Modeling the Demand and Supply of Reconditioned Electric Vehicle Batteries under Consideration of Stakeholder Interests

The transition to electric mobility is expected to significantly increase the number of electric vehicle batteries (EVBs) reaching end of life (EoL). As EVBs contain critical raw materials and their production is associated with significant environmental impacts, it is essential to manage them in a sustainable manner. Circular EoL strategies include recycling EVBs at the material level, reconditioning them for reuse as spare parts in electric vehicles, and repurposing them for use in non-automotive applications. While recycling and repurposing have received considerable attention in both academia and industry, reconditioning remains insufficiently understood, particularly in terms of its economic viability and its acceptance among key stakeholders. The objective of this thesis is to support EoL decision-making by developing a tool that evaluates how stakeholder decisions and technical developments influence the supply and demand of reconditioned EVBs.

To this end, four studies (A–D) were conducted. Study A presents a simulation model estimating the volume of reusable EVBs as well as the number of electric vehicles requiring battery replacement. ... mehrThe estimation is based on the lifetime mismatch between batteries and vehicles and considers constraints such as battery quality and the age of the host vehicle. Study B investigates consumer demand through a discrete choice experiment. It analyzes how product attributes, such as price, life expectancy, and environmental footprint, affect preferences for replacing a failed EVB with either a new or reconditioned battery, or for retiring the vehicle entirely. Study C integrates the findings from Studies A and B and incorporates additional insights from stakeholder interviews to develop a discrete event and agent-based simulation model. The model includes 49 parameters, reflecting trends in recycling, repurposing and reconditioning costs, battery technology developments, consumer and workshop preferences, and legal requirements such as recycled content targets. The model outputs projections of reconditioned EVB supply and demand up to 2050. Study D addresses the complexity and computational burden of the integrated model by using machine learning methods to approximate its input-output relationships. It shows that different key performance indicators can be predicted with varying accuracy depending on the machine learning approach applied.

The studies show that reconditioning can be a valuable component of a circular battery economy, provided that key technical, economic and collaborative conditions are met. The simulation approach developed in this thesis serves as a flexible tool for exploring future scenarios and supporting strategic decisions by manufacturers and policymakers. Future research could examine additional stakeholder dynamics and technical challenges of reconditioning, and analyze its ecological impacts in greater detail.