Back to search

Georgia Institute of Technology

Machine learning approaches for considering decentralized EVB pre-processing facilities with respect to end-use sectors and a potential second-use location

Abstract

dc:description.abstract

Lithium-Ion Batteries (LIBs) at End-of-Life (EoL) pose several safety risks, as LIBs have the potential to self-ignite during transportation, release toxic compounds during incineration, and can leach contaminants into landfills. To reduce these safety risks in US, LIBs are labeled Class 9 hazardous materials under the Code of Federal Regulations. This causes LIBs to be subject to numerous policies, including the requirement of certified personnel and companies to pack and ship the items, and regulatory processing with government agencies involved in transport. Efforts to improve LIB recycling focus on reducing costs to make recycling economically lucrative. Hence, there is a significant emphasis on improving recycling processes; however, transport cost alone has been identified to be on average 41% of the total cost of LIB recycling. This thesis aims to provide a methodology for choosing a Spatially Constrained Multivariate Clustering Analysis (SCMCA) heuristic which could be implemented in a variety of applications. A case study is analyzed which investigates network optimization for potential decentralized pre-processing facility locations of Electric Vehicle Batteries (EVBs) in California. The decentralized facilities aim to minimize the transportation distance and costs of shipping intact EVBs between end-use sectors, the facilities, and potential seconduse locations. The methodology consists of a clustering analysis comparison of unsupervised SCMCA Machine Learning heuristics followed by location analyses of potential pre-processing facilities. The freight capacity of the solutions under different transportation scenarios is utilized as the primary criteria to determine an appropriate SCMCA heuristic for the case study. Following this, a sensitivity analysis is implemented to evaluate the volatility of the solutions presented by the various heuristics. Finally, a staged development scenario is proposed for the construction and expansion of facilities in California to manage the increasing rate of EoL EVBs from 2024 to 2030.

Degree

thesis:*
Level thesis:degree_level
Masters
Department dc:contributor.department
Mechanical Engineering
Grantor dc:publisher
Georgia Institute of Technology
Year dc:date.issued
2022

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Haynes, Megan
Advisor dc:contributor.advisor
  • Hatzell, Marta C.
Committee members dc:contributor.committeemember
  • Wang, Yan
  • Braun, Paul

Subjects

dc:subject × 5

Rights

Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Handle dc:identifier.uri
http://hdl.handle.net/1853/67291

Chain of custody

source
Harvested from
Georgia Tech
Base URL
repository.gatech.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Haynes, Megan. Machine learning approaches for considering decentralized EVB pre-processing facilities with respect to end-use sectors and a potential second-use location. Masters thesis, Georgia Institute of Technology, 2022. http://hdl.handle.net/1853/67291