USG Corporation relies on SAS to improve its manufacturing process and reduce downtime, costs and energy consumption
From skyscrapers and houses to the factory floor, USG Corporation is using advanced data to not only create new products – like its ubiquitous Sheetrock® brand drywall – but also to optimize every step of the manufacturing process for maximum quality and profitability.
We help the construction industry build stronger, safer and more sustainable communities,” says Paul Reed, Principal Technical Manager at USG. “We hold more than 2,900 patents for our products, like life-saving fireproofing systems, mold and moisture-proof systems that keep air cleaner, and eco-friendly products that help builders meet standards for coveted environmental ratings. To do that, we have to fully understand how our products are made and how they work. That requires big data and predictive analytics.”
To create the best quality products in the most efficient way, we have to fully understand how our products are made and how they work. That requires using big data and predictive analytics. Paul Reed Principal Technical Manager USG
Removing guesswork with analytics
The benefits of analytics in manufacturing directly address the challenges. Facing a barrage of global competition, USG and other manufacturers must produce high-quality products at an affordable price. This requires confidently detecting, resolving, predicting and preventing quality and reliability issues while minimizing costs.
Easier said than done, notes Reed. Offline product testing requires a tremendous amount of labor, time and materials. In addition, less-than-ideal production results can require a complete reset of the line, creating unnecessary delays and an additional cost burden in labor and scraps. Plus, an inefficient production routine drives up energy costs.
“Raw materials are one of our biggest input costs, so even the smallest improvements are very valuable to us,” Reed says.
To enhance its production process, USG is applying the power of analytics. Using SAS, USG can remove guesswork and optimize its production investments, while balancing the need to maximize choices for its data scientists and analysts to manage data, navigate discovery and ensure deployment.
A mathematically optimal procedure
USG uses the predictive modeling capabilities of SAS to streamline manufacturing while predicting product quality.
In the past, the manufacturer would test different materials and make adjustments to ensure products met quality standards. Some of the quality tests can take 24 hour or longer to perform. With SAS, USG can now predict the test result for the line operators in real time.
USG can analyze plant inputs, such as flow rates and raw material additives, to predict quality outcomes before production even starts. The manufacturer does this by deploying optimization models using SAS Model Manager – a solution that enables USG data scientists to manage their repository of models and automatically deploy the best model into production.
Using SAS Model Manager, USG can test the performance of challenger models and select the top performer. USG can single out the optimal formulation in raw materials and adjust its production process in near-real time. This allows the company to manufacture products that meet its quality standards at the lowest possible price.
“Over a six-month period, our product and manufacturing methods are completely different depending on what materials and goals we have at that time. SAS is adaptable through the choices it affords in techniques, data sources, deployment and even programming languages, while also delivering the speed and scalability that allows us to control outcomes.
“Use of predictive analytics has improved the quality, efficiency, safety and cost of our products. We’ve established a procedure that is mathematically optimal. By having the automated system calculating results on a regular basis, our plants always have the best information to drive manufacturing processes.”
USG – Facts & Figures
revenue in 2018
Building on success
What started as a successful pilot was rapidly deployed to around 20 USG plants within a year.
The true power of the predictive analytics and optimization system quickly became evident in the field, where the cost of raw materials and production styles vary by location. “SAS enabled us to analyze local plant costs alongside local production options to find the lowest-cost formula that meets our critical-to-quality parameters at each plant,” Reed notes.
While the operational personnel are thrilled with the quality results, the IT personnel are similarly pleased with the performance of the computing environment. To generate the necessary computing power, USG relies on SAS Grid Manager for a centrally managed grid computing environment with workload balancing, high availability and faster processing.
“We have SAS making predictions 24 hours a day, seven days a week. It’s a tremendous amount of computational effort to provide these answers for the plants.”
To gain a transparent view of production results and operational metrics, operators and executives at USG use SAS Visual Analytics, which is accessible and easy to use for everyone from line personnel to data scientists. The interactive dashboards enable business users to be self-sufficient in trend analysis and allow Six Sigma programs to excel.
“If performance drops in any area, the prescriptive functions of SAS Visual Analytics tell us what we need to change to maintain compliance,” Reed says. “It empowers our operators with the array of choices needed to deal with a wide variety of situations, and all that goes right to the bottom line.”
USG prides itself on applying the latest technology to ensure its manufacturing plants are automated, modernized and ready to meet demand. “We have to maintain our future-focused approach,” Reed says. “For instance, the Internet of Things – connected devices and machines. We’re exploring how we want to expand our products to include smart functions. There are so many more opportunities out there, and analytics works hand in hand with them.”
The results illustrated in this article are specific to the particular situations, business models, data input, and computing environments described herein. Each SAS customer’s experience is unique based on business and technical variables and all statements must be considered non-typical. Actual savings, results, and performance characteristics will vary depending on individual customer configurations and conditions. SAS does not guarantee or represent that every customer will achieve similar results. The only warranties for SAS products and services are those that are set forth in the express warranty statements in the written agreement for such products and services. Nothing herein should be construed as constituting an additional warranty. Customers have shared their successes with SAS as part of an agreed-upon contractual exchange or project success summarization following a successful implementation of SAS software. Brand and product names are trademarks of their respective companies.