Question 5 of 7: Design, materials, manufacturing and Canadian innovation (15 marks)

Question 5 — Design, materials, manufacturing and Canadian innovation (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.

Given. A Canadian plant before and after a design-led improvement programme: output 48 000 units per year from 62 000 labour-hours before, and 58 000 units from 60 000 hours after; pre-change unit price CAD 180, post-change price CAD 165; labour rate CAD 38/h; purchased materials CAD 3.10 M before and CAD 3.48 M after; energy CAD 140 000 and CAD 150 000; annualised capital services CAD 620 000 and CAD 900 000.

Find. How design, materials and manufacturing developments drive innovation, quantified through labour and multifactor productivity computed at constant prices, and the diagnosis this implies for Canada's innovation gap.

Approach. Treat the innovation question as a productivity question, since that is how the gap is actually measured; compute labour and multifactor productivity at constant prices, show what a current-price measure would wrongly report, and use the result to argue that design raises the numerator rather than only shrinking the denominator.

A. How design influences innovation

Innovation is not invention. Invention is a new technical idea; innovation is that idea reaching a user and creating value, and design is the discipline that spans the gap between them. Its influence operates in five ways.

Design decides where the cost is committed. The familiar result is that some 70 to 80 % of a product's life-cycle cost is committed by the end of the conceptual stage, when only a few per cent has been spent. Innovation that is attempted downstream — by squeezing suppliers or by process improvement alone — is working on the 20 % that remains uncommitted, which is why it produces diminishing returns while a design change on the same product produces step changes.

Design converts a technical capability into a use. A new sensor, alloy or battery chemistry is a capability with no market until a product architecture makes it usable. The iPad argument of Question 3 is the general case: the components existed and the innovation was the architecture. Countries with strong research output and weak design capability reliably export the capability and import the product.

Design differentiates on something other than price. A firm competing on price competes on capital intensity and wage rates, which is a competition Canada cannot win against much larger and lower-cost economies. A firm competing on designed attributes — usability, reliability, service life, appearance, integration — competes on value added per unit, which is a competition it can win.

Design de-risks the innovation itself. The systematic process of Question 2, with its iteration and its explicit specifications, is what turns a promising idea into a launchable product at an acceptable probability of failure. Much of what is described as a shortage of innovation is a shortage of successful commercialisation.

Design determines whether the innovation can be made and serviced. Design for manufacture and assembly, design for reliability and design for the environment decide whether the concept survives contact with a factory and a customer.

B. The role of materials in innovation potential

Materials set the boundary of the achievable, and moving that boundary opens designs that were not previously available at any price. Their influence runs through several channels. A material's property combination defines the feasible performance envelope: the aluminium airframe, the lithium-ion cell, the carbon-fibre pressure vessel and the wide-bandgap semiconductor each enabled a class of product rather than improving an existing one. Because the performance of a load-bearing part scales with a material index rather than with a single property — $E^{1/2}/\rho$ for a stiff, light beam, as in Question 7 — a new material is worth adopting only when it advances the index, and a material that is stronger but proportionately heavier changes nothing.

Materials also determine cost structure and supply risk, and here Canada's position is distinctive: the country is materials-rich, with aluminium, nickel, cobalt, lithium, potash, forest products and a substantial critical-minerals endowment, and this is precisely the point at which the innovation gap becomes visible. Exporting a mineral concentrate captures a small fraction of the value that is captured by exporting the cathode, the cell or the vehicle, and closing that gap is a design and manufacturing problem, not a mining one. Materials further enable substitution and function integration — a moulded polymer or composite part can replace an assembly of a dozen metal parts, which is simultaneously a materials innovation, a design innovation and a cost innovation — and they increasingly set the environmental and regulatory envelope, because recyclability, embodied carbon and restricted-substance compliance are now specifications rather than preferences.

The corresponding brake on innovation is qualification time. A structural or medical material may need five to ten years of testing before it may be used, so materials innovation has a long lead time that must be started before the product programme that will need it.

C. How manufacturing process developments impact innovation

Process development affects innovation in three distinct ways, and conflating them is the source of much confused argument. First, new processes make new geometries possible. Additive manufacturing permits internal conformal cooling channels, lattice structures and consolidated assemblies that no subtractive or formative route can produce; micro-moulding and semiconductor lithography define whole product categories by what they can resolve. Here the process is the innovation.

Second, process economics decide which innovations can be sold. The unit cost of a component is $c(n) = T/n + u$, where $T$ is tooling and $u$ the variable cost, so a process with high tooling and low variable cost only wins above the break-even volume $n^{*} = (T_2-T_1)/(u_1-u_2)$. Developments that lower $T$ — additive tooling, rapid die change, reconfigurable fixtures — shift $n^{*}$ downward and thereby make small-volume and customised products viable. That is a profound change in what can be innovated, because it removes the requirement that every new product be a mass-market product.

Third, process capability sets achievable tolerance and therefore achievable function. A design cannot specify what the factory cannot hold, so improvements in capability, measured as $C_p = T/6\sigma$, quietly enlarge the design space upstream. Automation, in-line metrology and statistical process control also shorten the iteration loop of Question 2, raising the learning rate per pass.

The Canadian productivity arithmetic. Innovation is measured, in the aggregate, as productivity, and this is where Canada's gap is located. Take the plant in the Given data, where a design-led programme cut part count and assembly time. Labour productivity is $LP = Q/H$:

$$LP_{0} = \frac{48\,000}{62\,000} = 0.774\ \text{units/h}, \qquad LP_{1} = \frac{58\,000}{60\,000} = 0.967\ \text{units/h}, \qquad \boxed{+24.9\ \%}$$

Expressed as value added per hour, the measurement convention decides the answer. At the pre-change price of CAD 180 — the constant-price basis, which is the correct one —

$$\frac{Q_0 p_0}{H_0} = \frac{48\,000(180)}{62\,000} = 139.36\ \text{CAD/h} \;\longrightarrow\; \frac{Q_1 p_0}{H_1} = \frac{58\,000(180)}{60\,000} = 174.00\ \text{CAD/h}$$

the same $+24.9$ %. But if the firm passes some of the gain to customers as a price cut to CAD 165 and the analyst uses current prices, the measured figure is $58\,000(165)/60\,000 = 159.50$ CAD/h, an apparent gain of only $\boxed{+14.5\ \%}$ — the plant looks half as improved as it is, purely because it shared the benefit. Real output must always be valued at a fixed base-period price; this is the single commonest error in productivity arguments and it systematically understates exactly the firms that are competing hardest.

Multifactor productivity, $MFP = Q p_0 / (Hr + M + E + K)$, tests whether the gain is real or merely bought with capital:

$$MFP_{0} = \frac{48\,000(180)}{62\,000(38)+3.10\text{M}+140\,000+620\,000} = \frac{8.64\text{M}}{6.216\text{M}} = 1.390$$

$$MFP_{1} = \frac{58\,000(180)}{60\,000(38)+3.48\text{M}+150\,000+900\,000} = \frac{10.44\text{M}}{6.810\text{M}} = \boxed{1.533}, \qquad +10.3\ \%$$

The multifactor gain is smaller than the labour gain because the programme consumed capital, and that difference is the honest measure of the innovation.

The diagnosis for Canada follows from the structure of these ratios. The gap is usually attributed to low capital intensity per worker, slow diffusion of technology into small and medium-sized enterprises, and a mix weighted toward resource extraction and low-value-added assembly. All three are design and manufacturing problems, not research problems: Canada's research output per capita is competitive, and the loss occurs at the point where research must become a product. The important implication of the arithmetic above is that part A's claim can be made precisely — design raises the numerator $Qp_0$ by making the unit worth more, whereas process improvement alone shrinks the denominator. A country that competes only on the denominator is competing on wage rates, and it will lose.