Role of Automation in Pharma
Published on : Monday 11-05-2020
Roadblocks to automation can actually be turned on their heads and they can potentially prove to be opportunities says Susanto Banerjee.
Pharmaceutical processes are complex and have multiple inter-linkages involving several functions for both Active Pharmaceutical Ingredient (API) and Finished Dosage Formulations (FDF) (Refer diagram 1). The processes work in a highly regulated environment and starting from formulating a process and its documentation to its adherence is open to scrutiny by multiple regulators depending on the markets the products would access.
Diagram 1
Each of these processes have intricate sub-processes, which have cross linkages. The manufacturing process itself has multiple cycles depending on the steps outlined in the manufacturing SOP following Good Manufacturing Practices (GMP). Each step is a process centre, governed by the documented protocols that are validated and need to be followed consistently. For example, there are multiple process routes in tablet manufacturing operation depending on the technology design of a product (see diagram 2). The manufactured tablets go for packaging which will have its own set of processes and protocols by packaging designs, complying with GMP norms. Each of these processes would be supported by Standard Operating Procedure (SOP) and any manufacturing and packaging operation will have countless SOPs which have to be maintained or updated with change control protocols.
Diagram 2
Pharmaceutical companies, around the world, have processes that are differentially automated. The Big Pharma companies, globally, have deep pockets and their Research to Manufacturing to Sales Marketing linkages are more linear which makes it obvious that they would be more automated or open to automation than their Generic counterparts. India is a branded generic market where >80% of the products are generic in nature but sold as a brand. India is also a major manufacturing hub for exports and there are more than 3000 drug companies with ~11000 manufacturing units. The automated environment of these companies or units vary anywhere between mostly manual to partially automated. Even those who are relatively more automated will most often have disparate platforms either due to various solutions pieced together over time or different platforms coming together through inorganic acquisitions and these platforms may not be talking to each other on a real time or without manual intervention. The use of robotics on shop floors and logistics management are being actively contemplated or even attempted by this industry.
The aim and objective of automation is to eliminate chances of human error. Even with hard training and rigorous programs to build awareness in the workforce engaged in manufacturing and supply chain, processes struggle to adhere to evolving compliance frameworks and many of these non-compliances are traced back to human error or negligence or ignorance. Starting from data consistency and integrity to shop floor practices, automation is becoming a key differentiator within the industry. The areas of human error are identified and prioritised for automation. Changing regulatory landscape, mergers and acquisitions, insufficient investible surplus, reluctance to change established processes and evolving technology platforms are potential roadblocks which may defer automation plans in pharmaceutical companies, but they are more likely to remain on the roadmap of these companies with a strong desire to accomplish. Roadblocks to automation can actually be turned on their heads and they can potentially prove to be opportunities and candidates for automation.
Changing regulatory landscape
Regulators around the world are continuously upgrading their expectations on compliance and their frameworks. This has not only prompted them to tighten their inspections on the ground but they come up with new frameworks for pharmaceutical organisations to comply. Regulators are mainly concerned with:
- Failure to control documents and records
- Failure to adequately investigate critical deviations and Out of Specifications (OOS) results
- Unauthorised access to electronic systems and data manipulation, and
- Process consistencies ensuring Good Manufacturing Practices (GMP) at all times.
While the first three bullets are fairly self-explanatory, they do lend themselves in some form to GMP. The GMP regulatory framework adopted by most International Organisations and Regulatory Authorities, including their main GMP protocols and supplementary guidelines, are essentially similar. Automation has made inroads into each of the aspects above, allowing the manual processes to bridge and complete the threads of activities. The extent of automation, however vary depending on the maturity and size of an organisation. Because of human interventions, and lack of appropriate inter-connectivity among disjointed automated platforms, following these protocols and following them consistently is a challenge for most pharmaceutical organisations.
Mergers & Acquisitions
As companies merge, disparate platforms come together with their own set of learning and further automation roadmap takes a pause. In most cases the merged entity will be revisiting the automation roadmap not to eliminate it but to modify it depending on their learning. In the short to mid-term, the merging entities may continue with their existing platforms but over a period of time, a transition to newer and more robust platforms will more often be the call to action.
Insufficient investible surplus
As organisations vary in size and maturity, their set of priorities and affordability dictate their automation roadmaps. Scalable systems and platforms based on SaaS models may be a good option to explore for small to midsize firms.
Reluctance to change established processes
Proposed automated environments do have a bearing on processes and sometimes these process changes can be significant, triggering upstream or downstream or lateral changes in multiple other processes. Every change control, governing process changes, especially in GMP scope, need to be validated with respect to all the activities in the thread, documented and are made available for regulatory scrutiny. Any inconsistency found can trigger a major observation and potentially can cause disruptions. Managements typically, would be reluctant to change established processes which can potentially impact many other processes. Innovative ideas and solutions which are flexible and nimble can work in these difficult situations.
Evolving technology platforms
As technology evolves, a harmonised automated architecture can easily become a disparate set of platforms over a period of time, creating chaos and data integrity issues. Diagram 3 attempts to present an example of a typical system architecture in a pharmaceutical company. The existence of multiple intricate processes have necessitated a complex architecture of multiple automated platforms operating in a single manufacturing unit. A change of a platform would typically involve validation across the landscape due to intense cross linkages within the architecture. Does it not present a huge scope to simplify this architecture?
Diagram 3
The Pharma industry is lagging behind on automation in general and robotics in particular, compared to many other manufacturing industries. The developed countries, in many ways have been a path-leader on automation within the industry and we have witnessed use of robotics in pockets like Fill & Finish processes. The industry leaders are increasingly seeing the benefits and are increasingly willing to extend artificial intelligence and robotics in other areas including Research & Development. This phenomenon in the developed countries has been driven by higher cost of labour in these countries, a pressing need to mitigate process compliance risks in evolving regulatory frameworks and efficient timing to market. The rest of the world is catching up as supply chains are continuously connecting the best of the markets with cheapest sources. Fill & Finish processes have witnessed a transformation, even in India, and the packaging lines are today mostly seen with automated cartonators and inspection units. Equipment, automated through computerised Programmable Logic Controllers (PLC), is no longer uncommon in manufacturing operations, besides packaging. These machines not only perform the process with precision but also record actions which are vitally important in a regulated environment. Block Chain technology is making its presence in Clinical Studies. A complete digital manufacturing and supply chain process (see diagram 4) may not be far away as we see this sector rapidly evolving and trending on that path.
Diagram 4
Robotics, may jump ahead in the queue and not remain behind as a step following complete digitisation. There are strong arguments on the benefits on robotics in specific areas like Clinical Studies, Drug Discovery and the Laboratory. Robots are also capable of performing tasks that are considered hazardous to human beings, like proximity to biological dangers, chances of radioactive contaminations, etc. Job losses because of Robotics, however, will be a challenge for many countries especially in the near to mid-term, post Covid-19 phase, and countries may come up with explicit policies discouraging it, at least in non-hazardous tasks. The immediate demand for Covid-19 vaccines and/or products to be used in its treatment will consume the attention of the pharmaceutical world in the near term but automation including artificial intelligence and robotics will continue its transformational journey to mitigate compliance risks and, to gain higher competitiveness and a faster timing to market.
Susanto Banerjee is a Change Catalyst who has served Corporates in Paints/IT/Pharma industries for ~27 years in various leadership roles including CFO and CHRO. He is an independent consultant in Financial and Business Transformation space. Susanto is the Founder of Ventures, a theatre based activity platform for Behavioural Coaching and pure play theatre where he coaches, directs and acts, and is associated with non-profit sector driving social impact.
