Verification and critical review of a cement EPD with a comparative claim
A ready-mix concrete producer commissions an EPD for CEM I 42.5 R cement, declared unit 1 t, to EN 15804+A2. The consultant uses ecoinvent 3.8 cut-off although the EPD template states 3.9.1; alternative fuels (tyres, biomass) are allocated by economic value; internal transport of limestone from the quarry to the kiln is excluded because it takes place 'within the site'; GWP-biogenic is reported but GWP-luluc is left blank. The draft states '30% lower GWP than the European average cement'. The proposed verifier is the owner of the LCA software used and also trained the consultant. The producer wants to publish in 3 weeks and asks whether the verifier can issue the verification after a desk check.
1. What type of review is required, who can perform it, and is the proposed verifier acceptable?
Read the worked answer
The review type is driven by the claim. For the EPD itself, EN 15804 requires independent verification by an approved verifier. The statement '30% lower GWP than the European average cement' is a comparative assertion disclosed to the public, which triggers ISO 14044 section 6.1: a critical review by a panel of at least three members, including a chair and at least one non-LCA expert (for example a cement technologist or a procurement specialist). Panel members must be independent of the study.
The proposed verifier is not acceptable. Owning the software and training the consultant creates a self-review conflict; independence is compromised even if the technical competence is high. A desk check by one person cannot satisfy either EPD verification or the ISO 14044 critical review. The producer should appoint a new independent verifier and, because of the comparative claim, a review panel before the 3-week deadline is considered. The panel must review goal and scope, data, allocation, LCIA, interpretation and the comparative claim itself.
2. List the technical nonconformities in the draft and decide whether verification can be issued.
Read the worked answer
Nonconformities:
- Dataset version mismatch: the model uses ecoinvent 3.8 but the EPD declares 3.9.1, so the result is not reproducible as declared.
- Alternative-fuel allocation by economic value may not follow the PCR; check whether the PCR requires energy content or system expansion for waste-derived fuels.
- Internal transport is not outside the system boundary merely because it happens on site; include it in A2 unless the PCR explicitly excludes it.
- GWP-luluc must be reported, even if zero; a blank field is a declaration error.
- The comparative claim lacks a reference EPD with identical scope and a critical review; it cannot stand as written.
- Declared unit and reference service life must be consistent across the compared products.
Verification cannot be issued. These are major nonconformities, not editorial corrections. Corrective actions are: rebuild or re-run with 3.9.1, revise allocation, add internal transport and GWP-luluc, remove or substantiate the comparison, then re-verify. The 3-week publication date is unrealistic.