@inproceedings{LorenzoLastMile2025,
abstract = {The COVID-19 pandemic significantly accelerated the growth of e-commerce, placing unprecedented strain on last-mile and first-mile reverse logistics (LM{\&}FMRL) operations. This surge led to increased service demand, requiring companies to expand their workforce rapidly. However, the long-term impact of this scaling on service effectiveness remains uncertain. This study examines the role of socio-economic factors in LM{\&}FMRL service completion rates using a machine learning regression approach. Data was collected from operational records of carriers across more than 400 postal codes in Spain, each with at least 200 attempted services in February 2022. Socio-economic indicators, such as average age, population density, income levels, and land use, were sourced from a proprietary database. The analysis includes multiple regression models, namely Linear Regression, Decision Tree, Random Forest, Support Vector Regression (SVR), and Gradient Boosting. Results indicate that the Random Forest model demonstrated the best performance, though overall predictive accuracy remained moderate. SHAP analysis provided insights into the socio-economic drivers of service disruptions. Understanding these factors could support strategic improvements in LM{\&}FMRL effectiveness.},
address = {Cham},
author = {Lorenzo-Espejo, Antonio
and Flores-Huam{\'a}n, Kenny-Jes{\'u}s
and Onieva, Luis
and Pegado-Bardayo, Ana},
booktitle = {Data Science, Challenges and Applications in Industrial Operations},
editor = {Barbadilla-Mart{\'i}n, Elena
and Robles-Velasco, Alicia
and Rodr{\'i}guez-Palero, Mar{\'i}a
and Cort{\'e}s, Pablo},
isbn = {978-3-032-10126-6},
pages = {121--127},
publisher = {Springer Nature Switzerland},
title = {Evaluating Socio-Economic Drivers of Service Completion in Last-Mile and First-Mile Reverse Logistics},
year = {2025}
}
Abstract
The COVID-19 pandemic significantly accelerated the growth of e-commerce, placing unprecedented strain on last-mile and first-mile reverse logistics (LM&FMRL) operations. This surge led to increased service demand, requiring companies to expand their workforce rapidly. However, the long-term impact of this scaling on service effectiveness remains uncertain. This study examines the role of socio-economic factors in LM&FMRL service completion rates using a machine learning regression approach. Data was collected from operational records of carriers across more than 400 postal codes in Spain, each with at least 200 attempted services in February 2022. Socio-economic indicators, such as average age, population density, income levels, and land use, were sourced from a proprietary database. The analysis includes multiple regression models, namely Linear Regression, Decision Tree, Random Forest, Support Vector Regression (SVR), and Gradient Boosting. Results indicate that the Random Forest model demonstrated the best performance, though overall predictive accuracy remained moderate. SHAP analysis provided insights into the socio-economic drivers of service disruptions. Understanding these factors could support strategic improvements in LM&FMRL effectiveness.