A system dynamics approach to life cycle assessment

Abstract

Cars consume a large amount of energy and material resources, and generate large environmental impacts, through their production, use, and disposal. Some automotive engineers use the life cycle assessment (LCA) method to estimate and understand these environmental impacts arising from their design decisions. The standard LCA method, however, makes some limiting assumptions that exclude changes in parameter values over time. This lack of the temporal dimension leads to uncertainty in estimates, especially for new, high-volume, long-life products, such as cars. The aim of this thesis is to explore the value and challenges of an LCA method that can account for changes in resource consumptions and environmental impacts over time. This thesis modifies the standard LCA method to include some dynamics-the way that the state of a system changes over time in response to internally-generated and externally-imposed forces. A case study considers the technological intervention of replacing steel with lightweight materials in passenger cars in Australia. It considers the dynamics of the wider 'car system' at the level of the product fleet and resource flows, but not at the level of the environmental impacts. This thesis takes a System Dynamics (SD) approach by developing a computational model, a hypothesis of how the system works. The model explains the growth in the Australian car-fleet fuel consumption despite persistent policy intervention by government to decrease fuel consumption, compliance and technology innovation by car manufacturers, and shifts in transportation preferences to non-car modes by travellers. The model also allows the exploration of future scenarios wherein some car manufacturers adopt lightweight-material components and battery-electric powertrains. The computed values are used in spreadsheet calculations of resource flows and environmental impacts. The simulations of the SD model estimate that many resource benefits of lightweight cars take decades to accumulate because steel cars drain out of the fleet slowly. Furthermore, if the population of car travellers continues to grow, then oil depletion and growth in urban density could cause rapid declines in driving intensity and in the size of the petrol car fleet. Low driving intensity makes it difficult to recover the high energy investment in lightweight materials. Finally, the adoption of battery-electric cars enables car travellers to avoid most effects of oil depletion, but traffic congestion could still cause moderate declines in driving intensity and in the car fleet. The inclusion of the dynamics in an LCA study leads to surprising insights-some computational parameters, usually assumed to be constants or fixed functions of time, are considerably nonlinear. The results of the case study suggest that mass-reduction and similar 'efficiency' interventions are less effective than intended because the balancing loops and buffers of the system prevent parameters from operating in critical ranges that shift feedback loop dominance. The considerable investment into such technological interventions could be redirected, for greater effect, to policy interventions that target fuel security and traffic congestion through the behaviour and decision-making of travellers. The superiority of such interventions can be difficult to identify with a linear worldview of complex systems.