Emission measurement of construction machines through IoT-based activity mapping

The project's aim is to reduce CO2 emissions from construction machinery globally through development work in Västernorrland by developing new opportunities for more efficient procurement models and planning systems. By making CO2 emissions visible, the opportunity is created to develop new KPIs and motivate new driving forces that invest in reducing CO2 emissions.

A construction vehicle can carry out one of a wide range of activities due to its versatility. Some activities are clearly different from each other, such as transport or digging. However, some activities will have greater ambiguity, such as excavation or earthworks, but which in a larger perspective can have a major impact on carbon dioxide emissions due to, among other things, load.

To be able to identify the vehicle's activity, several measurement values will be interesting. The project will mainly try to package the various sensors used in an activity sensor node that can be easily installed and carry out activity classification in the sensor node through Edge-IoT technology.

Activity classification is performed by comparing data collected from the activity sensor node with identified activities that construction vehicles can perform. The comparison takes place through machine learning and the trained model can then be integrated into the activity sensor node to minimize the data transfer need.

When the activity of the construction vehicle is classified, a number of interesting values can be calculated. In this project, the focus will be on CO2 emissions. This means that we can monitor the vehicle's CO2 emissions in real time based on ongoing activity, but also implement changes in the follow-up and procurement of construction vehicles that affect its CO2 emissions.