Leaf Business Consulting Services team of data scientists and data engineers started with the analysis of the Customer’s business objectives and requirements for the future software product. After that, they reviewed the existing software architecture and suggested the enhanced architecture in accordance with the Customer’s strategic and tactical goals. Leaf Business Consulting Services experts continued with the review of ML code and suggested creating ML models according to the following process:

To ensure the high accuracy of ML models, the consulting team recommended that the Customer

Leaf Business Consulting Services delivered an application that uses machine learning and visual recognition algorithms to detect drill bit defects in the images captured by the cameras and provide recommendations on required drill bit replacement and maintenance.

Once launched, the application preprocesses an image and analyzes it to recognize blades and single out individual cutters, as well as to detect blade surface and cutter wear.

The application employs an object detection neural network relying on the Hough Circle Transform method to identify cutters. A simple convolutional neural network, a primary algorithm applied for image recognition, is used to perform cutter state classification, and a separate Mask R-CNN, an algorithm for object instance segmentation, is applied to perform blades’ surface segmentation and detect blade surface defects. The application displays the results of the analysis as wear percentage and recommends an optimal date for drill bit replacement.

We worked with a small team who serve the two sides of the business: private equity and lending. They wanted to better understand their customer base and coverage across the UK, and use machine learning to explore use cases around credit scoring, risk management, fraud detection and price optimisation. In order to deliver such projects, we need to create a central data repository and onboard analysts within the company who could deliver future projects.

To achieve this, we built a ring-fenced Azure environment closely tied to the existing infrastructure where the raw data is stored as flat files in Azure blob storage, and ETL pipelines created in Azure Data Factory to move the datasets into a SQL Server database. The infrastructure and pipelines were thoroughly documented and presented back to the client in an on-site workshop.

After sourcing and loading data into the environment, we built and tested a cross-selling propensity model and an enhanced credit-scoring model, handing over reproducible code and whitepaper reports to internal stakeholders. The company is now able to rapidly iterate on analytical ideas and is at the forefront of technological innovation.

Results

15M Rows of data across 6 portfolio companies used to generate cross portfolio insights.