector. Industrial diversity also creates real technology transfer: cold-weather materials knowledge from the resource sector, corrosion practice from the coasts, lightweight structures from aerospace.

It improves the quality of decisions and catches errors. Homogeneous groups converge quickly and confidently, which is comfortable and dangerous; the same shared assumptions that make agreement fast also make the group's blind spots common to every member, so nobody is positioned to challenge them. A group that does not share every premise argues more, converges more slowly and reaches better-tested conclusions — provided the process gives dissent a legitimate channel, which is part B's subject.

Two further benefits are practical rather than cognitive. Geographic diversity gives a Canadian design team access to genuinely severe validation environments within its own country — prairie winters, coastal salt fog, northern gravel roads — so that durability testing need not be simulated. And a team that resembles its market has more credibility with that market, which shows up in better recruitment, better access to users during needs research, and fewer expensive misjudgements of what a market will accept. Engineers Canada's 30 by 30 initiative and EGBC's equity, diversity and inclusion commitments are grounded in exactly this argument, alongside the ethical one.

Part B — the challenges, and how to overcome them

Communication friction and unstated assumptions. Where team members do not share a first language, an engineering education or a standards culture, the same word carries different meanings: “tolerance”, “qualification”, “factor of safety” and even “approved” are all ambiguous across traditions, and an engineer trained to ISO or to EN practice will read a drawing dimensioned to ASME Y14.5 differently than its author intended. The remedy is to make the shared language explicit and written: a project glossary, one nominated dimensioning and tolerancing standard stated on the drawing, interface control documents for every boundary between subsystems, and design decisions recorded in writing rather than settled verbally. Written records also give a non-native speaker time to read and respond carefully, which a fast meeting does not.

Slower convergence and the risk of unproductive conflict. The same disagreement that improves decisions can stall them. The remedy is structure rather than suppression: an explicit divergent phase in which alternatives are collected and criticism is deferred, followed by an explicit convergent phase using the formal tools of question 1 — Pugh screening and weighted scoring — because a criterion-based method moves the argument off personality and rank and onto evidence. Clear decision rights, recorded before the argument starts, ensure that a decision is eventually made and that everyone knows who makes it.

Unequal participation, deference and unconscious bias. In mixed-seniority, mixed-culture groups the same few people speak, and the contribution of the quietest member is lost even when it is the best. Remedies that work are procedural: rotating facilitation, round-robin idea collection, written brainstorming (each participant records ideas independently before any are discussed) and anonymous scoring in the first pass of concept selection. Bias training helps at the margin, but process design is what actually changes the outcome.

Credential recognition and uneven onboarding. Canada's engineering workforce includes many internationally trained engineers whose technical depth is not in doubt but whose familiarity with Canadian codes, regulator expectations and professional-practice obligations may be. Treating this as a deficiency wastes the expertise; treating it as an onboarding task captures it. The remedy is a structured mentoring pairing, an explicit induction into the applicable Canadian standards and into the EGBC licensure framework and Code of Ethics, and clear supervision arrangements under the responsible professional engineer taking charge of the work.

Distributed teams and time zones. A team spread across Canada spans four and a half hours, and one spread to an offshore partner may share no working hours at all. The remedies are the same ones that serve language diversity: a single controlled source of truth for the design data, disciplined asynchronous documentation, deliberate overlap windows reserved for decisions rather than status, and physical or virtual prototypes circulated as shared reference objects, because a prototype communicates across a language barrier better than any specification does.

The common thread in all five remedies is that diversity pays only when the process is designed to collect what the diverse team knows. Left to informal practice, a diverse team converges on the loudest voice and delivers the homogeneous team's answer more slowly; given explicit methods, records and decision rights, it delivers a better one.