PipeSearch modernizes a demand fulfillment global inventory platform
Machine learning model helps steel-pipe marketplace achieve 80% confidence in one week, transforming supply chain
Our customer
PipeSearch is a U.S. company that uses technology to connect steel pipe suppliers with global supply and demand. It offers intelligence, analytics and a proprietary inspection and certification processes for asset quality and reliability.
PipeSearch aims to modernize the rigorous and outdated demand fulfillment process in the pipe and tube industry through its platform, which provides a comprehensive, searchable global inventory of new and unused tubular stock.
This digital trading platform matches technical requirements with market availability, assesses inventory quality assurance/quality control and determines execution likelihood within a given timeframe. The result is optimal material options for end-users based on requirements, location and delivery windows, saving costs and time for end-users and providing a digital sales channel for suppliers.
Industry
Oil & gas
Challenge
Build a searchable, standardized global inventory of tubular stock.
Solutions
Business/IT transformation, data modernization, IoT, AI and machine learning, Managed Public Cloud, Amazon S3, Amazon SageMaker
Platforms
Amazon Web Services (AWS)
The obstacles they faced
During the process of creating an inventory search and trade solution, PipeSearch found that each inventory holder used various methods to store information, including spreadsheets, PDF files and CSV files. The lack of a broadly accepted format for data presentation and inventory descriptions posed a significant challenge.
To modernize demand fulfillment, PipeSearch aimed to build a comprehensive, searchable global inventory of tubular stock. However, the lack of common formats required innovative solutions to standardize and streamline the data.
“Using a combination of an agile approach with sprints and common tools, Rackspace Technology helped us stay on track and delivered a successful solution.” — Briggs Thompson, Co-Founder, PipeSearch
How we helped
To address this complexity, PipeSearch partnered with Rackspace Technology to build a machine learning model that identifies and normalizes inventory attributes and descriptors.
The data flow plan involved using Amazon S3, AWS Lambda and a searchable database to provide customers with easy access to normalized inventory descriptions.
Initially, natural language processing (NLP) was proposed, but the team opted for a blank named-entity recognition (NER) model to avoid NLP biases. They trained the model using a sample containing 100,000 examples from 10 million raw data records, achieving over 80% confidence in the results within the first week.
The solution allowed for the identification of uncertain data, which was used to retrain and strengthen the model further.
“Working with Rackspace Technology was a positive experience. The company’s expertise in machine learning and data analytics helped us simplify the difficult problem of normalizing non-standard raw inventory data.”— Briggs Thompson, Co-Founder, PipeSearch
What we achieved together
PipeSearch implemented the completed solution using Amazon SageMaker, a fully managed Amazon Web Services (AWS) service that provides everything required to build, train and deploy ML applications.
In having a thorough understanding of the problem, and through fast work with a functional proof of concept, the Rackspace team helped PipeSearch create a solution to expand its platform’s capabilities. Now customers have a more streamlined way of understanding market options for steel pipe beyond the mills.
“Working with Rackspace Technology was a positive experience,” said Briggs Thompson, Co-founder at PipeSearch. “The company’s expertise in machine learning and data analytics helped us simplify the difficult problem of normalizing non-standard raw inventory data. Using a combination of an agile approach with sprints and common tools, the Rackspace team helped us stay on track and delivered a successful solution.”
About Rackspace Technology
Rackspace Technology is a hybrid, multicloud solutions expert. We combine our expertise with the world’s leading technologies — across AI, applications, data and security — to deliver end-to-end solutions. We have a proven record of advising customers based on their business challenges, designing solutions that scale, building and managing those solutions, and optimizing returns into the future.