Time-explicit life cycle assessment: a flexible framework for coherent consideration of temporal dynamics.

Purpose: A well-known limitation of conventional Life Cycle Assessment (LCA) is the lack of temporal considerations, particularly the temporal distribution and evolution of processes, emissions, and environmental responses. While these aspects have been explored to some extent in dynamic and prospective LCA, a comprehensive approach for considering both temporal distribution and evolution is currently missing. We introduce a novel framework for time-explicit LCA that integrates the temporal distribution and evolution of product systems in the Life Cycle Inventory (LCI) phase and supports dynamic characterization of emissions in the Life Cycle Impact Assessment (LCIA) phase. Methods: The proposed approach expands the conventional LCA matrices to incorporate timing and time-based changes. We use a best-first graph traversal to derive an absolute timeline of intermediate flows by convolving relative temporal distributions at the process level. These timings are then integrated into the LCA matrices by adding time-specific row-column pairs in the technology matrix. Temporal markets are used to distribute product demands to the most-suitable processes in time-specific background databases. New rows in the biosphere matrix represent time-specific elementary flows. By preserving the timing of elementary flows during inventory calculation, time-explicit LCA enables dynamic alongside conventional LCIA. The proposed framework can be used for assessing any product system and impact category. An implementation of time-explicit LCA is provided in the open-source python package Results: We demonstrate the framework with a simplified case study of an electric vehicle (EV). For a Paris-Agreement-compatible scenario, which assumes strong decarbonization over time, time-explicit LCA determines the EV's total Global Warming Impact to be half that of a 2020 conventional LCA and nearly double that of a 2040 prospective LCA. These differences arise because time-explicit LCA uses time-specific inventory data for each timestep, depending on the timing of processes in the supply chain, contrasting the conventional or prospective cases, which rely on a single inventory database. To further demonstrate dynamic characterization, we show the instantaneous and cumulative radiative forcing over the EV life cycle. Conclusions: Overall, time-explicit LCA can provide more representative results compared to conventional LCA, by considering when processes and emissions occur and what the state of the systems is at these timings. This is particularly valuable for long-lived products in temporally variable or fast-evolving systems. Future research should focus on filling data gaps and connecting time-explicit LCA with spatial LCA or dynamic material flow analysis. Graphical Abstract: