A one-to-one mapping between classical Life Cycle Assessment (the ISO 14040/14044 tradition, as practised in ecoinvent, GaBi/Sphera, SimaPro, OpenLCA) and EaaA (Everything as an Activity) — the attributional LCA framework this platform implements.
The aim is that a practitioner from either side can read across and see exactly which term or step corresponds to which.
The one-line difference
Classical LCA models a product system: you draw a system boundary, compile an inventory of exchanges up to that boundary, and multiply the boundary-crossing elementary flows by characterisation factors. Precision comes from where you stop.
EaaA models a graph of activities that never stops until it touches nature. Every produced thing resolves to the activity that produced it; the boundary is not drawn, it is emergent; and where knowledge runs out, the model shows a low accuracy score rather than a cut-off. Precision is replaced by visible, improvable uncertainty.
Same underlying arithmetic — flow × characterisation factor, summed over an upstream chain. A different discipline about where a chain is allowed to end.
The three reframes that separate them
Everything below follows from three founding principles of EaaA:
| Principle | What it removes from classical LCA |
|---|---|
| Everything is an activity. Anything created, produced, grown, mined, refined, or transported is an activity; a produced thing is never a terminal node. | The "background dataset as opaque endpoint." A produced input can no longer sit in the model with its cradle-to-gate burden hidden behind it. |
| No cut-offs. Missing or generic data is represented as a low accuracy score on the affected flows, never truncated to manufacture a usable number. | The convenient system boundary and the false precision it buys. |
| Money is a resource. A monetary magnitude is a money flow, denominated in a currency unit (ISO 4217), scored on the same gradient as carbon or water. | The idea that price is a separate discipline from environmental accounting. |
Terminology map
Read across. Left is what a classical LCA practitioner says; centre is the EaaA term; right is what actually changes.
| Classical LCA | EaaA | What changes |
|---|---|---|
| Unit process / process | Activity | Same idea — a transformation with inputs and outputs. In EaaA every production is one, industrial or natural ("factory or forest"). Burden lives on the activity. |
| Product system | Activity graph | A directed graph of activities linked by flows. No privileged foreground/background split — any activity may reference any other as input or output. |
| Exchange (activity ↔ flow relationship) | Flow with an explicit direction (input / output) | The abstract "exchange" relationship is replaced by an input or output declared directly on the flow. |
| Elementary flow (resource/emission crossing the boundary with nature) | Elementary flow — a flow targeting a natural sphere | Retained, but the compartment is a typed link to a sphere, not a free-text name like "Emission to air, unspecified". |
| Intermediate / product / technosphere flow | Intermediate (product) flow — targets the Technosphere | Retained. Activity-to-activity linkage is declared explicitly, not inferred by an implicit technosphere-matching step. |
| Technosphere (implicit holding space for product flows) | The Technosphere sphere every intermediate flow targets | No hidden linking space; a product flow explicitly targets the Technosphere and is wired to its producer. |
| Reference flow / reference product | The output flow of the activity | The activity's declared output; downstream consumers wire to it by shared resource identity. |
| Functional unit | The output flow value you query the graph at | Scaling is by the output flow of the activity you ask about; there is no separate functional-unit object. Reference year and geography are activity facets. |
| System boundary / cut-off criteria | (no boundary object; no cut-offs) | The chain runs until it reaches natural spheres. Incompleteness is not a drawn line; it is a low accuracy state on the flows that lack upstream depth. |
| Background database (as a supplier of opaque unit processes) | Imported activity — a rung on the accuracy gradient, superseded by richer producers for the same identity | An imported dataset (e.g. Rice from EXIOBASE) is a starting point with breadth and one rung of depth, not an opaque endpoint. |
| Substance / flow nomenclature | Resource — a typed identity anchored to an external authority (PubChem, IMA, GBIF, GPC…) | A resource labels a flow for validation and matching; it is never the endpoint of a chain. The same identity (e.g. gold) can be sourced from nature or from an activity. |
| Compartment / sub-compartment (air / water / soil…) | Sphere (geosphere / hydrosphere / atmosphere + sub-spheres, plus Technosphere) | A typed, hierarchical compartment the flow targets — the model's only two terminals are a natural sphere or another activity. |
| Allocation (mass / economic / system expansion) | (largely avoided) — burden follows real flows; co-products wrapped in a purpose-built output activity | The most contested classical-LCA decision is minimised. |
| Cut-off / substitution / recycled-content method | Descriptive routing — waste flows to collection → transport → processing; recovered material flows onward carrying its real upstream burden | No avoided burden, no system-expansion credit. The graph "contains what happened." |
| Pedigree matrix (reliability, completeness, temporal, geographic, technological) | Accuracy — three components: source, geography, temporal | Pedigree is a conversion input at import, then discarded. Its dimensions map into the three components. EaaA keeps accuracy, never the imported pedigree. |
| Data-quality / uncertainty (lognormal σ²) | Accuracy score + grade, propagated along the chain | Uncertainty is a first-class, per-flow, per-component output surfaced to the user, not a hidden statistical annotation. |
| Characterisation factor / LCIA method (IPCC GWP100, ReCiPe…) | Characterisation factor keyed to an impact-category method | Same concept. Applied to elementary flows over the upstream chain. |
| Cradle-to-gate / cradle-to-grave | base → leaf | A complete path from a base (draws only from spheres) to a leaf (emits only to spheres). Base-ness and leaf-ness are emergent graph properties, not declared flags. |
| Monetary value on a flow (economic allocation, price data) | Money flow — a financial resource in a currency unit | A flow's identity (what it is) and its monetary measure (what it is worth) become two distinct flows. |
Process map — ISO 14040/14044 phases vs the EaaA workflow
Classical LCA runs four phases (ISO 14040/14044). EaaA re-expresses each; the mapping is one-to-one but the emphasis shifts.
| ISO 14040 phase | Classical LCA | EaaA equivalent |
|---|---|---|
| 1. Goal & scope | Define the functional unit, draw the system boundary, set cut-off criteria, choose allocation rules. | Declare the activity by four facets — What (output resource) · How (process) · Where (geography) · When (year). No boundary is drawn and no cut-off criteria are set: the boundary is "all the way to nature," and missing depth shows as low accuracy. Allocation choices largely disappear. |
| 2. Life Cycle Inventory (LCI) | Collect unit-process data; link technosphere exchanges; compile boundary-crossing elementary flows; solve the inventory. | Traverse the activity graph. The graph is the inventory: intermediate flows are wired producer→consumer; the upstream chain is the recursive set of upstream activities; elementary flows are the ones targeting natural spheres. No produced input may terminate the chain. |
| 3. Life Cycle Impact Assessment (LCIA) | Classification → characterisation (× CF) → optional normalisation & weighting, per impact category. | Same math, same place. impact = flow value × characterisation factor, applied to elementary flows over the upstream chain, per impact-category method. |
| 4. Interpretation | Contribution & sensitivity analysis; uncertainty (Monte Carlo over pedigree); conclusions against the goal. | Accuracy, per flow and per component, propagated along the chain and surfaced as scores + a grade — replacing a bolt-on uncertainty pass. Consequential questions ("recycling vs landfill?", displacement) are answered by building two declared graphs and comparing them. |
Where the two genuinely diverge
-
Cut-offs. Classical LCA's apparent precision is partly an artefact of chains truncated at convenient boundaries and background datasets treated as exact. EaaA forbids the truncation; a shallow or generic model is valid and simply scores low. Early users are expected to score poorly — that is the design working.
-
Terminal produced things. In classical LCA a product input can rest on a background unit process whose internals are opaque. In EaaA a produced input must point at a producing activity, so its cradle-to-gate burden is structurally forced into the calculation and cannot silently vanish.
-
Recycling & end-of-life. Classical LCA offers avoided-burden / system-expansion / cut-off / substitution methods — the most contested allocation area. EaaA resolves it descriptively: recovered material flows to whoever uses it, carrying its real upstream burden; if unused, it routes to a disposal activity. No credit mechanism.
-
Uncertainty. Classical LCA treats the pedigree matrix and lognormal uncertainty as an add-on for Monte Carlo. EaaA converts pedigree into accuracy at import and makes it a first-class, per-flow, per-component, propagated output — the headline number a consumer sees is a value and a grade.
-
Money. Classical LCA uses price only instrumentally (economic allocation, monetary MRIO sectors). EaaA makes money a resource on the same footing as carbon or water, so "environmental cost as visible as monetary price" is structural.
-
Data as attribution, not authority. EaaA does not assert facts; it records what a source (PubChem, IMA, GBIF, EXIOBASE…) says, scored and supersedable. A better producer for the same identity supersedes a worse one at calculation time.
Where they converge
EaaA is attributional by design and deliberately compatible with classical LCA at the numbers. The same activity graph is intended to produce impact scores comparable to conventional LCA results when the same system boundary and characterisation factors are applied — the differences being that EaaA results carry explicit accuracy scores, and the model is flow-typed and sphere-targeted rather than encoding compartments in free-text names.
See also: Core concepts · Accuracy & confidence · Glossary.