SphereGen

SphereGen Logo

Automating the Intake of Material Certificates

SphereGen Case Study

The intake and recording of Material Certificates was a highly manual, time intensive process which was prone to error for a Connecticut Manufacturer. Their goal was to automate the process and increase accuracy.

SphereGen used Intelligent Document Processing with AI and Machine Learning to automate the intake and validation of Material Certificates from multiple suppliers. This automation transformed the Materials Certification process by removing the need for manual data entry, saving hours of time every day and increasing accuracy.

uipath silver partner badge
technology globe

Automating
the Intake of
Material Certificates

SphereGen Case Study

The intake and recording of Material Certificates was a highly manual, time intensive process which was prone to error for a Connecticut Manufacturer. Their goal was to automate the process and increase accuracy.

SphereGen used Intelligent Document Processing with AI and Machine Learning to automate the intake and validation of Material Certificates from multiple suppliers. This automation transformed the Materials Certification process by removing the need for manual data entry, saving hours of time every day and increasing accuracy.

uipath silver partner badge

O V E R V I E W

Our Manufacturing client accepts materials from multiple suppliers. Every materials shipment is associated with Material Certificates which are used as the reference point for recording and receiving data about the shipment. Each supplier’s Material Certificate has a unique layout and may represent pertinent information differently.

The traditional method of recording the materials intake is to enter all materials data manually, to their system destinations such as ERP, Quality, Procurement, etc. This method is incredibly time intensive and prone to errors due to the large amount of data which must be hand keyed.

Challenges

Our client wanted to improve their Materials Certificate intake to reduce the amount of time required to complete the process and the percentage of errors caused by manual data entry. The major obstacle in improving this process was the variation of certificate layouts per supplier and the unique data representation of data elements.

SOLUTION

SphereGen built an automation using Intelligent Document Processing with AI and Machine Learning to recognize each supplier’s Materials Certificate and look for data elements based on supplier recognition. The automation robot (bot) is trained to identify supplier certificates through a training process. Once the supplier is identified, the automation looks to certain areas of the certificate to pull the correct data. The data is sorted to a file based on the system that receives it. The files can then be loaded to the appropriate system and the next certificate is processed. If certificates are blurred and cannot be read by the automation, the certificate is tagged as an exception and put aside for manual review. The status of all processed certificates is tracked for reporting purposes. This automation has increased the accuracy of data entering the ERP system and has reduced manual interaction within the process, saving days worth of effort a week.

RESULTS

Enhanced task
productivity

Time saved for
Materials Certificate intake

Improved
Accuracy

THE DETAILS​

In automating the Materials Certification intake process, there were many steps to be researched and setup. All supplier certificates had to be identified and then the bots trained to read them.

OUR APPROACH

Identifying All Supplier Certificates

The first step of development required researching all supplier certificates flowing into the process. The layout of each certificate had to be translated into a recognition pattern so the bot could identify the supplier and which fields to pull for entry to each receiving system. Each supplier certificate is unique, so the pattern is different for every supplier. These patterns are coded to a script which the bot can use to determine how to read the certificate based on the supplier ID. In addition to each supplier’s certificate being unique, sometimes one certificate for a supplier may contain information for multiple departments. The design process and pattern recognition also had to be determined for these certificates as well.

Training the Bot

Once the certificates were identified and the patterns recorded, the bots were trained to recognize the certificates. This was accomplished by continually feeding certificates through the automation process until the bot has a high confidence level it can recognize the layout of the certificate and properly translate the data.

In the case of Material Certificates for our client, the bot must recognize the symbols of chemical elements and the compositional makeup of each element within the material. If certificates are not printed properly – for example numbers don’t line up with field headings, or the copy is blurry – then the bot has trouble recognizing the values. Properly training the bot is key to increasing the ability to overcome these issues and reducing the percentage of certificates which fail document recognition.

Once the bot was trained to a high degree of success, then the data files which the automation produced were used to feed the appropriate destination system.

Integration to the Receiving System

After the automation bot pulls the necessary materials data from the certificate, the data is stored for upload to the receiving system. These systems include ERP, Quality, and Procurement. Once all certificates in the batch have been successfully processed, the automation updates the materials received and notifies the appropriate departments that materials certificates have been successfully loaded to their system.

Exception Errors

If any certificates cannot be intelligently processed by the bot with a high degree of confidence, the pdf files are moved to a folder or Action Center for manual review. Staff are notified that exception files need attention in the Action Center. These exceptions contain the original file and the resulting values which the bot attempted to translate. Upon review, staff can either accept the data translation as is, or fix the incorrect values. This exception process is important, as the bot learns from these corrections – training it for the next time a similar issue may be encountered.

Each exception is logged for tracking throughout the system and analysis reporting.

CONCLUSION

By automating the intake and recording of Material Certificates, our client was able to recognize considerable time savings and increased data accuracy.
  • Less Data Entry time – With Intelligent Document Processing, the majority of Material Certificates can be processed automatically. A very small percentage of exceptions require manual review. Because many staff members were previously required to enter data manually, this time savings is multiplied across the board.
  • Improved Accuracy – With a large percentage of the data processed digitally, the chance for human error has been significantly decreased. When you increase the accuracy of data as it enters the system, all processes down the line benefit from this improvement.
  • Increased Productivity – As a result of less time entering data and fixing errors, overall productivity increases across multiple departments.

    Automating the processing of pdfs is just one way you can use AI based automation functions to streamline workflows and improve productivity. If you have any questions regarding automation and its benefits, we would be happy to discuss any potential opportunities for automation you may have!

    SphereGen is a unique solutions provider that specializes in cloud-based applications, custom web/mobile apps, RPA (Robotic Process Automation), and Extended Reality (AR/VR/MR). We offer full-stack custom application development to help customers employ innovative technology to solve business problems.

View and download this case study here

WE ASSIST
SMALL + MEDIUM
SIZED BUSINESSES

As a privately-held small business, we understand the goals, needs, and challenges of organizations your size. SphereGen leverages 10+ years of experience to find the ‘best technology fit’ for clients whether you build, buy or customize software and services.