Climate change and reduction in biological diversity are global challenges for the 21st century. A significant contribution to climate-change related greenhouse gas emission origins from our society’s demand for transport and logistics.
Meeting the urgent need for reductions in mobility and logistics is challenging due to their dependence on many and complex relations, which makes it difficult to model analytically which is required for traditional optimization methods. In a new whitepaper, the researchers outline how data driven machine learning methods, which is a sub-branch of artificial intelligence (AI), can overcome these problems and be crucial in creating CO2 reduction initiatives.
Large potential for CO2 reductions
Traditional applications of machine learning in climate research include sophisticated data driven climate models, analysis of large data sets, image analysis, and analysis of data from remote sensing devices. This is extremely valuable for predicting and estimating the effect of our actions today and provide politicians with scientifically sound information for staking out strategies and political actions.
The mobility and logistics sectors are sectors with significant potential for increased efficiency and large CO2 reductions by using these methods.
While replacing raw materials or resources with environmentally friendly counterparts is not always feasible, another solution is to improve the overall efficiency so less resources are wasted, or emissions reduced. This often comes with an economical benefit too.
Environmentally friendly solutions with AI
In this whitepaper you can read about how SINTEF and partners are using AI and machine learning to create more environmentally friendly solutions in the mobility and logistics sectors.
For example, together with Skanska, researchers from SINTEF are working on reducing emission from road construction through better coordination between excavators and dumpers. In another project with Distribution Innovation the aim is to provide last-mile package delivery with more precise information to plan the shortest and most efficient routes.
Read more about the data driven machine learning methods and environmentally friendly projects in the SINTEF whitepaper written by the scientists Signe Riemer-Sørensen, Alexander Johannes Stasik, Anne Marthine Rustad, Dag Kjenstad, Petter Arnesen and Milan De Cauwer.
