Machine Learning for Pharmaceutical Process Control and Monitoring:

Revolution or Evolution?


30 March - 2 May, Crowne Plaza Chester, UK

Furqan Tahir - Perceptive Engineering

Register / Learn More

APACT

APACT is an open forum for the presentation and discussion of recent scientific and engineering advances relevant to process analytics and control technologies. Plenary and keynote speakers will report recent advances in the development and application of novel process analytics, predictive modelling and control technologies, and will review the benefits achieved. 
Following the success of previous conferences, APACT 19 will be a 3-day meeting featuring plenary and parallel sessions on topics crucial to the achievement of manufacturing excellence.


Process Analysis and Control

The financial, safety and environmental benefits of process analysis and control (PAC) technologies have never been more important to modern manufacturing in the pharmaceuticals, chemicals, materials, petrochemicals, biotechnology, food industries, and many others. The benefits of in-process measurements, real-time data mining and predictive modelling include: 
  • Improved process efficiency or conversion 
  • Reduction of waste products 
  • Better safety 
  • Improved fundamental understanding of your process 
  • Reduced “time-to-market”


ABSTRACT

Four terms that have achieved a high level of both media and industry interest over the past three years are Industry 4.0, Internet of Things (IOT), Artificial Intelligence (AI) and Machine Learning (ML). The algorithms underpinning ML are designed to overcome following strictly static program instructions by making data-driven predictions or decisions [1]. The power of Machine Learning in particular, is often described as the ability to leverage value from data. For the Pharmaceutical industry, statistical and mechanistic models are already routinely used to obtain value from data, by predicting process and product behaviour in real-time.

Advanced Process Control (APC) and Multi-Variate Analysis (MVA) are well established tools within the process industries. These technologies use data driven techniques such as regression and classification to maximise product quality whilst improving process robustness and agility. When we compare these “traditional” data-driven techniques with Machine Learning we see the same algorithms being applied and a common goal of improved decision making through data analysis and observation being aspired to.

In this talk, case studies from the Pharmaceutical and other industries are used to demonstrate the application of several forms of supervised learning for process control and optimisation. The case studies present applications for the rapid development of synthesis, granulation and crystallisation processes.

Our Clients & Partners

Selection of our clients and key partners we work with to improve process efficiency