Advanced process control using intelligent self-learning systems to optimise process conditions is key to sustainable chemical production.
The future will see “cognitive plants” using intelligent self-learning systems able to build models based on data information from multiple inline sensors. These systems will keep on learning and broaden the scope of the models for both continuous and batch-processes.
High performance computing will boost integrated simulation and forecasting capabilities. These cyber-physical systems for complete real-time plant control will improve the monitoring of quality and environmental parameters and enable plants to achieve new ambitious targets and to optimise maintenance scheduling. These advances will be at the heart of new flexible, modular, miniaturised and delocalised plant concepts, in particular when it comes to processing for the bioeconomy.
In parallel we will need to develop the capabilities of plant engineers and operators to use digital technology for process optimisation, control, smart data applications and plant maintenance. Thy will have to be able to use advanced control algorithms and developments like augmented reality technologies to develop intuitive and user friendly human-machine interfaces.
Digital links between plants will also be key to the widespread adoption of industrial symbiosis, for instance ensuring that “waste” flows of resources and energy from one process are appropriately delivered in a timely manner to other processes and can be used as “raw materials” for value added operations. Digital technologies and data flows will be essential to a truly circular economy.
Finally, the development of computational modelling methods for the rational development of materials and the screening of candidate molecules will transform drug discovery and many other areas of sustainable chemistry. In the near future no sustainable chemistry researcher will likely start an experiment or synthesis without having previously modelled the experiment and final product.