The design process applied to a sustainable green bin: 22-Mec-B5 Product Design and Development, May 2014, Q1

Question 1: The design process applied to a sustainable green bin (40 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.

Product selected: (iii) the green bin for organic waste — the 120 L wheeled cart that a Canadian municipality issues to each household for the curbside collection of food and yard organics. It is chosen because its sustainability is governed as much by a societal variable (whether households actually use it) as by a technological one (what it is made of), which is exactly the tension the question sets up. The answer follows one design direction from end to end.

The design process used in this answer, with each exam part placed on the phase it belongs to. The loop from F back to D is deliberate: selection normally re-opens one or two specifications.

Part A — Establishing the current level of sustainability

Sustainability cannot be improved until it is measured, so part A is a measurement exercise, not a design one. The method is a screening life-cycle assessment following the ISO 14040/14044 framework, cut down to what one engineer can complete in a week. Four decisions come first, and they matter more than the arithmetic that follows. The functional unit is one household served with organics collection for one year, not one bin — comparing bins is meaningless if one lasts twice as long as the other. The system boundary runs cradle to grave and deliberately includes the collection and treatment that the bin enables, because a container whose only job is to make diversion happen must be credited or debited with the diversion it causes. The inventory is a bill of materials plus the fate of each material at end of life. The impact categories are kept to global warming potential, primary energy and diverted tonnage, because those are the three a municipal client will act on.

Applied to the incumbent bin, the inventory is short.

Screening inventory for the bin in service today.
Component	Material	Mass (kg)	Cradle-to-gate GWP (kg CO₂e per kg)	Contribution (kg CO₂e)
Body, lid and wheels	Virgin HDPE	9.20	1.90	17.48
Axle and hinge pins	Zinc-plated steel	0.55	2.30	1.27
Lid gasket	EPDM	0.15	3.20	0.48
Latch and fasteners	Nylon 6,6	0.10	7.60	0.76
Total	—	10.00	—	19.99

Two numbers come out of this and both are needed later. The embodied burden is about 20.0 kg CO₂e per bin, of which 87 per cent sits in the HDPE simply because the HDPE is 92 per cent of the mass. The observed service life on the municipality's replacement records is six years, dominated by lid-hinge and latch failures rather than by wear of the body, so the annualised burden is

$$\text{annualised GWP} = \frac{19.99\ \text{kg CO}_2\text{e}}{6\ \text{yr}} = 3.33\ \text{kg CO}_2\text{e per year}.$$

The societal half of the baseline is measured separately and is the more important of the two. On the study route of 40 000 households, audits show a participation rate of 48 per cent — slightly under half of the bins issued are actually set out with organics in them. The reasons given in the survey are not aesthetic: raccoons and rodents defeat the latch, and the resulting mess and odour are what stop a household using the bin a second time. A baseline is therefore not a single number but a pair: 20.0 kg CO₂e of embodied burden and 48 per cent participation, and part B has to attack both.

Part B — Two design changes that enhance sustainability

Change 1 — make the bin a single polymer. The steel axle, the EPDM gasket and the nylon latch together are only 8 per cent of the mass but they are 100 per cent of the reason the bin cannot be granulated and re-moulded at end of life; a recycler will not hand-strip a metal axle out of 40 000 carts. The change is to design them out: a moulded-in live hinge in place of the pinned lid, a snap-fit wheel retention in place of the through axle, a compression bead moulded into the lid rim in place of the gasket, and a latch moulded as part of the lid. The polymer is then specified as HDPE with 50 per cent post-consumer recycled content, which the application can carry because the bin is a non-food, non-pressure, thick-walled moulding with generous tolerance on colour and impact.

Change 2 — make the bin defeat wildlife. A gravity latch that engages under the bin's own lid weight and releases only when the lid handle is lifted vertically, in combination with a continuous compression seal at the lid rim, addresses the failure mode the survey identified. This is a societal change delivered by a technological means: it does nothing to the material burden and everything to whether the product performs its function at all.

The two changes are not independent, and that is why they were chosen together. The moulded-in latch of change 1 is the natural place to embody the gravity mechanism of change 2, and the moulded rim bead that replaces the gasket is the seal that change 2 needs. One tool change buys both.

Part C — Impact on long-term sustainability

The two changes have to be tested against the baseline of part A, and they behave very differently, which is the finding worth reporting.

Change 1 acts on the embodied burden. With a recycled fraction $f = 0.50$, virgin HDPE at 1.90 and recycled HDPE at 0.45 kg CO₂e per kg, the blended factor is

$$\text{GWP}_{\text{blend}} = f\,\text{GWP}_{\text{rec}} + (1-f)\,\text{GWP}_{\text{virgin}} = 0.50(0.45) + 0.50(1.90) = 1.175\ \text{kg CO}_2\text{e per kg}.$$

Designing the fasteners out costs a little material back — the live hinge and the moulded latch need local thickness — so the mono-material bin masses 9.60 kg rather than 9.20 kg. Its embodied burden is $9.60 \times 1.175 = 11.28$ kg CO₂e, a saving of 8.71 kg or 43.6 per cent. More importantly, removing the two components that caused the failures takes the service life from six years to twelve, so on the functional unit that actually matters

$$\boxed{\ \text{annualised GWP}: 3.33 \rightarrow 0.940\ \text{kg CO}_2\text{e per year} \quad (71.8\ \text{per cent lower}).\ }$$

Note how much of that improvement is durability rather than material substitution. Recycled content alone would have bought 43.6 per cent; doubling the life is what turns it into 71.8 per cent. This is the general lesson of design for the environment and it is worth stating explicitly in an answer: lifetime is usually a larger lever than material choice, and it is the lever a mechanical designer controls directly.

Change 2 acts somewhere else entirely, and it dwarfs change 1. The route generates $40\,000 \times 0.30 = 12\,000$ tonnes of household organics a year. At 48 per cent participation, 5 760 t are diverted; if the wildlife-resistant latch lifts participation to 68 per cent, 8 160 t are diverted, an increase of 2 400 t per year. Organics kept out of landfill avoid roughly 0.35 t CO₂e per tonne in avoided methane, so

$$\boxed{\ \text{operating benefit} = 2\,400 \times 0.35 = 840\ \text{t CO}_2\text{e per year}.\ }$$

Set that beside change 1: the embodied saving across the whole route is $40\,000 \times 8.71/1000 = 348$ t CO₂e, and that is a one-off saving spread over a twelve-year life, or about 29 t per year. The latch is therefore worth roughly thirty times the recycled content. The long-term consequence for the design direction is unambiguous: every subsequent trade-off in this project is resolved in favour of whatever keeps households using the bin, and material improvements are taken only where they do not compromise that. An answer that optimised the polymer and left the latch alone would have improved the least important number on the page.

Part D — Engineering specifications to implement the change

Specifications turn the intent of part B into numbers a supplier can be held to and a test lab can check. Each row below states the metric, the target and, critically, how it will be verified; a specification with no verification method is a wish. Three of the targets are derived rather than asserted, and the derivations are given underneath.

Target specifications for the revised green bin.
Metric	Target	Basis	Verification
Nominal capacity	120 L	Two-week set-out for a four-person household	Water fill to the seal line
Design payload	55 kg	120 L at 0.40 kg/L organics, plus 15 per cent margin	Static load, 24 h, no permanent set
Gross lifted mass	64.8 kg	Payload plus 9.60 kg bin	Weigh loaded cart
Lid-opening force	≤ 25 N	Under half the 55 N fifth-percentile one-hand pull	Force gauge at the handle, ten cycles
Latch release force	≤ 20 N	Same basis; must still resist a raccoon's pull	Force gauge, plus wildlife-resistance trial
Recycled content	≥ 50 per cent post-consumer	Change 1	Supplier declaration plus resin traceability
Single-polymer fraction	≥ 99 per cent by mass	Change 1; no metal, no elastomer	Full teardown and weigh
Comb-lift bar height	1000 ± 10 mm	EN 840 tipper interface on the existing fleet	Gauge check against the truck comb
Service temperature	−30 to +50 °C	Canadian climate envelope	Conditioned drop and latch cycling at both limits
Drop resistance	1.2 m onto concrete at −30 °C, loaded	Winter set-out abuse	Three drops, no crack, latch still functions
Weathering	1500 h QUV, ΔE ≤ 3	Ten summers of UV	Accelerated weathering coupon
Design life	≥ 12 yr	Part C; the number the whole business case rests on	Accelerated hinge and latch cycling to 12 yr equivalent

The payload derivation is the only one with any arithmetic in it. Loose household organics run about 0.40 kg per litre, so a full 120 L bin carries $120 \times 0.40 = 48$ kg; adding a 15 per cent margin for wet yard waste and the occasional frozen block gives $48 \times 1.15 = 55.2$ kg, specified as 55 kg. The operating forces are set from anthropometric data rather than from feel: the fifth-percentile adult female one-hand pull is about 55 N, and CAN/CSA-B651 practice is to leave at least a factor of two, which is where the 25 N lid target comes from. The comb-bar height is not negotiable at all — it is fixed by the trucks the municipality already owns, and a bin that misses it is worthless however sustainable it is. That row is a reminder that a sustainability specification lives inside a set of interface constraints it does not get to change.

Part E — A methodology for comparing the design alternatives

The comparison is done in two passes, which is the standard concept-selection sequence and is used because the two passes answer different questions.

The first pass is Pugh screening. The incumbent bin is made the datum, every concept is scored against it on each criterion as better (+), same (0) or worse (−), and the net score is taken. Screening is deliberately coarse: its purpose is to eliminate, to expose concepts that are worse than what already exists, and to suggest hybrids, and it needs no weights at all. A concept with a negative net score does not go forward.

The second pass is weighted objectives scoring. Criteria are given weights $w_i$ that sum to unity, surviving concepts are rated $r_{ij}$ on a one-to-five scale, and the total for concept $j$ is

$$S_j = \sum_i w_i\, r_{ij}, \qquad \sum_i w_i = 1.$$

Two disciplines make the difference between this being a decision tool and being a way of dressing up a decision already taken. First, the weights are agreed and written down before any concept is rated, and they are agreed with the client, because the weights are a statement of what the municipality values and are not an engineering judgement. Second, a sensitivity pass is run afterwards: the weights are deliberately shifted to represent a different but defensible view of the project, and the ranking is recomputed. If the winner survives, the recommendation is robust and can be defended. If the ranking inverts, the honest conclusion is that the decision turns on the weighting rather than on the engineering, and that has to be escalated to the client rather than buried.

The three architectures generated for the revised green bin. They differ in architecture, not in styling, which is what makes a selection exercise meaningful.

Part F — Applying the methodology: rank and select

Given. Three concepts — C1, a 120 L two-wheel roll-out cart with a moulded-in hinged lid and gravity latch; C2, an 80 L hand-carried nesting tote with a clip-on lid; C3, a structural frame carrying a swappable rigid liner under a separate lockable lid. Six weighted criteria, agreed with the municipality before rating, and ratings on a one-to-five scale where five is best.

Find. The ranking of the three concepts, the selected concept, and whether the selection is robust to a defensible change in the weighting.

Approach. Screen with Pugh against the incumbent bin as datum, then score the survivors on the weighted matrix, then re-weight to triple the importance of unit cost and see whether the ranking holds.

Screen the three concepts against the incumbent bin. Scoring each of the six criteria as better, same or worse than today's cart gives C1 four pluses, one same and one minus, for a net of $+3$; C2 two pluses, one same and three minuses, for a net of $-1$; and C3 three pluses, one same and two minuses, for a net of $+1$. C2 is worse than the datum overall — an 80 L tote that has to be carried to the curb loaded is a step backwards on exactly the handling criterion the project exists to improve — so screening would normally stop it here. It is carried into the scoring pass anyway, because the municipality asked for a hand-carried option to be costed for apartment buildings, and it is more useful to show why it loses than to drop it silently.
Fix the weights, with the client, before any concept is rated. Wildlife resistance and odour control takes 0.22 because part C showed participation is the dominant lever; handling effort and accessibility 0.20; mono-materiality and recyclability 0.18; durability over the twelve-year life 0.15; collection-fleet compatibility 0.15; and unit manufacturing cost 0.10, low because the cart is a twelve-year municipal asset rather than a retail purchase. The weights sum to 1.00.
Rate the concepts and compute the weighted totals. Applying $S_j = \sum_i w_i r_{ij}$ to the matrix below gives C1 4.43, C2 3.19 and C3 3.61. Working the winning column out in full as a check: $0.22(4) + 0.20(5) + 0.18(5) + 0.15(4) + 0.15(5) + 0.10(3) = 0.88 + 1.00 + 0.90 + 0.60 + 0.75 + 0.30 = 4.43$.
Rank, and record the margin. The order is C1, then C3, then C2, and the margin between first and second is $4.43 - 3.61 = 0.82$ on a five-point scale — wide enough that no plausible re-rating of a single cell overturns it.
Run the sensitivity pass. A finance-led view of the same project would weight unit cost far higher. Tripling the cost weight from 0.10 to 0.30 and taking the difference out of wildlife resistance (0.22 to 0.12) and fleet compatibility (0.15 to 0.05) gives C1 4.13, C2 3.79 and C3 3.11. C1 still leads, so the selection is robust, but two things change and both are worth reporting: the margin more than halves, from 0.82 to 0.34, and the runner-up swaps from C3 to C2, because the cheap tote is the concept that benefits most from a cost-driven view.

Weighted concept scoring. The last column spells out the winning concept row by row.
Criterion	Weight	C1 roll-out cart	C2 nesting tote	C3 frame and liner	C1 weighted
Wildlife resistance and odour control	0.22	4	2	5	0.88
Handling effort and accessibility	0.20	5	3	3	1.00
Mono-materiality and recyclability	0.18	5	5	2	0.90
Durability over a 12-year life	0.15	4	3	5	0.60
Collection-fleet compatibility	0.15	5	2	4	0.75
Unit manufacturing cost	0.10	3	5	2	0.30
Weighted total	1.00	4.43	3.19	3.61	4.43

Weighted totals under the agreed weights and under the cost-dominated sensitivity weights. The winner is unchanged; the runner-up is not.

Selection. Concept C1, the 120 L mono-material roll-out cart with a moulded live hinge and an integral gravity latch, is selected. It carries both of the part B changes, it is the only concept that both defeats wildlife and remains a single polymer, and it is the only one that meets the fixed comb-lift interface without modifying the collection fleet. Its one weakness — it is the most expensive of the three to make — is real, and the sensitivity pass quantifies what it would take for that weakness to matter, which is the answer the client actually needs.

Question 1 — results.
Quantity	Result
Baseline embodied burden	19.99 kg CO₂e per bin
Baseline annualised burden (6-year life)	3.33 kg CO₂e per year
Revised embodied burden (50 per cent recycled HDPE)	11.28 kg CO₂e per bin
Revised annualised burden (12-year life)	0.940 kg CO₂e per year, 71.8 per cent lower
Extra organics diverted on the study route	2 400 t per year
Operating benefit of the wildlife-resistant latch	840 t CO₂e per year
Embodied benefit of change 1 across the route	348 t CO₂e once, about 29 t per year
Design payload / gross lifted mass	55 kg / 64.8 kg
Weighted scores C1 / C2 / C3	4.43 / 3.19 / 3.61
Ranking, agreed weights	C1, C3, C2 (margin 0.82)
Ranking, cost-dominated weights	C1, C2, C3 (margin 0.34)
Selected concept	C1 — 120 L mono-material roll-out cart with integral gravity latch

Topic. Design for the environment inside a defined design process, and concept selection by Pugh screening followed by weighted objectives.

Key relations. Blended impact factor $\text{GWP}_{\text{blend}} = f\,\text{GWP}_{\text{rec}} + (1-f)\,\text{GWP}_{\text{virgin}}$; burden on the functional unit $= \text{embodied}/\text{life}$; weighted concept score $S_j = \sum_i w_i r_{ij}$ with $\sum_i w_i = 1$; Pugh net score = (number of pluses) − (number of minuses) against a datum.

Why this works. A life-cycle assessment answers a question only after the functional unit fixes what is being compared, and here the choice of "one household served for one year" rather than "one bin" is what lets a doubling of service life show up as a benefit at all. Weighted scoring works for the same structural reason: the weights encode the client's values and the ratings encode the engineering, and keeping them apart is what stops one being used to smuggle in the other. The sensitivity pass is the honesty check on the whole method — it costs one recomputation and it tells you whether you are reporting an engineering result or a weighting choice.

Common pitfall. Students optimise the material and never touch the use phase. Here the recycled content is worth about 29 t CO₂e a year across the route and the latch is worth 840 t, and a candidate who spent the whole answer on polymer selection would have improved the smaller number by a wide margin. The second pitfall is rating concepts before agreeing weights, which lets the weights be tuned until the preferred concept wins; agree and record them first.

Source. Ulrich & Eppinger, Product Design and Development, ch. 8 (concept selection: screening and scoring matrices) and ch. 12 (design for the environment); Dieter & Schmidt, Engineering Design, ch. 6 (decision making) and ch. 8 (life-cycle considerations); ISO 14040/14044 for the assessment framework; CAN/CSA-B651 for the operating-force limits.

Question 1 of 7: The design process applied to a sustainable green bin

Reference texts for this subject