In the design and execution of infrastructure works, the allocation of equipment and workers on the different work fronts has a frankly notorious impact on costs. Since fuel consumption is one of the factors with the greatest impact, it is relevant to accurately forecast and monitor in different scenarios of load, route (sinuosity, distance, etc.), or regularity/pavement constitution. That is why this pilot project of parameterizable estimation of truck fuel consumption in infrastructure works at BUILT CoLAB has emerged.
Scanning in Construction is one of BUILT CoLAB’s main bets and this is an example of the advantages of its use. The sensorization, application of machine learning algorithms, and real-time communication allow the estimation and monitoring of the consumption of each piece of equipment in a realistic way. This approach is noninvasive and agnostic to the manufacturer and model of the equipment in which it is applied since it does not depend on the existence of additional intelligent systems installed in the vehicle.
At this time the first prototype is collecting data in a material transport vehicle, where they are being registered to allow the training of the machine learning algorithm. Thanks to the development of a mechanism for communication and remote synchronization of the collected data, it was possible to analyze it and conclude, from now on, that there is a strong correlation between the collected data and fuel consumption, which allowed to validate the hypothesis formulated, paving the way for a truly adapted and reliable model.
This project was part of an internship project of the Instituto Superior de Engenharia do Porto (ISEP) of the student Manuel Sampaio, where Gonçalo Pereira (IoT Specialist), Manuel Parente (Head of AI), and João Moutinho (Business Director) of BUILT CoLAB were involved and has the collaboration of the Instituto Superior Técnico (IST) and JJR who, within the scope of the internship, provided full support for experience (vehicle, data collection, execution of experimental procedures).