Objectivity Case Studies Digital Transformation of Paper-Based Processes for a Major European Airline
Edit This Case Study Record
Objectivity Logo

Digital Transformation of Paper-Based Processes for a Major European Airline

Objectivity
Analytics & Modeling - Machine Learning
Cybersecurity & Privacy - Intrusion Detection
Buildings
Construction & Infrastructure
Quality Assurance
Counterfeit Product Identification
Fraud Detection
Testing & Certification
The client, a major European airline, was grappling with a cumbersome document verification process. Their ground handling office was processing over 10,000 invoices a month, delivered by nearly 600 independent companies and contractors. These invoices came in different formats and languages, making the process time-consuming and requiring the involvement of multiple people and teams. This not only delayed data analysis and subsequent actions but also made the entire process prone to errors. The client sought to optimize this process and engaged Objectivity to develop a solution for automated verification of documents and easier error detection. The team was tasked with building a Proof of Concept (PoC) using Microsoft’s technology stack to validate documents from different sources and formats.
Read More
The client is a major European carrier, established in 1928, making it one of the world's oldest operating airlines. With a fleet of almost 100 aircraft, the client flies to over 120 destinations across Europe, Asia, and North America. The airline provides services to millions of passengers and cooperates with numerous business partners. Their ground handling office processes over 10,000 invoices a month, delivered by nearly 600 independent companies and contractors, making them a high-volume, high-traffic business with extensive data sets.
Read More

Not disclosed

Read More
Objectivity decided to use Azure Form Recognizer for document extraction. They built a tool for specific visualization of the structure of PDF files to test the results from the Form Recognizer against the actual structure of the invoices. Over 200 sample invoices were checked and collected as the reference data set for further testing. The team developed methods for pre- and post-processing of the extracted data and the correction of tabular data structure where needed. The solution was then tested and fine-tuned to achieve over 95% accuracy in data detection in the test set of invoices. The digitalization mechanism included initial pre-processing of documents, utilizing Azure Cognitive Services and the Form Recognizer to extract data, post-processing of the extracted data, and restoring previously missing data. The structure of tabular data was corrected where necessary using Machine Learning algorithms.
Read More
The developed PoC provided the client with a clear understanding of the potential improvements document digitalization could bring to their processes. It demonstrated how Microsoft technologies could meet their needs effectively. The PoC allowed the client's teams to explore automation possibilities for their invoice processing and proved that the designed solution could handle their extensive data sets with efficiency and accuracy. Once deployed to production, manual correction of data would only be needed for data marked as unreliable by the application. The automated transfer of data to the client's SAP system would significantly speed up the entire process and allow for extended data analysis. The PoC enabled the client to make informed decisions regarding the direction and scope of their process digitalization, in line with their business priorities.
Over 10,000 invoices processed monthly
Nearly 600 independent companies and contractors' invoices handled
Over 200 sample invoices checked and collected as the reference data set
Download PDF Version
test test