GAF is North America’s largest roofing and waterproofing manufacturer and a part of Standard Industries, a global company focused on building materials.
GAF’s products include a comprehensive portfolio of roofing and waterproofing solutions for residential and commercial properties as well as for civil engineering applications. The full GAF offering is supported by an extensive national network of factory-certified contractors. GAF continues to be a leader in quality and offers comprehensive warranty protection on its products and systems. The company’s success is driven by a commitment to empowering its people to deliver advanced quality and purposeful innovation. Learn more at www.gaf.com.
GAF and Made Smarter Technology Accelerator
Clayton McGratty, VP Corporate Communications , Said: “GAF never stops researching, testing, prototyping and creating innovative new roofing and waterproofing solutions. With technology advancement at the core of Made Smarter Technology Accelerator being an Industry Challenge Owner enables us to drive this innovation even further and explore the opportunities that advanced digital technologies has for our business as well as being at the forefront of industry. As part of Standards Industries this partnership enables us to take part in innovation with the chance to take this into a global scale of adoption.
Challenges brought to you by Industry Challenge Owner – GAF
GAF has two challenges for relevant startups. Applicants must choose only one challenge from the programme’s 14 challenges.
Read on to find out about our challenges:
- Challenge 1: Asphalt material characterisation
- Challenge 2: Machine vision systems for product conformance and machine condition
Challenge 1: Asphalt material characterisation
GAF is the leading roofing manufacturer in North America with manufacturing plants located throughout the United States. Its diverse range of commercial roofing products are sold world wide, of which a number are reliant on the production and manufacture of asphalt shingles. It is this production process which GAF is looking to innovate through engaging with Made Smarter Technology Accelerator.
GAF is interested in developing a technology based-solution to characterise, identify and understand the critical to quality (CTQ) results and parameters (material, spectral or otherwise) associated with asphalt materials so as to inform downstream manufacturing procedures for a more dynamic process.
As asphalt is produced through the refining of petroleum, it is composed of heavy, long hydrocarbon chains from crude oil. Historically, asphalt streams relevant to the roofing industry have been characterized solely based upon falling within the ranges of specific physical properties for acceptable use. While various asphalt sources may indeed fall within these ranges and even meet the targets defined within these ranges, there exists inherent differences in how these raw materials are further processed and converted. These differences may be associated with crude slates which vary geographically as sourced by regional refiners, different refining techniques or processing capabilities as managed by the refiners to meet market demand within their product portfolios or as mandated by seasonal operating constraints. Even the batch to batch inconsistencies of like asphalt streams can pose substantial challenges when further processing in configuring the machinery in manufacturing processes. This can result in waste, delays in production and impact on the final product quality.
For this challenge and from applications:
Any proposed solution should expect to characterise asphalt beyond simple material standards such as softening point, penetration and viscosity. The identified critical characteristics could be grouped as ‘families’ of discrete ranges across a broader spectrum. Known technologies exist in the areas of physical, analytical and rheological material testing.
The solution must fit within a typical laboratory environment and be suitable for technical staff to operate. Consideration must also be given to how a solution will fit within the current operating infrastructure, utilities and permits for operation; support communication with an industrial internet of things (IIoT) platform and provide event flagging/annunciation for action level and nonconforming measure.
Finally, the solution must deliver results in a timely manner of less than two hours. Any solution cannot require special permitting or cause operator safety hazards and should not require special certification based skills.
This understanding will be used to predict and train models on how the ingredients are varying over time and, in turn, be used to inform modifications to the manufacturing process. The desired outcome is to develop a solution that can enable manufacturing to run more consistently and deliver a more consistent product without jeopardising the speed and operating efficiency of the machines.
Challenge 2: Machine vision system – supporting product conformance
The manufacture of roofing products is an integral part of the GAF process. It is a complicated and high speed operation involving many variations. It consists of many attributes that are compulsory to maintaining acceptable product performance where each requires operational management to ensure presence, location, alignment and verifiable dimensional characteristics. Several of which may be impacted by the compounding effect of one variation upon another. Quality attributes include verifying the presence of discreet and continuous color patterns as well as the capture of any surface anomalies or defects. Employing real time and in-process measurement is highly desirable from a quality, cost and efficiency standpoint.
For this challenge and from applications:
GAF is looking to employ machine vision and sensor technology to assess product conformance to specification; thresholding and characterising surface anomalies and identify the process conditions when they occur; and evaluating critical machine components for continuance of use and end of life.
The desired solution must be able to demonstrate improved performance over today’s human and machine applications in relation to accuracy, repeatability, reproducibility, speed/frequency by using in line measurements at high frequency. The ability to threshold, identify and/or measure critical product attributes in both static mode (discreet unit by unit) and dynamic mode is important. Desired surface patterns are to be recognised and flagged for variation or missing elements and critical attributes are measured within 0.001″ in X, Y and Z dimensions. For dynamic assessment, surface anomalies greater than 1.0 sq. in. size to be flagged and identified via lookup database.
The successful solution cannot interfere with product manufacture or cause operator safety hazards. It should not require any contact with any product or machine; require any human response or intrusion to maintain operation. It must fit within current operating infrastructure and utilities and permits to operate; support communication with an industrial internet of things (IIoT) platform; provide event flagging/annunciation for action level and nonconforming measures.
If possible, the solution shcould integrate with robotic pick, reorient, read/measure and replace for static mode (does not currently exist); utilise grey scale or other color technology; included integrated dimensional measurement instruments and platforms.
The expectation is to improve product quality and consistency, reduction in the cost of poor quality and in related customer claims. Other expected benefits are eliminating latency associated with manual processes by providing real time information for manufacturing process control and troubleshooting, improved operating effectiveness (OEE & TEEP) and improved operating safety related to machine contact measurement by removing frequent manual and repetitive tasks and positions of people. Finally, to advance the digital capabilities for Smart Manufacturing in the areas of predictive and prescriptive analytics for artificial intelligence (AI) and machine learning (ML).
Challenge focused FAQs
We’ve had a couple of questions related directly to our challenges and GAF, please see below:
Would the team have to work on-site? If so, where?
We do not have an absolute need to have the team work on-site. However, if the solution requires some online work, it can be worked out on a case by case basis. Right now no on-site requirements.
In addition, because you’re placed in the US, what kind of testing is feasible?
Depending on the technology we can pilot is on-site or in our labs.