Skip to main content Skip to main navigation menu Skip to site footer
Article ID: 101708 Article

Last-Mile Logistics Optimization in West Africa: A Machine Learning-Driven Case Study of Ghana and Burkina Faso

  • Eugene Ansah Owusu Ampaw
    School of Traffic & Transportation Engineering, Central South University, Changsha 410083, China Department of Theory and History of International Relations, Peoples’ Friendship University of Russia, Moscow 117198, Russia
    Author
  • Qin Jin
    School of Traffic & Transportation Engineering, Central South University, Changsha 410083, China
    Author
  • Maxwell Ako Owusu-Ampaw
    Department of Theory and History of International Relations, Peoples’ Friendship University of Russia, Moscow 117198, Russia
    Author
  • Isaac Richard Malan
    School of Computer Science and Engineering, Central South University, Changsha 410083, China
    Author

Abstract

In Sub-Saharan Africa, the emergence of e-commerce and the rising population of cities is placing novel strain on logistics systems. The efficiency of last-mile delivery (LMD) is an important factor for improving the economic output and sustainable development of cities, particularly in fast urban development areas. This study investigates LMD efficiency in urbanizing Accra, Ghana, and Ouagadougou, Burkina Faso for the effects of urbanisation. It also explores the various urban problems such as persistent mass traffic jams, woefully inadequate infrastructure, and a growing number of informal settlements characterized by a lack of addressing systems that collectively impede the operation of LMD. The mixed methods approach integrates Geographic Information System (GIS) analysis with stakeholder surveys and statistical modelling to uncover that unmanaged traffic conditions, coupled with poor road quality, are leading contributors to inefficiency. It increases delivery times, increases operating costs due to fuel use and vehicle degradation, and aggravates environmental damage through the emission of greenhouse gases. Furthermore, informal economies and a lack of formal addressing aggravate these logistical problems within the research, the researchers highlight. An important contribution of this research is the development and empirical validation of a contextualised Automatic Machine Learn ing (AutoML) framework. The results revealed in a systematic case study simulation showed that routes optimized with AutoML achieved a 28% reduction in total delivery time, an estimated 22% reduction in travel distance, and a predicted reduction in fuel consumption by 22.1% compared to routing using regular routing approaches. This paper suggests specific policy recommendations to cities in Sub-Saharan Africa.

Keywords:

Last-Mile Logistics, Automatic Machine Learning (AutoML), Delivery Time Estimation, Route Optimization, Sustainable Transport

References

    Issue

    2025 Vol.3 No.2

    Copyright & License

    Copyright (c) 2025 Eugene Ansah Owusu Ampaw, Qin Jin, Maxwell Ako Owusu-Ampaw, Isaac Richard Malan

    ×