Darryl Peterson, business development manager, Antares Vision Group, highlights the critical parameters to consider when installing an Automated Visual Inspection system per specific User Requirements Specifications (URS).
Antares Vision Group
In parenteral manufacturing, visual inspection of vials is crucial, as all units must be inspected to ensure the precision quality assurance the mission-critical format mandates. Visual inspection can be performed with the human eye by a trained inspector under controlled conditions, or via automation using advanced cameras and computer technology.
Since visual inspection is a vital function of any manufacturing process, keeping current with technological advancements and global regulations is necessary to meet increasing production demands. The first step is continuously validating an incorporated system with high confidence.
The evolution of automated visual inspection (AVI)
The notion of inspecting injectable compounds to maintain consistent quality is over a century old; the first mention dates to 1915 when a document called USP IX insists such protocols must entail "true solutions”. Out of such vagueness have evolved methods and acceptance criteria that matured along with our understanding of risk and, just as significantly, advancements in inspection and manufacturing technologies. As more standards developed around what constituted "good or bad" inspection, attempts were made to finetune or supplement the human eye's limitations via tools like statistical analysis, known particle size defect standards, and consistent inspection conditions.
These capabilities were advanced further by the advent of AVI machines. AVI technology allowed a company to perform inspections faster and with more repeatable results that could be validated, substantiated, and reproduced. As with any new paradigm, vendors have continued to push the frontiers of technology in response to pharmaceutical manufacturers' various (and increasingly discerning) requirements.
What we see today in the automated visual inspection field are deployments of advancements in processing time, artificial intelligence (AI), and deep learning that promise to revolutionise the product inspection space and free up human inspectors to pursue less retina-taxing monotony.
The Advantages of AVI Systems
The disadvantages of human inspection have long been understood. Manual visual inspection leads to frequent needs for breaks, and even then, human nature invariably leads to long-term difficulties maintaining consistency - a must-have for increasingly tight tolerances. In short, there are too many human variables to meet modern requirements.
By contrast, AVI systems have traditionally been hampered by one factor: speed. Promisingly, AVI machines have made tremendous strides toward meeting the escalated production throughput needs of 21st-century pharmaceutical manufacturing and packaging lines. Compared to cameras manufactured as recently as ten years ago, today's AVI systems can employ line scan or gigabit ethernet (a.k.a. "GigE") cameras with vastly superior processing speed and image acquisition capabilities. The relatively newfound ability to take multiple images per second and then compare them to preset standards allows several criteria to be computed instantly, and a determination made as to "good or bad product" in mere fractions of a second. Today's machines can process several inspections of typical 10–20ml vials at a rate approaching 30,000 per hour.
Well-designed AVI systems are as precise as they are rapid. They can inspect vials from multiple angles for more detailed analyses of vial bases, bottoms, necks, and heels. Combined with backlighting, these camera perspectives can reveal details regarding glass cracks or vial deformities so minute that the human eye would struggle to notice them. Further, employing rapid rotation of vials allows for 360-degree inspection by stitching together images into a continuous "panorama" view of liquid or even lyophilised (freeze-dried) products.
Minding the Q’s
We define validation as a quantitative approach to prove the quality, functionality, and performance of a pharmaceutical/biotechnological manufacturing process. This approach is applied to individual pieces of equipment and the overall macro-level manufacturing process. While regulatory authorities set guidelines for validation, its specifics are too customised for comprehensive government oversight and must, therefore, be carefully compiled by the individual manufacturers. There is no one-size-fits-all handbook for validation, and AVI systems are certainly no exception to this notion.
Validation is broken down into three phases we call the three Q's: Installation Qualification (IQ), Operational Qualification (OQ), and Performance Qualification. IQ and OQ are the foundation for Factory Acceptance Testing (FAT) and Site Acceptance Testing (SAT). The vendor of the automated inspection machine should supply pharma manufacturers with a protocol plan for the testing, as this takes several weeks before initiating the process. From here, it is the manufacturer's responsibility to review and approve these proposals to ensure they align with their specific production and quality assurance needs.
Installation Qualification (IQ): This protocol ensures that the AVI equipment and its various components are installed correctly and to the original manufacturer's specifications. Calibration of major equipment, accessory components, and any adjacent utilities also should be performed in this step. In addition to ensuring proper electrical and compressed air supply setups, attention should be given to calibrating vacuum sensors, which perform not only vacuum leak checking but also help properly position vials on star-wheel conveyors. This is important, as these gauges tend to drift over time.
Operational Qualification (OQ): OQ commences following the IQ's completion. Here, tests are performed on the critical parameters of the system and the more extensive process that it drives – in other words, functions and variables associated with system equipment. If the system is the hub, this step considers the hub and various spokes, examining how it impacts the larger manufacturing environment around it.
During OQ, all test data and measurements must be documented to set a baseline for the equipment's performance. Here is where manufacturers should ensure that vials run smoothly through the inspection machine and that the intended operation and inspections work as intended.
Performance Qualification (PQ): The third and final validation phase tests the ability of the process to perform over long periods within tolerances deemed acceptable. PQ is performed on the manufacturing process as a whole by the company and is usually devised and executed by the pharma company to conform to specific line requirements. Some vendors may help craft this protocol and offer assistance during PQ. However, at minimum, the company should always direct and perform a PQ wherever possible to avoid conflicts of interest.
Individual components of the system are not tested individually during PQ, but the overall synergy and performance are run through their paces and compared to anticipated results or desired manufacturing goals.