Synthetic port scene with a container truck, freight train and cargo ship beside gantry cranes under low sunlight.
Conceptual transport modes, not a measured consignment, actual transfer record or verified emissions result.
Diagram text description

An unbranded container truck is at the left of a port scene, a container freight train runs along tracks through the centre, and a cargo ship sits beneath cranes beside the water. There are stacked containers, distant equipment and low sunlight. No shipment identifier, carrier name, cargo weight, route, distance, emissions figure or assurance statement is legible. Depicting these modes together does not establish a real connected journey, fuel performance, customer software capability or functioning carrier data integration.

A logistics emissions calculation becomes useful when the reader can see which movements it covers, how each result was obtained and what remains missing. Start with the physical shipment chain, not a single factor applied to an undifferentiated journey. Road collection, rail haulage, port handling, a sea crossing and final delivery can have different carriers, activity records and allocation methods.

The workflow is straightforward: map the chain and facilities; gather evidence for each operation; choose matching methods and factors; align units, gases and fuel boundaries; then add compatible, non-overlapping results. Keep unknown activities visible and retain the versions needed to reproduce or revise the total. An attractive calculator output cannot supply those missing decisions.

This guide follows a fictional consignment through that process. Its numerical coefficients are invented teaching inputs, not factors for real reporting. The examples distinguish a transport subtotal from a wider service result, then show why a separate facility addition and inventory-classification decision are still necessary.

Map the chain before calculating

Assign one shipment identifier, then give every movement and facility operation its own reference. Record origin, destination, service period, mode, carrier and the reporting organisation's relationship to that service. Check collection and delivery against booking and receipt records: a port-to-port quote may omit both the first and last mile.

In the illustrative R17 chain, L1 is road collection from Factory A to Depot B; L2 is rail from B to Port C; H1 is handling and temperature-controlled dwell at C; L3 is the sea crossing to Port D; L4 is road delivery to Customer E. H1 is a facility operation, not another tonne-kilometre leg. The diagram keeps those references attached to one chain, rather than implying that changing carrier creates a new shipment.

R17 follows road L1, rail L2, port facility H1, sea L3 and final delivery L4, each with its own reference. Sea and final-delivery calculations remain missing.
One illustrative R17 chain, May 2026. H1 is a facility operation; L3 and L4 are mapped but not yet calculated.
Diagram text description

Five connected labelled boxes map one illustrative shipment. The upper row runs left to right from Road L1 through Rail L2 to Port H1. The path descends to Sea L3, then moves left to Delivery L4. The port box identifies a distinct facility operation rather than another distance-based leg. The path records continuity and handovers, not geographical scale, emissions magnitude or proof of actual movement. Only road, rail and a limited hub-electricity contribution have invented arithmetic in the article; the sea and final-delivery data and refrigerant leakage remain unresolved. All examples are fictional, not verified service results.

Reconcile handover points before calculating. Who includes terminal movements, loading, transfers and storage? A carrier's broad “door-to-door” label is not sufficient evidence of coverage. Where R17's sea and delivery records have not arrived, keep L3 and L4 in the register as missing. Removing their rows would make an incomplete calculation look complete; duplicating a transfer in two providers' records would overstate it.

Choose an activity method for each leg

Choose a method separately for each leg, according to the available evidence and the decision the result supports. Traceable operational information can reduce modelling assumptions, but only if the carrier explains its covered service, allocation and emissions boundary. A fuel total for an entire vehicle is not automatically the footprint of one customer's load.

MethodEvidence neededMain qualification
Allocated carrier operationsRelevant fuel or electricity records, service coverage and a defensible shipment shareEstablish which operations and gases are included; retain the allocation denominator
Mass-distance modelShipment mass, relevant route distance and an appropriate mode-specific factorFactor assumptions about utilisation, vehicle type and empty running affect representativeness
Spend screenTransport expenditure, matching economic factor, currency and price-year treatmentPrice changes and bundled services can alter the estimate without changing physical activity

Table 1, shown one row at a time. Every field is retained.

Row 1

Method
Allocated carrier operations
Evidence needed
Relevant fuel or electricity records, service coverage and a defensible shipment share
Main qualification
Establish which operations and gases are included; retain the allocation denominator

Row 2

Method
Mass-distance model
Evidence needed
Shipment mass, relevant route distance and an appropriate mode-specific factor
Main qualification
Factor assumptions about utilisation, vehicle type and empty running affect representativeness

Row 3

Method
Spend screen
Evidence needed
Transport expenditure, matching economic factor, currency and price-year treatment
Main qualification
Price changes and bundled services can alter the estimate without changing physical activity

GHG Protocol's transport guidance describes fuel-, distance- and spend-based approaches; spend is particularly qualified as a screening route. Do not treat a more detailed-looking spreadsheet as better evidence simply because it contains more digits. For an important recurring lane, identify what better carrier information would replace the current estimate and who will request it. Different methods across a chain are possible, but their outputs still need compatible boundaries before aggregation.

Keep tonne-kilometres and factors compatible

Match the activity units to the factor before doing multiplication. In a fictional illustration, a packaged load of 500 kilograms is 0.5 tonne. Moving it 100 kilometres gives 0.5 × 100 = 50 tonne-kilometres. A factor expressed as kilograms CO2e per tonne-kilometre can then multiply that activity directly: the tonne-kilometre units cancel, leaving kilograms CO2e.

Using 500 as though it were tonnes introduces a thousandfold mass error. Equally, a factor per vehicle-kilometre, container-kilometre or already allocated shipment is not interchangeable with one per tonne-kilometre. Use the activity driver the factor actually requires; do not multiply an allocated shipment result by mass and distance again.

State whether mass includes packaging and pallets and whether it changes between legs. Use relevant travelled distances when available, retaining assumptions where routing is modelled. Check whether the selected factor already incorporates average loading, empty running or return-trip treatment before adding another correction. For bulky or containerised freight, a volume-related allocation may be more appropriate; cargo weight alone does not establish the right denominator for every vehicle or vessel.

Align the gas, life-cycle and service boundary

Align three separate boundaries: the greenhouse gases represented, the energy life cycle covered and the physical service included. CO2 counts carbon dioxide; CO2e combines specified gases using a stated global-warming-potential basis. Converting kilograms to tonnes does not convert a CO2-only result into CO2e.

Fuel-use emissions describe the use stage. A wider fuel-life-cycle estimate also includes upstream fuel production and supply. The fuel-use CO2 component can sit within that broader CO2e total, but it is not an extra quantity to add to it. Likewise, a full fuel-life-cycle factor for a sea leg does not automatically include a road pickup, port facility or final delivery. Fuel boundary and service coverage answer different questions.

Record factor vintage, geographical and technological fit, gases and warming-potential basis alongside each leg. Where a provider supplies a differently bounded number, obtain a compatible replacement or retain it as a separately labelled alternative. Do not bridge missing gases or stages with an unexplained percentage uplift. The worked subtotal below assumes compatible fictional fuel-life-cycle CO2e inputs; it does not establish current GLEC or ISO conformity.

Calculate leg results and sum only compatible totals

Fictional worked calculation: R17 carries 0.5 tonne of packaged temperature-sensitive components in May 2026. L1 covers 100 kilometres by road; L2 covers 800 kilometres by rail. Distances, load, period, carriers and every coefficient below are invented. The example demonstrates reproducible arithmetic, not measured emissions or published modal performance.

Assume fictional factor register FICT-F1 defines road coefficient R as 0.12 kg CO2e per tonne-kilometre and rail coefficient L as 0.02 kg CO2e per tonne-kilometre. Both cover energy supply plus use on the same stipulated gas and warming-potential basis. They already contain the example's assumed utilisation and empty-running treatment. Neither includes H1 facilities, other transport legs, vehicle manufacture or infrastructure construction.

OperationActivity transformationFictional matching factorResult
L1 road collection0.5 tonne × 100 km = 50 tonne-kmR = 0.12 kg CO2e/tonne-km50 × 0.12 = 6 kg CO2e
L2 rail haulage0.5 tonne × 800 km = 400 tonne-kmL = 0.02 kg CO2e/tonne-km400 × 0.02 = 8 kg CO2e
Compatible transport subtotalAdd the two emissions results, not unlike activity driversSame stipulated fuel-life-cycle CO2e basis6 + 8 = 14 kg CO2e

Table 2, shown one row at a time. Every field is retained.

Row 1

Operation
L1 road collection
Activity transformation
0.5 tonne × 100 km = 50 tonne-km
Fictional matching factor
R = 0.12 kg CO2e/tonne-km
Result
50 × 0.12 = 6 kg CO2e

Row 2

Operation
L2 rail haulage
Activity transformation
0.5 tonne × 800 km = 400 tonne-km
Fictional matching factor
L = 0.02 kg CO2e/tonne-km
Result
400 × 0.02 = 8 kg CO2e

Row 3

Operation
Compatible transport subtotal
Activity transformation
Add the two emissions results, not unlike activity drivers
Fictional matching factor
Same stipulated fuel-life-cycle CO2e basis
Result
6 + 8 = 14 kg CO2e

The general relationship is 50 × R + 400 × L. For a real shipment, source and qualify R and L separately; do not use the invented values above. Adding activities first and applying one average factor would erase the mode-specific calculation. Keep intermediate values and convert a result in grams to kilograms before summing if necessary.

A separate hypothetical carrier note reports 2 kg CO2 from combustion for L1. That is a non-additive alternative covering the same road service on a narrower gas and fuel boundary: do not add it to either the 6 kg road estimate or the 14 kg subtotal. It is also not a directly comparable replacement without boundary reconciliation. H1 electricity and leakage are outside these transport factors and are treated separately below.

Open decision: obtain the L3 sea and L4 final-delivery activities and compatible factors, or explicit allocated carrier results, before calculating the complete chain. Consequence: report 14 kg CO2e only as the L1–L2 transport subtotal; missing legs remain unresolved, not zero.

Treat facilities and cold chain as explicit additions

Fictional facility addition: R17 is the same 0.5-tonne shipment in May 2026, including L1 road and L2 rail above. Assume H1's temperature-controlled zone uses 60 kWh during the relevant two-hour dwell window. R17 occupies 0.2 cubic metre of an average 4 cubic metres stored over that same window. A stipulated equal-service volume-time allocation gives 0.2 ÷ 4 = 5%, so its allocated electricity is 60 × 5% = 3 kWh.

Assume a separate fictional electricity coefficient of 0.40 kg CO2e/kWh includes generation and upstream energy supply on the same gas basis as the example transport factors. H1 electricity is then 3 × 0.40 = 1.2 kg CO2e. This allocation assumes the same storage duration and temperature service for the represented stock; it is not a universal warehouse rule or evidence from an actual meter.

Included operationReproducible fictional transformationContribution
L1 road0.5 tonne × 100 km × 0.12 kg CO2e/tonne-km6 kg CO2e
L2 rail0.5 tonne × 800 km × 0.02 kg CO2e/tonne-km8 kg CO2e
H1 allocated electricity60 kWh × 5% × 0.40 kg CO2e/kWh1.2 kg CO2e
Known-operation subtotal6 + 8 + 1.215.2 kg CO2e; still incomplete

Table 3, shown one row at a time. Every field is retained.

Row 1

Included operation
L1 road
Reproducible fictional transformation
0.5 tonne × 100 km × 0.12 kg CO2e/tonne-km
Contribution
6 kg CO2e

Row 2

Included operation
L2 rail
Reproducible fictional transformation
0.5 tonne × 800 km × 0.02 kg CO2e/tonne-km
Contribution
8 kg CO2e

Row 3

Included operation
H1 allocated electricity
Reproducible fictional transformation
60 kWh × 5% × 0.40 kg CO2e/kWh
Contribution
1.2 kg CO2e

Row 4

Included operation
Known-operation subtotal
Reproducible fictional transformation
6 + 8 + 1.2
Contribution
15.2 kg CO2e; still incomplete

The assumed transport factors exclude H1, so this electricity line is a distinct addition, not a second count of transport energy. If an actual carrier's chain total already included the facility, use the supported complete result or a reconciled component breakdown, not both. Onboard refrigeration can likewise be embedded in vehicle energy; establish its treatment before adding a generic cold-chain uplift.

Refrigerant leakage is another possible activity, not electricity consumption. It needs evidence of leakage, refrigerant identity, the appropriate emissions basis and a defensible allocated share. The example has no such evidence. Open decision: ask H1's operator for the leakage record and allocation, and obtain the outstanding L3–L4 data. Consequence: retain leakage and those legs as unresolved gaps; 15.2 kg CO2e is not the complete cold-chain or door-to-door footprint.

Compare options without changing the service

Compare proposals only after defining the service that both must deliver. Hold the consignment, final endpoint, covered activities, temperature conditions and feasible delivery window constant. Then align the gas and energy boundaries, allocation and data quality. A slower route that fails the product's service needs is not automatically a usable lower-emissions alternative.

For R17, one proposal might show a small port-to-port sea figure while another covers collection, terminal activity, sea transport and final delivery. Those headline numbers cannot rank the complete services. First restore the omitted activities or narrow the comparison explicitly to the common segment. Do not quietly assume that extra storage or refrigeration is identical when schedules differ.

After establishing comparability, test practical changes such as consolidation, routing or mode choice against actual loading and handling constraints. Show both emissions and service outcomes, and keep material uncertainty beside the recommendation. If the better-looking option depends on an unsupported load factor or unconfirmed connection, make obtaining that evidence part of the decision rather than declaring a quantified saving.

Record assumptions so another analyst can reproduce the total

Keep an activity-and-factor register that lets another analyst reproduce the result without reconstructing the shipment from emails. Store original carrier documents, allocated results and factor references alongside the calculation version. Separate inputs supplied by a carrier from estimates introduced by the analyst, and identify the reviewer responsible for checking that distinction.

Register groupMinimum useful record
Identity and serviceShipment and leg/facility references, carrier, reporting entity, period, origin, destination and included connections
ActivityMass or volume basis, packaging treatment, travelled or assumed distance, meter/fuel data and units
Factor and allocationSource, edition, geography, technology, gases, energy boundary, warming-potential basis and shipment-share denominator
Limitations and reviewMissing activities, exclusions, uncertainty, reviewer, unresolved findings and supported use
Version controlCalculation version, change reason, update trigger, owner and affected recipients

Table 4, shown one row at a time. Every field is retained.

Row 1

Register group
Identity and service
Minimum useful record
Shipment and leg/facility references, carrier, reporting entity, period, origin, destination and included connections

Row 2

Register group
Activity
Minimum useful record
Mass or volume basis, packaging treatment, travelled or assumed distance, meter/fuel data and units

Row 3

Register group
Factor and allocation
Minimum useful record
Source, edition, geography, technology, gases, energy boundary, warming-potential basis and shipment-share denominator

Row 4

Register group
Limitations and review
Minimum useful record
Missing activities, exclusions, uncertainty, reviewer, unresolved findings and supported use

Row 5

Register group
Version control
Minimum useful record
Calculation version, change reason, update trigger, owner and affected recipients

For R17, the register must preserve missing L3, L4 and leakage entries alongside the calculated lines. When better data arrives, issue a new version and explain which assumptions it replaces. Check whether an apparent reduction reflects operational change or merely a revised factor, boundary or allocation rule; retain enough history to tell the difference.

Classify the result in the right inventory

Classify the physical result for a named reporting company rather than assigning one universal scope to the journey. A carrier burning fuel in its own controlled vehicle normally records that combustion in scope 1. Purchased electricity for its controlled electric operations belongs in scope 2. A customer purchasing the third-party service addresses a different organisational relationship, usually within its relevant scope 3 accounting.

GHG Protocol's category guidance includes purchased third-party transport services in Category 4, including outbound services purchased by the reporting company. Category 9 addresses relevant downstream transport not paid for by that company. Physical direction alone therefore does not settle the category, and leased operations or transport already embedded in purchased-product inventories need separate consideration.

Keep the corporate inventory boundary distinct from the broader energy-life-cycle service subtotal. A total containing upstream energy contributions cannot simply be copied into a category defined around a provider's allocated operational emissions without checking the required treatment and possible overlap elsewhere. The same emissions can appear in different companies' inventories through their relationships; that is different from counting one activity twice within the same reporting company's total.

Use the baseline to improve the next shipment

Use the baseline to choose a manageable next improvement, not merely to generate a larger reporting file. Prioritise recurring lanes where emissions, service risk or evidence gaps matter. Ask carriers for the specific missing inputs that would change a decision, such as allocated fuel data, actual routing or an explanation of a transfer already included in their total.

Test feasible changes against a stable service definition. Better consolidation can change utilisation; a revised mode can add handling or storage; improved delivery reliability can avoid repeat movements. Recalculate the affected chain rather than assuming that a favourable factor for one mode proves a better end-to-end outcome. Preserve temperature and delivery performance alongside the emissions comparison.

For R17, completing L3, L4 and the facility evidence comes before announcing a whole-chain saving. The next repeat shipment can then be compared using aligned boundaries and documented source quality. If better records change the estimate, explain that evidence improvement separately from a physical reduction so that procurement does not reward a reporting artefact.

Connect the calculation to product-data records

Connect logistics records to the product or supplier references that the receiving analyst actually uses. An organised bill of materials, SKU revision and supplier record can identify which shipment data belongs with a product study. They do not supply a missing carrier allocation, choose a freight factor or establish that a shipment total has been independently assured.

Prepare one neutral handoff: the product reference, relevant shipment identifiers, calculation and source versions, coverage statement, unresolved inputs and responsible contacts. Check whether a product inventory already contains some transport before importing another logistics subtotal. This prevents an apparently helpful data transfer from duplicating activities or mixing a component boundary with a complete delivered-product boundary.

Before relying on any software or service, demonstrate that its actual functions preserve those relationships and limitations. Confirm calculation scope, supported exchange fields and any assurance engagement directly; a structured-data description is not proof of a freight calculator or carrier integration. The finished decision should be traceable from physical activity to qualified result, without asking a product record to promise more than its evidence can support.