A Leading Facility Management Firm Reduced Data Errors by 70% with nCircle Tech’s Machine learning based OCR

Summary

For a leading facility management firm, nCircleTech built an ML-based OCR solution making data verification and collection process faster, easier, and considerably more accurate to reduce 70% data errors.

About the Client

The Client provides the facility management platform organizing customer data in a facility asset library to harness for the successful planning, management and optimization. The company has evolved into a SaaS platform provider that serves as a bridge between the design, construction, operations and maintenance stages of construction and/or renovation of a facility. They have been working with more than 1,200 customers providing service in over 2,000 buildings comprising 90 million sq. ft. of space.

The Goal

Client’s field engineers and contractors manually verify and copy the text data available from asset nameplates and labels which results in human errors and waste of time. They required a process which could identify and extract desired data across infinite variations of nameplate styles and layouts.

The client needed to streamline important data across various entities to verify the submittal data against the asset nameplate to confirm the correct asset is installed with additional data fields reducing the human errors and wasted time. A solution that would complement the client’s workflows and make the process of data verification and collection faster, easier, and considerably more accurate was required.

Accounting for the less-than-optimal conditions, the client also needed utter intelligence through an OCR system that can easily be preprocessed using machine learning. The existing OCR solutions available in the market are not working out for the client

Solution

• We developed a step ahead solution compared to Google and Azure OCR making better sense of the data extracted
• Machine Learning solution was utilized to correlate random letter and map these characters to correspond to the model number, manufacturer and so on
• OCR Engine implemented specific to construction data providing higher accuracy utilizing thousands of nameplates to train the OCR engine to logically identify the data
• Curated an easy to edit, expand or share format to boost productivity and improve documentation resulting further to improve overall efficiency by 2 folds

Benefits

• Captured and converted tons of construction images into a well-categorized text format with better accuracy in the interpretation of data
• Formulated an ability to read text despite multiple alignment or orientations in one document
• Improved data processing by 95% by saving more than 1611 hours
• With the new system in place, the client saw more than 70% reduction in data errors

Impact

• 2x growth in overall efficiency with improved productivity and documentation. Improved reporting efficiency by saving 1611 hours in processing. Generally KTrack processes around 50,000 Asset nameplates.
• 70% reduction in data errors

Want to know more about our work with Ktrack? Contact us and we can schedule a call to give you a detailed rundown on our ML-based OCR solution!

Watch the Demo Video of Machine learning-based OCR developed Ktrack – https://youtu.be/2b57sVToywM

Contact us

nCircle Tech (inCorporated in 2012) empowers passionate innovators to create impactful 3D visualization software for desktop, mobile and cloud. Our domain expertise in CAD-BIM customization driving automation with the ability to integrate advanced technologies like AI/ML and VR/AR; empowers our clients to reduce time to market and meet business goals. nCircle has a proven track record of technology consulting and advisory services for the AEC and Manufacturing industry across the globe. Our team of dedicated engineers, partner ecosystem and industry veterans are on a mission to redefine how you design and visualize.

Over the last 7+ years, the organization has worked on more than 150 large and complex projects for 50+ customers across 15+ countries.