Enabling digitisation of continuous manufacturing processes : the role of image analysis

Cardona, Javier and Ferreira, Carla Sofia and McGinty, John and Hamilton, Andrew and Agimelen, Okpeafoh and Cleary, Alison and Chen, Yi-Chieh and Sefcik, Jan and Michie, Walter and Atkinson, Robert and Andonovic, Ivan and Tachtatzis, Christos (2017) Enabling digitisation of continuous manufacturing processes : the role of image analysis. In: Network Plus: Industrial Systems in the Digital Age Conference 2017, 2017-06-20 - 2017-06-21, James Watt Building South, University of Glasgow. (In Press)

Full text not available in this repository.Request a copy


Digitisation of manufacturing processes under the umbrella of Industry 4.0 is a multi-faceted challenge, with requirements ranging from ensuring all relevant data is captured and has meaning, to combining and analysing data streams correctly, through to creating useful and intuitive real-time user interfaces. One such data stream that has more recently become of interest due to improved processing powers allowing near real-time, and in some cases real-time analysis, is image data. We will demonstrate as a case study how images can be used to help the pharmaceutical industry in the transition from batch to continuous processing through providing near real-time analysis of in-situ crystal images taken during the crystallisation process. Traditionally, particle size and shape measurements in pharmaceutical production have been made offline, therefore introducing the risk of altering particle properties during the procedure of taking samples, drying and measuring. More recently, instruments such as the Particle Vision and Measurement (PVM) imaging tool which can provide real time, in-situ, qualitative particle image data have seen widespread uptake both in academia and industry. Here we will show an interactive tool that has been developed with the purpose of providing users of such instruments with quantitative information and statistical data as the crystallisation process develops, allowing control measures to be taken in near real-time. Combined with other simultaneously-acquired data streams, the quantitative information extracted through the imaging user tool provides both user feedback and also further enhanced control information through the use of the data as input to analysis from other methods, such as the inversion of particle chord length distributions. This work was carried out as part of the EPSRC ‘Intelligent Decision Support and Control Technologies for Continuous Manufacturing of Pharmaceuticals and Fine Chemicals’ project, EP/K014250/1.


Cardona, Javier ORCID logoORCID: https://orcid.org/0000-0002-9284-1899, Ferreira, Carla Sofia ORCID logoORCID: https://orcid.org/0000-0002-0592-8540, McGinty, John ORCID logoORCID: https://orcid.org/0000-0002-8166-7266, Hamilton, Andrew ORCID logoORCID: https://orcid.org/0000-0002-8436-8325, Agimelen, Okpeafoh ORCID logoORCID: https://orcid.org/0000-0002-0844-965X, Cleary, Alison ORCID logoORCID: https://orcid.org/0000-0002-3717-9812, Chen, Yi-Chieh ORCID logoORCID: https://orcid.org/0000-0002-8307-0666, Sefcik, Jan ORCID logoORCID: https://orcid.org/0000-0002-7181-5122, Michie, Walter ORCID logoORCID: https://orcid.org/0000-0001-5132-4572, Atkinson, Robert ORCID logoORCID: https://orcid.org/0000-0002-6206-2229, Andonovic, Ivan and Tachtatzis, Christos ORCID logoORCID: https://orcid.org/0000-0001-9150-6805;