End-of-life battery route frontiers for the Andean electric-vehicle and bus fleet (2030–2045): deterministic pipeline, data and analysis code

Reproducible data pipeline and analysis code behind the paper "No regional LFP black-mass refinery, freight-positive low-margin export: end-of-life battery route frontiers for the Andean electric-vehicle and bus fleet (2030–2045)" (submitted to Transportation Research Interdisciplinary Perspectives). The deterministic "Deflusso" cohort-survival model estimates formally captured, cell-equivalent end-of-life traction-battery feedstock for Peru, Chile, Colombia and Ecuador (2030–2045) and tests it against four route gates: captured mass, legal admissibility, conforming black-mass classification, and refining/offtake economics. It reproduces the figures, tables and crossing years reported in the manuscript, including the per-tonne-of-black-mass economics (yield 0.50), the domestic break-even lithium-carbonate threshold (~US$39,000–47,000/t), second-life modelled as a 5–10 yr return delay, the BEV/PHEV and e-bus/light-duty stream splits with stream-specific collection, and the per-country, tornado and Monte-Carlo results (aggregate 2045 ≈ 20,254 t cell-equivalent; operative 17,000 t cell-equivalent/yr crossing in 2044, all-high case only). Deterministic and reproducible: the frontier uses no RNG; the Monte Carlo uses a single fixed seed (numpy default_rng(20260614)). Each stage writes a byte-stable MD5 manifest (reproduced with Python 3.10, numpy 2.2.6, pandas 2.3.3). Third-party components are cited, not redistributed: the EverBatt cost model (Argonne National Laboratory) and proprietary Fastmarkets price assessments (cited by assessment and date). Code: MIT. Data and documentation: CC BY 4.0. See LICENSE-CODE.txt, LICENSE-DATA.txt and DATA-NOTICE.md.