Systematic lifecycle design and management are promising approaches for constructing sustainable product lifecycle systems. Lifecycle simulation (LCS) has been used to evaluate a product lifecycle in the design phase from both the environmental and economic perspectives. Based on material flows through each process of the product lifecycle, the LCS calculates the time variation in environmental loads, cost, and profit. In each process of the LCS model, functions that regulate the behaviors of the process, called behavior functions, are set, and these functions control material flows. Previously, we proposed a data-assimilated LCS method that combines data assimilation (DA) with LCS to realize adaptive management based on actual states of the product lifecycle. In this previous development, the DA mechanism modified the material flows of an entire lifecycle in the simulation model based on actual flows observed in each process at the time of the DA. However, because process behaviors were not modified, the gap between material flows predicted by the simulation and the flows of the actual lifecycle increased over time. To overcome this limitation, in this study, we propose a new DA mechanism that modifies the behaviors of un-observed processes based on observed material flows. The proposed DA mechanism uses the response surface methodology to estimate the behaviors while tracing the causal relation in the LCS model in reverse. A case study on a photovoltaic panel reuse business showed that the DA mechanism successfully merged the observed data into the process behaviors in the LCS model including the processes where no data were observed, thereby improving the accuracy of the simulation for future pre-diction. Systematically analyzing the current and future process states of the product lifecycle can support decision-making in lifecycle management.
ASJC Scopus subject areas