To support the plastic industry in the adoption and development of novel programmed biodegradable materials, new manufacturing guiding tools are needed. Artificial Intelligence (AI) can be used to develop such tools, operating by leveraging machine learning algorithms and deep learning networks to process and analyse vast amounts of data. AI identifies patterns, learns from data inputs, and makes decisions or predictions without explicit human intervention.
The integration of AI in the plastic industry represents a transformative shift, enhancing efficiency, sustainability, and innovation.
Programmed biodegradation represents a new forefront area in material science, focusing on designing materials that degrade at a predetermined rate under specific conditions.
In recent years, a few digital tools have emerged to support programmed biodegradation (e.g., DeepBio, Material Nexus). However, such early-stage tools suffer from three main limitations: material complexity, environmental variability, and limited available data. Firstly, materials are intricate with varied chemical compositions, morphologies, and surface properties. Accurately simulating their biodegradation behaviour using molecular modelling tools is computationally expensive and requires significant data on material characteristics.
Secondly, biodegradation is highly influenced by environmental factors like temperature, pH, and nutrient availability. Often, digital tools cannot factor in these complexities and incorporate environmental data to provide reliable predictions. Finally, comprehensive data on the biodegradability of various materials and the capabilities of diverse microorganisms is scarce. It limits the training and accuracy of AI models and hinders the development of robust digital tools.
MAGICBIOMAT will pioneer groundbreaking advancements in AI that will holistically address these limitations and enable the design and manufacturing of plastics whose biodegradability will be programmed to match their intended use and lifetime, as well as their receiving environments.
