Predicting Delivery Failures in Technical Courier Services: A Survival Learning Approach for Sustainable Last-Mile Logistics

Purpose: Driven by direct field experience in technical courier services, this study introduces a proactive predictive framework based on survival analysis to address critical vulnerabilities, operational disruptions, and fleet emissions in sustainable last-mile logistics. Recognizing that physical IoT infrastructure is often absent or constrained in daily operations, this research investigates how surrogate telemetry data engineering can enable real-time risk prediction, supply chain resilience, and operational control. Design/methodology/approach: Utilizing a structured synthetic benchmark of 60,000 delivery orders engineered to simulate multi-sensor telemetry, the research compares penalized semi-parametric Cox models with non-parametric machine learning ensembles. The core innovation feeds an Enhanced Random Survival Forest (RSF) architecture with dynamic proxy indicators---including customer reachability streams, material stock depletion, real-time milestone progression, ambient conditions, and ISO 14083:2023 carbon metrics. The top-performing model configuration was subsequently deployed and evaluated on real operational field data. Findings: The Enhanced RSF model demonstrates superior predictive discrimination ($C\text{-index} \approx 0.84$) and calibration over traditional Cox baselines and static machine learning approaches. Crucially, empirical testing on real operational data closely mirrors the synthetic validation metrics, confirming that proxy telemetry engineering successfully operationalizes predictive risk estimation and supply chain resilience even prior to physical IoT hardware retrofitting. Practical implications: Aligned with technology-enabled supply chain management, the framework equips fleet operators and TMS environments with real-time actionable intelligence, shifting logistics from reactive exception handling to proactive resilience control. It enables dynamic mid-route rerouting, preemptive customer interventions, a drastic reduction in supply chain disruptions, and automated ISO 14083:2023 green logistics reporting. Originality/value: This work bridges technology-enabled last-mile logistics, supply chain resilience, and survival analysis. It proves that surrogate telemetry synthesis can bypass current hardware and confidentiality barriers, establishing a scalable, mathematically rigorous paradigm that translates directly into real-world operational execution and provides a standardized benchmark for future physical IoT rollouts. Keywords: Technical last-mile logistics, Supply chain resilience, Survival analysis, Delivery failure prediction, Green logistics. 🎯 Objective: This study proposes a dataset and proactive predictive framework based on survival analysis to evaluate and mitigate delivery failure risks in last-mile technical courier services, addressing operational inefficiency, elevated costs, and redundant emissions. RQ1. Which factors exert the greatest influence on the occurrence of failed deliveries in technical courier services? RQ2. Are advanced machine learning models capable of accurately predicting the time-dependent risk of delivery failure prior to service execution? RQ3. What extent can predictive analytics enhance the sustainability of logistics operations by reducing failed deliveries and minimizing unnecessary vehicle movements?

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-19
DOI
https://doi.org/10.5281/zenodo.24375728
Primary Topic
Urban and Freight Transport Logistics
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Predicting Delivery Failures in Technical Courier Services: A Survival Learning Approach for Sustainable Last-Mile Logistics

Nunziato Garofalo
Zenodo (CERN European Organization for Nuclear Research)
Urban and Freight Transport Logistics
preprint

Predicting Delivery Failures in Technical Courier Services: A Survival Learning Approach for Sustainable Last-Mile Logistics

Nunziato Garofalo
preprint en

Abstract

Purpose: Driven by direct field experience in technical courier services, this study introduces a proactive predictive framework based on survival analysis to address critical vulnerabilities, operational disruptions, and fleet emissions in sustainable last-mile logistics. Recognizing that physical IoT infrastructure is often absent or constrained in daily operations, this research investigates how surrogate telemetry data engineering can enable real-time risk prediction, supply chain resilience, and operational control. Design/methodology/approach: Utilizing a structured synthetic benchmark of 60,000 delivery orders engineered to simulate multi-sensor telemetry, the research compares penalized semi-parametric Cox models with non-parametric machine learning ensembles. The core innovation feeds an Enhanced Random Survival Forest (RSF) architecture with dynamic proxy indicators---including customer reachability streams, material stock depletion, real-time milestone progression, ambient conditions, and ISO 14083:2023 carbon metrics. The top-performing model configuration was subsequently deployed and evaluated on real operational field data. Findings: The Enhanced RSF model demonstrates superior predictive discrimination ($C\text{-index} \approx 0.84$) and calibration over traditional Cox baselines and static machine learning approaches. Crucially, empirical testing on real operational data closely mirrors the synthetic validation metrics, confirming that proxy telemetry engineering successfully operationalizes predictive risk estimation and supply chain resilience even prior to physical IoT hardware retrofitting. Practical implications: Aligned with technology-enabled supply chain management, the framework equips fleet operators and TMS environments with real-time actionable intelligence, shifting logistics from reactive exception handling to proactive resilience control. It enables dynamic mid-route rerouting, preemptive customer interventions, a drastic reduction in supply chain disruptions, and automated ISO 14083:2023 green logistics reporting. Originality/value: This work bridges technology-enabled last-mile logistics, supply chain resilience, and survival analysis. It proves that surrogate telemetry synthesis can bypass current hardware and confidentiality barriers, establishing a scalable, mathematically rigorous paradigm that translates directly into real-world operational execution and provides a standardized benchmark for future physical IoT rollouts. Keywords: Technical last-mile logistics, Supply chain resilience, Survival analysis, Delivery failure prediction, Green logistics. 🎯 Objective: This study proposes a dataset and proactive predictive framework based on survival analysis to evaluate and mitigate delivery failure risks in last-mile technical courier services, addressing operational inefficiency, elevated costs, and redundant emissions. RQ1. Which factors exert the greatest influence on the occurrence of failed deliveries in technical courier services? RQ2. Are advanced machine learning models capable of accurately predicting the time-dependent risk of delivery failure prior to service execution? RQ3. What extent can predictive analytics enhance the sustainability of logistics operations by reducing failed deliveries and minimizing unnecessary vehicle movements?

Zenodo (CERN European Organization for Nuclear Research)
Urban and Freight Transport Logistics
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon