(370t) Going through the Disruptive Transformations in Chemical Industrial Practices : Impact of Expert Systems and Artificial Intelligence on Automation and Decision Making | AIChE

(370t) Going through the Disruptive Transformations in Chemical Industrial Practices : Impact of Expert Systems and Artificial Intelligence on Automation and Decision Making

Authors 

Yerolla, R. - Presenter, National Institute of Technology Calicut
Besta, C. S. Sr., National Institute of Technology Calicut
Kulshreshtha, S., National Institute of Technology Calicut
With machine learning, we can better engineer products that match desired results and specifications. Aside from medicines and materials, this could also be a huge game-changer in energy and environmental technologies. With changing time, the demand of the consumers has also evolved in Quality & Quantity. Industries need to adapt to the changing consumerisms and risks. On Spending of millions of hours and money in Research & Development, we have moved closer towards producing Real time solutions for production. One of the biggest disruptive technologies which have surfaced in the last decade–the perfect blend of sweet and savoury- is undoubtedly Artificial intelligence in the control systems and IOT networks of modern day industries.

Expert Systems are being designed for both on-line process control and off-line decision support in the handling of hazardous materials. There are possibilities for both positive and negative impacts on chemical industry fire/explosion loss potentials. It is a known fact that Chemical Industries and its dependencies claim maximum lives worldwide due to accidents and mishappenings. More often-negligence, poor handling, human error and faulty safety arrangements are the causes behind them getting out of control and risking thousands of lives and billions of dollars.

The prime reason that the industries are not relying on A.I. systems lie in the risk factor involved, but now due to the concepts like Reinforcement Learning, Transfer learning, Evolution Networks and Plastics networks it has become easy to customize and train the system in a pseudo environment till it achieves a success rate higher than Humans and then transfer the system at its best performance to the real task. Running on the virtual environment can become a part of R&D as it will provide us the flexibility to learn about the impacts of a change in the long run and mitigating the issues.

Such excellent predictive modelling will aid us to analyse the risk factor behind a decision and rolling it out in the best format. Outsourcing to a company specializing in AI models provides you with ready-made solutions to the challenges you are facing, without having to build the necessary infrastructure in-house. Companies working in the realm of AI solutions are willing and able to execute results, provided that the problems are measurable and traceable.

Organizations are hiring teams of data scientists before they have even put their AI strategy together. The well-known consulting firms, also, are looking to partner with AI and machine learning leaders as they seek to grow their practices and help corporations craft strategies and commence projects. Analysts have often claimed that the chemical-focused companies had fallen behind other industries in terms of exploitation of Technology.

However some expert systems have been in public domain lately which boasts about the autonomous decision making core that can handle all operational and emergency situations in day-2-day operations of production and refining units. In this paper we have discussed about creating and deploying such expert systems.