Eva Weber

	With competition as fierce as ever and rigorous administrative regulations on the up, the pharmaceutical sector is under pressure to remain compliant with good manufacturing practice (GMP) and deliver high-quality products to consumers. Crucial to GMP is pharmaceutical packaging. Packaging must not only appeal to the consumer, with clearly-described ingredients and guidelines on safe consumption and side effects, but it must also be compliant with the increasingly strict regulations.

	In 2019, new packaging measures to guarantee the safe trade of pharmaceuticals are set to come into force under the European Union’s Falsified Medicines Directive (FMD). Each unit of medicine under the Directive’s remit must have an anti-tampering device and a unique identifier code on its packaging. To ensure compliance, vendors must alter their packaging by early February 2019. Applying technologies that power industrial automation can help streamline production lines to achieve this goal.

Teaching machines to “read” and “react”

	In a factory, integrating machine vision with optical character recognition (OCR) technologies enables manufacturers to ensure regulatory compliance and the quality of both product and packaging, while meeting rapid demands on the production line. In short, machine vision gives computers the ability to “see,” and OCR adds the ability to “read” information and subsequently react as humans would.

	Quality assurance and production systems relying on visual information use machine vision and computer vision systems, which combine a computer-processing unit and an industrial camera to simulate human visual control (e.g., to spot products with errors). Implementing OCR within machine vision adds the capability to react to written information, such as detecting mistakes in text on packaging.

	Checking information printed on bottles, tablet packaging, or boxes-such as expiry date, barcode, or lot number-requires inspection camera systems that are able to capture and process images on the production line. Just as inspecting the quality of a product is important, label and packaging information is equally integral to overall quality assurance. Integrating OCR technologies into robotics, computer vision, and quality control systems enables the data to be extracted-from photographed barcodes and text to scanned images-and converted into a machine-readable format such as XML or plain text data. The robots then process the information and take immediate action if quality standards fall short. This way, any anomalies or errors, particularly critical data such as expiration dates, dosage, or side effects, can be singled out automatically before leaving the site, thus, reducing the margin for error.

	Moreover, using machine vision to automatically read barcodes, lot numbers, or strings of text on packages ensures that products can be automatically routed though the production line, and that the right label will be attached at the end of the process. This streamlines the entire production process and ensures quality.

	Armed with these technologies, pharmaceutical manufacturers can put their machines to work to meet strict new regulations and safeguard both manufacturers and consumers.

	Digital transformation is the only option for traditional companies hoping to remain at the forefront of the pharmaceutical industry. Ensuring accuracy, speed, and legislative compliance is imperative to pushing ahead in the face of competition.

	As working in harmony with machines becomes commonplace as part of Industry 4.0, it is important to understand which technologies underpin industrial automation and drive productivity as well as their impact on the pharmaceutical industry as a whole. Teaching machines to read and react on production lines where speed, accuracy, and quality are crucial will fundamentally save both time and money and play a major role in achieving consumer satisfaction and regulatory compliance.

Articles

Machine vision and optical character recognition technologies enhance inspection of packages and labels.