Digital Data and Automation: Working Together
Posted on April 9, 2019 by
Jacqueline Hora, Global Digital and Data Analytics Lead
Automation Design & Data
Effective automation design can have a positive impact on data analysis and routine process monitoring across multiple sectors. However, it remains a mystery to most pharmaceutical data scientists and process engineers and as a result, end users may face several challenges in accessing useful data. So, how exactly can automation design & data benefit the pharmaceutical supply chain and the wider pharmaceutical industry? To reap the benefits of automation, effective implementation is key so that digital data derived from platforms can be used to inform established processes.
Any system involved in the automation process should flawlessly deliver data communications and records, as the connectivity between equipment helps operational teams to move forward appropriately. This is usually standard in new facilities, but ageing facilities can struggle with technology connectivity.
To facilitate flawless communication delivery, it’s important that companies take the necessary steps during the automation design phase to prevent a disconnect between early stage decision makers involved in automation systems and the end users who will need the data. This is because automation engineers take lead from design documents that often lack input from process engineers (the end users). As a result, useful data may be removed for time and cost purposes and the value of having certain information may only be realised a couple of years after automation design and data transfer is complete.
It can often take up to four years from the creation of design documents to the period a plant is running commercial volumes. This is when the data becomes pivotal to process improvements and preventative maintenance. To solve this, companies could ensure data is contextualized and putting forward supporting process requirements could be a solution.
Contextualizing data provides a lot of opportunity for manufacturers to align data across all systems with common identifiers (i.e. batch numbers). The amount of time spent manually inputting context can be significantly reduced if key pieces of information, such as text descriptions, are included when editing the layers.
It is also critical for timestamps to be representative of an activity. However, a combination of timestamp formats can make it difficult to cross reference and align data derived from multiple sources.
To this end, reflective meta data and reference points will add value to process data, as process engineers do not have to manually filter or interpret their current standpoint– the system does it for you. Automation engineers have the capacity to solve these issues when implementing new processes, however it is not standard procedure as potential challenges are not often acknowledged in the set-up stage. We’d expect this to be adjusted when design teams start to interact with end users and perceive data analytics to be a major component within the design phase.
Access to contextualised and relevant quality data generated through well-connected automation systems has great potential to improve and optimise processes in pharmaceutical manufacturing. It may be a while before all segments of the pharmaceutical industry have full access to this information, but all firms should be working towards this to increase operational efficiency.
This article is taken from Innovations in Pharmaceutical Technology April 2019, pages 9-11. © Samedan Ltd. View the full article here: www.iptonline.com where a digital page-turning version is available on the homepage.
Comments are closed.