Chapter 7 of 8

Sustainability Measurement in Logistics

3 sections · ~9 min read · 3 practice questions · part of Supply Chain Analytics

Learning objectives

  • Estimate freight emissions from activity data using tonne-kilometres and an emission factor, and state the units at every step
  • Distinguish an intensity metric from an absolute total and explain when each one misleads
  • Compute a marginal abatement cost per tonne of carbon dioxide equivalent against the correct baseline

Activity data times an emission factor

Freight emissions estimates almost always follow the same structure: an activity quantity multiplied by an emission factor. For transport, the usual activity measure is the tonne-kilometre, the product of freight mass carried and distance travelled, and the factor is expressed in kilograms of carbon dioxide equivalent per tonne-kilometre. Carbon dioxide equivalent is a combined measure that converts several greenhouse gases onto a common basis using warming potentials, which is why results carry the suffix equivalent rather than being reported as carbon dioxide alone. Unit discipline is everything in this calculation. Mixing metric tonnes with short tons, kilometres with miles, or grams with kilograms produces errors of a factor of two or more that are invisible in the final number because the result still looks plausible. Write the units on every line and confirm they cancel. The factor itself is the weakest link: it depends on vehicle type, fuel, terrain, load factor, and empty running, and any factor used should be recorded with its source and vintage rather than treated as a constant of nature. Because factors are uncertain, comparisons between options computed with the same factor set are far more reliable than absolute totals, which is a strong argument for using emissions analysis to rank alternatives rather than to make absolute claims.

Intensity versus absolute, and the load factor trap

An intensity metric divides emissions by a unit of output, such as grams per tonne-kilometre or kilograms per unit shipped. An absolute metric is the total for the period. They answer different questions and can move in opposite directions: a firm that grows volume 30 percent while improving intensity 20 percent has reduced its intensity and increased its total. Both belong on a report, because intensity measures operating efficiency while the absolute total measures actual atmospheric impact. A specific trap deserves attention. When the emission factor is expressed per tonne-kilometre, the calculation is blind to how full the vehicle is, because both the numerator and denominator scale with the load. Under such a factor, one full truck and two half-full trucks carrying the same total freight produce identical estimated emissions, which is plainly wrong, since the two half-full trucks burn roughly twice the fuel. Capturing load factor requires a per-vehicle-kilometre factor instead, applied to the distance each vehicle actually travels including empty return legs. Whenever a proposed initiative is about consolidation, filling trailers, or reducing empty running, a tonne-kilometre model cannot measure the benefit and a vehicle-kilometre model must be used.

Marginal abatement cost and the baseline problem

To compare emission-reduction options on a common footing, compute the marginal abatement cost: the incremental cost of an option divided by the tonnes of carbon dioxide equivalent it avoids, expressed in currency per tonne. Ranking options from cheapest to most expensive builds an abatement cost curve that shows how far the firm can go at what price. Options with a negative abatement cost save money and reduce emissions at once, and they should be executed before anything is spent on the rest of the curve. The critical methodological point is the choice of baseline. Abatement cost must be measured against the option the firm would otherwise choose, not against the worst available option. Comparing an expensive low-carbon alternative to a high-carbon incumbent, when a cheaper mid-carbon option is already available and preferable, credits the expensive option with reductions it did not cause and can make a poor investment look attractive. The correct sequence is to identify the best option on the existing decision rule, adopt it as the baseline, and then price every further step against that. Costs and reductions should also be quoted over the same period, typically one year at the actual shipment frequency, so that a per-shipment saving is not silently compared to an annual cost.

Worked example

Problem

Marlow Instruments moves 18 tonnes of finished goods 1,450 kilometres on a lane it runs 240 times per year. Option A is a single truck for the full distance, costing 2,650 dollars per shipment, assumed emission factor 0.095 kilograms of carbon dioxide equivalent per tonne-kilometre. Option B is intermodal: 1,380 kilometres by rail at an assumed 0.028 kilograms per tonne-kilometre plus 120 kilometres of truck drayage at 0.095, costing 2,380 dollars per shipment. Option C is a low-carbon fuel service costing 3,100 dollars per shipment with assumed emissions of 350 kilograms per shipment. All factors are illustrative values used for comparison, not measured figures. Compute emissions per shipment for each option, the intensity in grams per tonne-kilometre for A and B, the annual difference between A and B, and the marginal abatement cost of B and of C.

Step by step

  1. Tonne-kilometres per shipment = 18 tonnes x 1,450 kilometres = 26,100 tonne-kilometres.
  2. Option A emissions = 26,100 tonne-kilometres x 0.095 kilograms per tonne-kilometre = 2,479.5 kilograms = 2.4795 tonnes of carbon dioxide equivalent per shipment.
  3. Option B rail leg = 18 tonnes x 1,380 kilometres = 24,840 tonne-kilometres, times 0.028 = 695.52 kilograms.
  4. Option B drayage leg = 18 tonnes x 120 kilometres = 2,160 tonne-kilometres, times 0.095 = 205.2 kilograms. Note that Option B's two legs total 1,500 kilometres against the 1,450-kilometre road distance, which is deliberate: rail routings are rarely as direct as the highway, and that extra circuity is distance the freight genuinely travels and must be charged emissions for.
  5. Option B total = 695.52 + 205.2 = 900.72 kilograms = 0.9007 tonnes per shipment.
  6. Reduction from A to B = 2,479.5 - 900.72 = 1,578.78 kilograms = 1.5788 tonnes per shipment, a 1,578.78 / 2,479.5 = 63.67 percent reduction.
  7. Intensity, using the same 26,100 tonne-kilometre denominator for both since the freight task is identical: Option A = 2,479.5 kilograms x 1,000 grams per kilogram / 26,100 = 95.00 grams per tonne-kilometre, which simply recovers the factor. Option B = 900.72 x 1,000 / 26,100 = 34.51 grams per tonne-kilometre.
  8. Annual figures at 240 shipments: reduction = 1.5788 tonnes x 240 = 378.91 tonnes of carbon dioxide equivalent per year. Cost change = (2,380 - 2,650) x 240 = negative 270 x 240 = negative 64,800 dollars, meaning Option B saves 64,800 dollars per year.
  9. Marginal abatement cost of B relative to A = incremental cost divided by tonnes avoided = (2,380 - 2,650) / 1.5788 = negative 270 / 1.5788 = negative 171.02 dollars per tonne. A negative value means the option pays for itself.
  10. Choosing the correct baseline for Option C: because B is both cheaper and cleaner than A, B is what the firm would otherwise choose, so B is the baseline.
  11. Option C reduction versus B = 900.72 - 350 = 550.72 kilograms = 0.5507 tonnes per shipment. Incremental cost = 3,100 - 2,380 = 720 dollars. Marginal abatement cost = 720 / 0.5507 = 1,307.38 dollars per tonne.
  12. For contrast, computing C against the wrong baseline A gives reduction 2,479.5 - 350 = 2,129.5 kilograms = 2.1295 tonnes and cost 3,100 - 2,650 = 450 dollars, for 211.32 dollars per tonne, a figure roughly six times better and entirely misleading.

Answer. Option A emits 2.4795 tonnes per shipment at 95.00 grams per tonne-kilometre; Option B emits 0.9007 tonnes at 34.51 grams per tonne-kilometre, a 63.67 percent reduction. Switching from A to B saves 378.91 tonnes of carbon dioxide equivalent and 64,800 dollars per year, a marginal abatement cost of negative 171.02 dollars per tonne, so it should be executed immediately with no further justification needed. Option C, priced against the correct baseline of B, costs 1,307.38 dollars per tonne of additional reduction. Against the wrong baseline of A it appears to cost only 211.32 dollars per tonne, and the difference between those two figures is entirely an artefact of baseline choice, not of anything the option does. The practical rule is to execute every negative-cost option first, then re-baseline before pricing anything further, because the cheap wins change what the next decision is being compared to.

Practice

Work each question before opening the solution.

  1. The traffic team proposes splitting the 18-tonne load into two 9-tonne truck shipments to improve delivery flexibility. Compute the estimated emissions under the tonne-kilometre factor and explain why the answer is not credible.

    Show solution for question 1

    Each 9-tonne shipment gives 9 x 1,450 = 13,050 tonne-kilometres, times 0.095 = 1,239.75 kilograms, and two shipments total 2,479.5 kilograms, exactly the same as a single full truck. This is not credible because two trucks driving 1,450 kilometres burn roughly twice the fuel of one. The tonne-kilometre factor is load-factor blind: both mass and the denominator halve together, so the model cannot see the change. Measuring this decision requires a per-vehicle-kilometre factor. At an assumed 0.95 kilograms per vehicle-kilometre, one truck emits 0.95 x 1,450 = 1,377.5 kilograms while two emit 2,755 kilograms, and the intensity rises from 1,377,500 grams / 26,100 tonne-kilometres = 52.78 grams per tonne-kilometre at full load to 105.56 grams per tonne-kilometre at half load.

  2. Marlow reports that its freight intensity improved from 95 to 34.51 grams per tonne-kilometre on this lane, while its total logistics emissions rose 12 percent year over year. Explain how both statements can be true and what should be reported.

    Show solution for question 2

    Intensity is a ratio and the total is an absolute. If shipment volume grew faster than intensity improved, the total rises despite genuinely better operations. For example, tripling the freight task while cutting intensity 64 percent gives roughly 3 x 0.363 = 1.09 times the prior total, an increase. Both figures belong in the report, with the volume change stated alongside, because intensity alone lets a firm claim improvement while its actual emissions grow, and the absolute alone gives no credit for real efficiency gains.

  3. A supplier offers a lane option costing 2,500 dollars per shipment with estimated emissions of 1.9 tonnes. Should Marlow take it, and what is its marginal abatement cost?

    Show solution for question 3

    No. Option B already costs less at 2,380 dollars and emits far less at 0.9007 tonnes, so the offer is dominated on both dimensions and needs no abatement calculation. Computing one anyway is instructive: relative to B, the offer costs 120 dollars more and emits 0.9993 tonnes more, so it has a negative reduction and the abatement cost is undefined in any useful sense. An option that is worse on both axes than the current baseline should be screened out before any per-tonne arithmetic, and the screening step is what a properly chosen baseline provides.