FEATURE Manufacturing Data & Acquisition
Firmly heading toward continuous manufacturing processes
By Martin Gadsby, CEO at Optimal Industrial Technologies, and Flavio Belvedere, Co-Founder of ABCS
T
he digital transformation is well and truly under way in manufacturing, as artifi cial intelligence (AI), machine learning and Big Data become widely adopted. On the factory fl oor, manufacturers are seeing the benefi ts of connected machines, processes and systems as well as the power of Big Data. There is the potential to create intelligent networks along the value chain that can control each other and optimise production. Effi cient, highly-productive, self- automated manufacturing processes that are self-monitoring are able to both fl ag up and address anomalies in quality in real time, resulting in optimal conditions, peak performance and high-quality products. Industry 4.0 strategies also allow businesses to make signifi cant advances in their manufacturing, as they are given a unique platform to shift from batch to continuous processing, which involves integrating all process stages, without any interruption and minimal intervention from one step to another. The approach heavily reduces downtime, in particular the time required for off -line quality control at the end of each manufacturing stage, which is replaced by continuous process verifi cation and quality assurance. As a result, cycle times are shortened and higher volumes of end products can be delivered. Consequently, moving away from batch manufacturing can further optimise operations by reducing production times and manufacturing costs while increasing throughput. Continuous processes operate in a
14 April 2021 | Automation
quasi-steady state, where the stages are not independent and fi xed, but rather dynamic and interconnected. Therefore, continuous manufacturing requires control systems that are able to constantly monitor product quality to ensure that a suitable output is fed to the following stage, from the characterisation of raw material being fed to the manufacturing line until the product is fi nished.
Industry 4.0 technologies that are necessary for continuous fl ow include real-time in-line and on-line analytics, Big Data semantics, factory and process automation as well as digital twins (or cyber-physical systems). The fi rst two are necessary to obtain actionable insight and predictions about processes and operations. The acquired knowledge is then used by digital twins and automation equipment. While cyber-physical systems support the evaluation of various continuous production scenarios, the selection of the most suitable automated process control is also key to promptly adjusting processes to maintain optimal operating conditions. The combination of these Industry 4.0 solutions for process improvement and intensifi cation have something else in common. They all fall under the framework of PAT.
The importance of PAT PAT is a Quality by Design (QbD)-driven approach that aims to deliver products of consistent and high quality by designing, analysing and controlling manufacturing through timely measurements of the CQAs
of raw and in-process materials as well as CPPs. These not only act as quality assurance, but also lay the foundations for process understanding and continuous process verifi cation. In eff ect, the large volume of end-to- end material and process data regularly collected on the production line allows businesses to continuously assess and validate this information against regulatory guidelines. In this way, they can verify that their processes are always in their validated state to ensure regulatory compliance. Even more, by correlating CPPs and CQAs, it is possible to determine how one infl uences the other as well as their impact on the output of each stage and, ultimately, on end product quality. As a result, process understanding can unlock statistical process control and quality prediction for manufacturers. More precisely, manufacturers can predict diff erences in product quality from variations in process conditions and ingredient properties. This ability in turn makes it possible to reduce product variability by maintaining target values within specifi c ranges.
Complexity as a resource The correlation between CQAs and CPPs is complex, and it is only possible to describe by using multi-factorial relationships. It is necessary to apply chemometrics – mathematical and empirical statistical methods – to physicochemical data. Therefore, businesses interested in adopting PAT need to use two main strategies to deal with the great complexity that may be
automationmagazine.co.uk
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46
