DIGITALISATION INTO PHARMA INDUSTRY

The pharmaceutical industry (PI) is one of the fastest-growing economic sectors with worldwide sales of more than $1228.45 billion last year in 2020. Since 2017, the pharmaceutical market has grown at the rapid annual rate of 5.8%. Worldwide revenue in the pharmaceutical market was 1143 billion US dollars in the year of 2017 and it will cross 1462 billion US dollars in 2021.

The industry 4.0 of the Pharmaceutical Industry will in the future contribute toward an intelligent automation technology and may support augmented manufacturing, such as a personalized medicine, additive manufacturing, localized 3D printing of treatments etc.

Keeping up with the latest technology advancements is key for pharmaceutical manufacturers but investing in new technologies can prove to be challenging.

Digitalizing Pharma Industry

The pharmaceutical manufacturing environment is highly sensitive and tightly regulated. The smallest of errors can result in life-changing patient outcomes and have a disastrous commercial, legal and reputational impact on the manufacturer. 

A few years ago, a global pharmaceutical giant had to recall over a half a million tablets because of packaging and human-monitoring errors in the plant. 

Digitalization and automation are now ensuring the company will not experience a similar error in the future and suffer the financial ramifications and negative brand impact it endured in the past. The company has introduced digital sensors and robotics and invested in high-availability computing to guard against data-transfer issues between units. 

This has created a fully automated production line that has the by-product benefits of making it much easier to maintain cleanroom processes, capture and manage electronic batch records, and analyse process performance (through root-cause analysis) to identify and implement improvements.

Digital information integration up the supply chain and down the distribution chain is also delivering greatly enhanced demand-supply management.

Investing in the future of pharma manufacturing

Inherent complexities in pharmaceutical manufacturing lines of modern industrial facilities make precise and timely detection of malfunction occurrences necessary. In fact, unpredicted malfunctions in a production line can often provoke a cascade of adverse effects that can occur everywhere in the production chain bringing the manufacturing line to a halt for undefined time periods. Such events can have unfortunate consequences that are not always confined to the damaged part itself but propagate throughout the production line. Nevertheless, modern production lines are equipped with a multitude of data sensors that enable the real-time and fine-grained monitoring of each constituent part of the production process providing a richness of information that can be exploited by intelligent data processing methods.

A deep learning-based model for monitoring real-time raw sensor data, deriving the condition of a pharmaceutical manufacturing line and predicting the next moment in time when a malfunction can occur. The model is further able to predict the severity of the next malfunction and can contribute adjunct information in corporate decision-making. The suggested approach exploits the capacity of deep transformer models for extracting both long- and short-term correlations as well as patterns in sequential data and, combined with a linear output layer, conducts both classification and regression. The proposed approach was tested on a real dataset comprising raw data from two manufacturing lines, and it achieved promising results.

SCALABILITY -SUCCESS FACTOR

Scalability across a site or multiple sites is a key success criterion for PPM deployment, and companies tend to evaluate this capability through a staged journey. Most often, the first phase is a pilot or proof of value over a 1-year period, where a target site or sites are chosen based on identified reliability issues, instrumentation availability, and alignment with internal digital initiatives. The scope is typically limited to critical equipment, and by the end of this initial phase, organizations have gained confidence, quantified the value based on predefined success criteria and integrated PPM into their daily operations.

The second phase is centered on expanding deployment to multiple sites, upskilling existing personnel, identifying and understanding key reliability initiatives, and aligning objectives at each site. In parallel, a continuous value capture process is conducted, based on the targeted business challenge and the associated asset within the production process. This serves as the business case framework and includes the targeted failure modes and anomalies that predictive maintenance will illuminate, and equally important, what actions should be taken after detection and the time that would require. The value capture process concludes with the quantification of business impact.

The most robust use cases for predictive maintenance deployments are typically associated with reoccurring losses, limited equipment availability, and process-based contributions to machine failure. These considerations are key differentiators of the value of PPM compared to traditional condition-based monitoring.

PREDICTIVE AND PRESCRIPTIVE MAINTENANCE

Pharmaceutical companies are seeing the benefits of deploying PPM at multiple sites, such as monitoring and alerting on factors causing equipment degradation, optimizing equipment availability, and ensuring on-time delivery of medicines across product lines. A recent example5 details how a large multinational pharmaceutical company ensured security of supply by implementing a PPM program that eventually rolled out to 30 sites. In one of the many applications implemented, PPM applied to a bead mill enabled seal replacement to be extended from every eighth to every twenty-fifth batch on average. The result was an estimated saving of $10,000 per seal. In addition, the greatest benefit came from increased production uptime due to the avoided maintenance, which delayed the need for capital investment into additional production capacity.

CONCLUSION

Despite pharmaceutical companies being at varying digital maturity levels, leadership teams are continuing to recognize how enabling technologies that break down data siloes are fundamental to furthering digital transformation across an enterprise. Implementing PPM programs help minimize production disruptions from equipment degradation and can gain millions of dollars in saved batches and increased uptime, improve consistent on-time delivery of high-quality medicines, and deliver on the collective responsibility in achieving global sustainability goals.

While pharma is unique in some ways, there are generally more similarities to other manufacturing industries than differences. You need to have the data to be able to compare moving forward.

A very straightforward way to do this is utilizing a standard predictive maintenance AI tool to understand what a healthy machine looks like. You can compare the performance of your pumps in China versus your pumps in Texas. Why does one fail more often? Are our standard operating procedures the same? Different operators? Do we have different bearing suppliers? Did a specific batch have different characteristics that impacted performance? This is what the top manufacturers are asking and learning today. If you don’t have contextualized data, it becomes very difficult to benchmark.