OR WAIT 15 SECS
Analyzing process and equipment data provides insights that can improve quality and productivity of pharmaceutical manufacturing.
The level of automation in the bio/pharma industry varies and continues to evolve. Many are investing in Industry 4.0 concepts such as digitalization and automation to improve manufacturing efficiency, quality, and capacity. According to Amos Dor, general manager of the Applied Materials’ Automation Products Group, one such capability for improving operations is advanced analytics, which takes manufacturing data and provides insights into how the process and equipment are holistically performing. Analytics-combined with advanced scheduling solutions-can be used to minimize downtime and identify additional capacity within existing manufacturing constraints, says Dor. To further examine current trends in pharmaceutical manufacturing, Pharmaceutical Technology spoke with Dor and with Amy Doucette, manager for the Americas for Applied Materials’ Automation Products Group, Pharma Solutions.
PharmTech:What are some ways to address the problem of how a lack of manufacturing capacity or quality impacts drug shortages?
Dor (Applied Materials): We see that the pharmaceutical industry is investing in digitalization of their manufacturing facilities. They are trying to extract value from the data being collected on the plant floor by using machine learning and advanced analytics to gain insight into how their process and equipment are holistically performing. For example, our advanced analytics and control solution applies analytics to automate business processes end-to-end to increase quality and productivity in pharmaceutical manufacturing.
In addition, the ability to integrate directly to the plant floor enables an advanced scheduling digital twin with optimization algorithms and expert rules to manage the plant, ensuring schedules are realistic and optimal for the next schedule horizon.
PharmTech: How are digital twins being used in pharmaceutical manufacturing?
Doucette (Applied Materials): Pharma is taking a phased approach to implementing digital twins. Some companies are starting in the quality control (QC) lab because QC is integral to ensuring that products are released on time. Others are taking initial snapshots of one manufacturing line and eventually building out to the entire manufacturing facility and its supply chain.
Digital twins allow scenario planning, using our scheduling solution, to go through different scenarios for how a company can debottleneck a process. Sometimes the process can be optimized by simply changing the schedule, and other times adding equipment or operators should be evaluated. Using these solutions, the workforce can work smarter to increase productivity and quality by increasing yields, reducing variability, minimizing downtime, and identifying capacity within existing manufacturing constraints.
PharmTech: What are some of the key lessons pharma can learn from other industries-such as semiconductor manufacturing-on modernizing manufacturing?
Dor (Applied Materials): As James Moyne from Applied Materials and co-authors wrote in 2017, the semiconductor industry, which operates at higher process capabilities and overall equipment effectiveness than the pharmaceutical industry, can serve as a role model. Semiconductor chip manufacturing has used advanced sensing and diagnostics, advanced process control, and robotics and automation, as well as advanced dispatching and scheduling, to achieve profitable productivity and quality (1).
Looking at the BioPhorum Operations Group (BPOG) plant maturity model (2), many pharmaceutical manufacturers are moving from digital silos (Level 2) to a connected plant (Level 3). The goal is a predictive plant (Level 4) and eventually progressing to an adaptive plant (Level 5). The semiconductor industry has been on this journey longer. In the pharma industry, we are working to help companies connect the silos. We have an agnostic platform that sits above the automation system and integrates with manufacturing execution systems (MES) and enterprise resource planning (ERP) systems. We are working with internal Applied Materials groups, equipment manufacturers, and sensor companies to enable a modular “plant in a box” that integrates predictive analytics with other layers of automation, such as inspection, MES, and ERP.
1. S. Romero-Torres, J. Moyne, and M. Kidambi, Am. Pharma. Review, 20 (1) 2017.
2. BPOG, Development of a Digital Plant Maturity Model to Aid Transformation in Biopharmaceutical Manufacturing, www.biophorum.com (2016).