Indium speeds up certificate processing by 4x for global pharma leader : Low-code (Mendix) & machine learning to the rescue

Understanding the needs of a pharmaceutical giant

Our client, a leading multinational pharmaceutical manufacturer in the US with a global presence, plays a vital role in bringing life-saving medications to patients worldwide. However, ensuring the quality and compliance of their products demands a meticulous process. One critical aspect involved managing certificates of completion (CoCs) for pharmaceutical products, which arrived from various labs in PDF format.

A maze of paperwork

Here’s where the challenges began:

Inconsistent formats

The PDFs lacked standardization, making data extraction a complex and time-consuming task.

Manual processing

The current process relied heavily on manual extraction and validation of information from the CoCs, leading to inefficiencies and potential errors.

Time constraints

This manual approach significantly slowed down downstream processes, hindering timely product approvals.

The need for a smarter solution

Our client needed a solution that could automate the extraction and validation of data from these CoCs, ensuring accuracy and streamlining the entire process. They required a solution that was:

Efficient

To expedite CoC processing and accelerate product approvals.

Scalable

To handle the high volume of documents they receive from various labs.

Accurate

To eliminate human error and ensure data integrity.

Indium's AI-powered cure for manual processing woes

Understanding the client’s need for efficiency and accuracy, Indium crafted a low-code and AI-powered solution specifically tailored to tackle their CoC processing challenges. Here’s how we approached it:

Leveraging the power of AI: We implemented a solution built upon Amazon Textract, a powerful AI service from Amazon Web Services (AWS). Textract utilizes machine learning models pre-trained to extract and analyze text from various document types, including PDFs.
Automated content extraction: Textract’s AI models were fine-tuned to intelligently extract critical information from the CoCs, such as product names, lot numbers, and expiration dates. This eliminated the need for manual data entry, significantly reducing processing time.
Streamlined validation with low-code development: We leveraged Mendix, a low-code application development platform, to create a user-friendly front-end interface. This interface presented the extracted data for human review and facilitated easy validation and approval by the client’s team.
Scalability and security: By utilizing AWS S3 for secure cloud storage, the solution could seamlessly handle the high volume of CoCs the client receives. Additionally, the cloud-based platform ensured scalability for future growth.

This solution effectively addressed the client's challenges by:

Automating data extraction: Eliminating manual data entry saved valuable time and resources
Enhancing accuracy: AI-powered extraction minimized the risk of human error, ensuring data integrity.
Streamlining workflows: The automated process accelerated CoC processing, leading to faster product approvals.

Overall, Indium’s low-code and AI solution transformed the client’s CoC processing into a swift, accurate, and efficient operation, paving the way for a smoother and more efficient product development pipeline.

Quantifiable success:
The impact of AI-powered automation

The impact of Indium’s solution is undeniable, delivering significant benefits for our client:

4x Faster Extraction: The AI-powered system streamlined data extraction, resulting in a remarkable 4x improvement in processing speed. This translates to substantial time savings and increased efficiency.
80% Accuracy Boost: Manual data entry is prone to errors. By leveraging AI, we achieved an impressive 80% improvement in data accuracy, ensuring the integrity of critical information.
End-to-End Digitization: The elimination of manual extraction completely digitized the CoC processing workflow. This saves time and reduces the risk of errors associated with manual data manipulation.

Contact us

Drop a message to our team to see how we can help you

Project start date * Immediate1-3 months6-12 months

Indium will use and protect your personal data pursuant to the terms of our Privacy Policy. You may withdraw your consent at any time. But if you do we may not be able to supply you with information related to, or necessary for providing the resource.