Oral Presentation Ninth International Symposium on Life-Cycle Civil Engineering 2025

Development of an energy consumption prediction model of a batch-mix asphalt plant: a machine learning approach (112328)

Joao Santos 1 , Filippo Giustozzi 2
  1. university of twente, Enschede, Netherlands
  2. RMIT University, Melbourne, VICTORIA, Australia

The road pavement network is a critical component of a nation's infrastructure, playing a vital role in ensuring the efficient functioning of its economy. However, the construction, maintenance, and rehabilitation of road pavements demand substantial quantities of raw materials and contribute significantly to greenhouse gas (GHG) emissions and air pollution. This environmental impact is increasingly incompatible with current policy frameworks that promote strict climate objectives.

In this context, the use of Reclaimed Asphalt Pavement (RAP) in new asphalt mixtures has been pointed as a straightforward strategy to reduce the consumption of virgin materials and potentially lower energy use and GHG emissions during asphalt production. However, the environmental benefits of incorporating high amounts of RAP are not universally acknowledged within the scientific community. This discrepancy arises from the reliance on simplified thermodynamic models that fail to account for the complexities of actual production environments, where numerous variables with varying and often conflicting impacts influence the environmental outcomes.

This paper addresses this gap by developing an energy consumption prediction model for batch-mix asphalt production, based on a year-long dataset collected from an instrumented batch plant. The study also identifies the material properties and operational conditions of asphalt plants that most significantly affect energy consumption. The findings offer valuable insights for the scientific community, road pavement practitioners, and asphalt plant manufacturers, contributing to more sustainable practices in the road construction industry.