LCA requires data on every process in your product system. Material quantities, energy consumption, transport distances, and emissions all need quantification. The challenge is gathering enough data for reliable results without spending months on data collection.
Primary Data for Foreground Processes
Foreground processes are those you directly control or have detailed knowledge about. These need primary data collected from actual operations.
Manufacturing Data
Production records provide most manufacturing data:
- Material inputs (kg, litres, m³)
- Energy consumption (kWh electricity, MJ natural gas, litres diesel)
- Water consumption
- Production yields (output per input)
- Scrap rates and waste generation
- Direct air emissions (if measured)
- Wastewater characteristics
Check utility bills, purchasing records, and production logs. Most manufacturers track this information for cost management. LCA repurposes existing data.
Packaging
Document all packaging materials:
- Primary packaging (immediate container)
- Secondary packaging (boxes, film)
- Tertiary packaging (pallets, stretch wrap)
Include weights and material types. “Cardboard box” isn’t sufficient – specify whether it’s virgin or recycled kraft paper, weights, and corrugation type.
Transport
Transport data needs distances and modes:
- Raw material delivery to your facility
- Product distribution to customers
- Internal logistics between facilities
Calculate tonne-kilometres: multiply mass by distance. Mode matters significantly – air freight creates far higher impacts per tonne-kilometre than sea freight.
Secondary Data for Background Processes
Background processes sit upstream in supply chains or represent generic activities. These typically use secondary data from LCA databases.
Material Production
Databases provide data for common materials:
- Metals (steel, aluminium, copper)
- Plastics (PE, PP, PET, PVC)
- Construction materials (cement, aggregates, timber)
- Chemicals and intermediates
- Packaging materials
Database entries specify geographical regions and production routes. European steel differs from Chinese steel. Primary aluminium differs from recycled aluminium. Select datasets matching your actual supply.
Energy Supply
Electricity impact depends on grid mix:
- UK electricity has lower carbon intensity than coal-heavy grids
- Renewable electricity creates different impacts than fossil generation
- On-site renewable generation needs separate accounting
Use regional grid data matching your production location. If you purchase renewable electricity with Guarantees of Origin, this might affect your carbon calculation but check whether other impact categories change.
Natural gas, diesel, and other fuels also need data representing extraction, processing, and distribution before combustion.
Waste Treatment
End-of-life scenarios require waste management data:
- Landfill processes include methane generation and leachate treatment
- Incineration includes energy recovery credits
- Recycling processes cover collection, sorting, and reprocessing
Match waste treatment data to actual disposal routes. UK waste management differs from scenarios in other regions.
Data Quality Requirements
ISO 14040 defines data quality dimensions:
Temporal Representativeness
How recent is the data? Manufacturing data from five years ago might not represent current production. Energy grids change as renewable capacity increases.
Specify acceptable data age. Current operations need recent data. Historical studies might deliberately use older data to represent specific time periods.
Geographical Representativeness
Does data match your location? European electricity data doesn’t represent Asian grids. Transport distances for global supply chains differ from local production.
Specify geographical scope. Global products might average across production regions. Regional products need local data.
Technological Representativeness
Does data represent actual technology? There might be multiple production routes for a material. Blast furnace steel differs from electric arc furnace steel.
Match technology to reality. If your supplier uses a specific process, seek data for that process rather than industry averages.
Common Data Gaps
Certain data commonly proves difficult to obtain:
Supplier Data
Upstream suppliers might not share detailed environmental data. Strategic materials need requests through procurement. Commodity materials rely on database averages.
Supplier engagement takes time. Start early and use procurement relationships. Frame requests around supply chain transparency rather than singling out specific suppliers.
Transport Details
Actual logistics involve multiple modes and warehousing. Simple point-to-point distance calculations miss complexity.
Track what you can measure. Major transport legs matter more than last-mile variations. A 5% distance error affects total impacts less than choosing air versus sea freight.
Use Phase Data
How customers use products determines use-phase impacts. This varies by user behaviour, maintenance practices, and operating conditions.
Define representative use scenarios. Typical usage patterns provide baseline assessment. Sensitivity analysis shows how variations affect results.
End-of-Life
Actual disposal routes vary by region and product type. Some percentage goes to each disposal method.
Use regional waste statistics for proportional splits. If 60% of plastic packaging gets recycled in your market, model 60% recycling and 40% alternative disposal.
Minimum Data Requirements
You can’t measure everything. Prioritise data collection:
Required Data
These elements must have good data:
- Material quantities for main components
- Energy for primary production processes
- Transport for major material movements
- Product use-phase consumption (if significant)
Without this core data, results are unreliable.
Important Data
These elements significantly affect results:
- Packaging materials and quantities
- Waste generation and disposal routes
- Secondary material inputs
- Significant auxiliary materials
Reasonable estimates work if measurement isn’t feasible, but document your assumptions.
Nice-to-Have Data
These elements refine results but missing data won’t invalidate conclusions:
- Minor material inputs below cut-off thresholds
- Detailed capital goods data for large-scale production
- Precise distances for short transport legs
- Small efficiency variations in background processes
Standard database values often suffice for these elements.
Data Collection Process
Systematic data collection improves completeness:
- Map the system – Create a process flow diagram showing all activities
- Identify data needs – List all materials, energy, transport, and emissions for each process
- Assign priorities – Determine which processes need primary data and which can use secondary data
- Engage data owners – Contact manufacturing, procurement, and logistics staff who have needed information
- Document sources – Record where each data point comes from and how recent it is
- Assess quality – Evaluate whether data meets your quality requirements
- Fill gaps – Use secondary data, estimates, or proxy data for gaps
- Document assumptions – Clearly state assumptions made due to data limitations
Data Formats
Organise collected data systematically:
Spreadsheets work for most data organisation. Create sheets for materials, energy, transport, and emissions. Link sheets to maintain consistency.
Process maps show material and energy flows visually. These help identify missing data and maintain system perspective.
Documentation should explain each data point. Where did the number come from? How was it measured or calculated? What uncertainty exists?
Estimating Missing Data
Perfect data doesn’t exist. Sometimes you need estimates:
Stoichiometry calculates theoretical material needs from chemistry. Yields account for process inefficiency.
Engineering calculations estimate energy consumption from equipment ratings and operating hours.
Proxy data uses similar processes as substitutes. If you lack data for a specific material, data for a similar material provides an estimate.
Literature values from research papers or industry reports fill gaps. Academic LCA studies often publish detailed process data.
Document all estimates clearly. Sensitivity analysis should test how these assumptions affect results.
Improving Data Over Time
First LCAs use available data. Subsequent iterations improve data quality:
- Replace estimates with measurements
- Collect primary data for processes initially using secondary data
- Update old data with current values
- Refine system boundaries based on initial findings
Iterative improvement balances resource investment against result accuracy. Stop when additional data collection doesn’t significantly change conclusions.
Data Verification
Check data reasonableness:
Mass balance – Inputs should roughly match outputs plus waste. Large discrepancies indicate errors.
Energy balance – Energy inputs should equal energy in products, waste heat, and losses.
Comparison to benchmarks – Do your values match published industry data?
Peer review – Have colleagues check calculations and assumptions.
Data errors compound through LCA calculations. Verification catches mistakes before they invalidate results.
Need LCA, EPD, or CBAM consultancy?
Or have a research proposal to collaborate on?
Global commercial consultancy • Horizon Europe, UKRI & Innovate UK research partner