A direct requirements table is a transformed and linked version of a Supply-use table representing a linear, homogeneous steady-state model of the economy and optionally also the environment. This mathematical structure serves as the computational foundation for Life Cycle Assessment calculations when working with input-output data or multi-functional processes.
The transformation from supply-use table to direct requirements table bridges the gap between economic statistical data and the requirements of LCA modelling. Whilst a supply-use table describes what activities produce and consume in aggregate form, a direct requirements table reorganises this information to specify the complete supply chains needed to deliver specific products. This transformation is governed by a System model, which defines the conceptual rules and mathematical procedures for linking activities and handling multi-functionality.
In a product-by-product direct requirements table, each column represents a single-product, interlinked implementation of an Activity dataset. This means that each column shows all the direct inputs required to produce one unit of a specific product, including both intermediate product inputs from other activities and elementary exchanges with the environment. The column therefore represents a fully specified unit process ready for use in LCA calculations.
A crucial property of the direct requirements table is that it contains the specification of as many Product systems as it contains activity datasets. Each column can be read backwards through the supply chain to reveal the complete product system needed to deliver the functional unit of that product. This makes the direct requirements table a powerful tool for consequential LCA, where understanding system-wide implications of changes in demand is essential.
The linear and homogeneous assumptions embedded in the direct requirements table mean that it assumes constant returns to scale and fixed input-output ratios. Whilst these assumptions simplify calculations and enable matrix algebra solutions, practitioners should be aware that real-world production may exhibit non-linear behaviour, particularly at the margins where LCA is most relevant for decision support.
