My question is with regard to the basics of LCA. There may be different ways to conduct an LCA or different approaches (different algorithms) to conduct an LCA. Which approach would be better than the others? A comparative study considering all the approaches would be helpful. Pls walk me through the steps of the algorithm that is being used in One Click LCA software.
How would AI/ML techniques (or any statistical models) help in conducting LCA for better accuracy, given the non-availability of missing data in the One Click LCA database?
Hi Chintan. Welcome to the One Click LCA Community and thank you for your question. Could you specify the type of LCA you are referring to? On our platform, we offer different types of LCA modules which can be used.
In regards to using e.g. AI/ML, this can help with data import where material labeling can be done with automation (already in place in One Click LCA), which speeds up the process of undertaking LCA quite a bit. However one should always manually confirm the choices as you are the one who controls your project’s data.
Non-availability or missing data is another topic altogether that may not be solved as easily as this is dependent on numerous key factors about the project and materials used. For buildings, Carbon Designer can be used to make assumptions on data for specific building parts (quantities and types of materials) which does not necessarily work with AI but rather preprogrammed models.
If your query was related to product LCA then this requires altogether a different approach which may not have a direct solution using AI or Machine Learning.