Question 2 of 7: Government Regulation and the Final Design

Question 2: Government Regulation and the Final Design (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 government regulations impact a final design. A regulation differs from every other input to a design in one respect that governs everything else: it is not negotiable against other objectives. A customer preference, a cost target and a schedule can all be traded; a regulatory requirement either is met or the product cannot legally be sold. It therefore enters the design as a constraint rather than as a criterion, and the first practical consequence is that it must never be placed inside a weighted decision matrix, where a good score elsewhere could buy off a failure that is in fact absolute.

The impact then reaches the design through four distinct channels. First, regulations set hard performance floors that size structure and systems directly — a crash pulse, a roof-crush load, an emission limit, a minimum stability margin. Second, they foreclose whole regions of the design space: once a requirement dictates a restraint system, a fuel-system containment scheme or a lighting arrangement, the architectures that cannot accommodate it are simply unavailable, however attractive they are on cost or mass. Third, they impose evidence obligations — the design must not only comply but be demonstrably compliant, which means prescribed tests, prescribed test articles, prescribed instrumentation and retained records. That evidence burden is often a larger share of the programme than the physical change: it consumes prototypes, tooling and calendar. Fourth, they fix the market architecture: because requirements differ between jurisdictions, a manufacturer must choose between a single world design carrying every jurisdiction's worst case, and regional variants that multiply part numbers and inventory.

A fifth impact is temporal and is the one most often missed. Regulations are announced years before they bite and are frequently phased in, so the design must be compliant not with today's rule but with the rule in force at the start of production and, for a long-lived product, throughout its production life. Designing to the current text of a standard that is already in revision is a well-known way to deliver an obsolete product on time.

Part B — Three automotive regulations that impact a car's design. In Canada, road vehicles are regulated federally under the Motor Vehicle Safety Act and the Motor Vehicle Safety Regulations, whose technical schedules are the Canada Motor Vehicle Safety Standards (CMVSS). Three examples, chosen because each drives a different part of the vehicle:

1. CMVSS 216 — roof crush resistance. The standard requires the roof structure to resist a force applied through an inclined platen without the platen displacing more than a prescribed amount. Given. A vehicle of unloaded mass 1 860 kg and a required strength-to-weight ratio of 3.0. Find. The design load the A-pillar, roof rail and B-pillar load path must carry. The requirement is expressed as a multiple of the unloaded vehicle weight, so $$F=\text{SWR}\times m\,g=3.0\times1\,860\times9.81=\boxed{54.7\ \text{kN}}$$ applied at 25° pitch and 5° roll. That single number sizes the pillar sections, dictates the use of hot-stamped boron steel in the B-pillar of most modern vehicles, and constrains the glazing area and roof-line styling — which is why a safety regulation ends up visible in the shape of the car.
2. CMVSS 208 — occupant restraint systems in frontal impact. This drives the seat-belt architecture, pretensioners and load limiters, the airbag system and its suppression logic, and, through the injury criteria it imposes, the entire front-end crush structure and steering-column collapse design. It is the requirement that most strongly couples styling to engineering, because the crush distance the occupant needs is bought from the length of the front overhang.
3. CMVSS 108 — lamps, reflective devices and associated equipment. Canada's version of this standard has required daytime running lamps on all new vehicles since December 1989, several years before comparable requirements elsewhere. It is a good illustration of the market-architecture impact: a manufacturer selling into both Canada and jurisdictions without the requirement must either fit and wire the function on every vehicle or maintain a variant, and in practice the industry chose the former. It also constrains the front-end design directly, since lamp photometry, mounting height and separation are all prescribed.

A fourth family worth naming, because it now dominates powertrain design, is the greenhouse-gas and fuel-consumption regulation made under the Canadian Environmental Protection Act, 1999, which sets fleet-average carbon-dioxide standards on a vehicle-footprint basis. Unlike the safety standards above, this one is met by the fleet rather than by the individual vehicle, which changes the design problem from a constraint on one car into an optimisation across the product portfolio.

Part C — Steps to improve compliance within the design process. Six steps, each of which moves compliance earlier or makes it more visible:

1. Build a regulatory requirements register at the start, and put it under configuration control. Every applicable clause of every applicable standard in every target market is entered as a numbered requirement with an owner, a design feature that satisfies it, and a means of compliance. It is maintained beside the customer requirements, not in a separate compliance department.
2. Convert each clause into an engineering specification with margin. A regulation states a pass or fail threshold; a specification should state a design target with explicit margin against it, because test-to-test scatter, build variation and material variation will otherwise put half the population on the wrong side of a limit that the nominal design just meets.
3. Use virtual compliance early and continuously. Explicit finite element crash simulation, thermal and emission models and photometric ray-tracing allow every clause to be assessed against a digital prototype long before hardware exists. The value is not that simulation replaces the certification test — it does not — but that it converts a late, expensive discovery into an early, cheap one.
4. Design the compliance evidence as part of the design. Test articles, instrumentation access, sensor bosses and the ability to run a subsystem in isolation are design features and must be specified as such, or the certification programme will need bespoke and unrepresentative test hardware.
5. Run compliance reviews at every gate, with the regulator engaged. In regulated industries a certification plan agreed with the authority in advance — what will be shown, by what means, on what article — is worth more than any amount of internal review, because it removes the risk that the evidence is rejected after it has been produced.
6. Monitor the regulatory pipeline. Assign an owner to track notices of proposed rule-making, standards under revision and international harmonisation, and to feed a forward view into the specification. Designing to the rule that will be in force at start of production is the only defensible target.

Part D — At what stage should regulations and standards be considered. They must be considered from the planning and concept phase, before any concept is selected, and then continuously at every gate thereafter. Two arguments make this precise rather than a platitude.

The first is architectural. Regulatory requirements are the requirements most likely to invalidate an architecture rather than a detail, and architecture is chosen in the concept phase. A concept selected without knowing that a fuel-system integrity or roof-crush requirement applies may be structurally incapable of satisfying it at any cost, and no amount of detail design recovers that.

The second is economic, and can be put as a number. Given. The well-established rule of thumb that the cost of a design change rises by roughly an order of magnitude at each successive phase, with a change costing about CAD 8 000 to make at the concept stage. Find. The cost of the same change discovered at production tooling, three phases later. With $$\begin{aligned} C_k&=C_0\,m^{k} \\ C_0&=8\,000,\ m=10,\ k=3 \end{aligned}$$ the cost is $$\boxed{C_3=8\,000\times10^{3}=\text{CAD }8.0\ \text{million}}$$ so a compliance requirement found late costs a thousand times what it costs found early, and that ratio ignores the schedule slip and the recall exposure entirely. In regulated industries the recall case is worse still, because a non-compliance discovered after sale is not a design change but a field campaign across the whole production population.