How GFT & Google Cloud Are Tackling Data Silos To Unlock AI

Manufacturing plants across the globe generate terabytes of data daily from machines, sensors and control systems, yet much of this valuable information remains trapped in isolated systems. This fragmentation represents perhaps the most significant barrier to the digital transformation that industry leaders have promised for the past decade.

For manufacturers eager to implement AI and machine learning, these data silos present a foundational problem that must be addressed before meaningful progress can occur. Recent research from Google Cloud reveals that nearly two-thirds of manufacturers are now deploying generative AI (Gen AI), indicating growing adoption despite these challenges.

“The manufacturing ecosystem is diverse and siloed, with different OT and IT providers, making it challenging to aggregate data,” explains Brandon Speweik, Head of Industry Sales & Strategy at GFT. “But thanks to innovations like those we’ve developed with Google Cloud, we are beginning to realise AI's full potential.”

This integration challenge sits at the heart of a partnership between the technology consultancy and Google Cloud, who since 2019 have collaborated on solving the persistent problem of disconnected manufacturing data.

Manufacturing Data Engine addresses integration challenges across factory systems

The current manufacturing landscape consists of a patchwork of legacy systems alongside newer digital technologies, resulting in data that exists in incompatible formats and isolated repositories.

“Factories rely on a mix of old and new systems that produce data in different formats, preventing a clear view of operations. This leads to inefficiencies and hampers productivity,” says Fabien Duboeuf, Industry Manager at Google Cloud, who is focused on cloud, AI and data analytics initiatives in manufacturing.

Fabien, who began his career as a research scientist before spending 15 years driving digital transformation at organisations including Autodesk, Elsevier and Siemens, acknowledges that manufacturing has lagged behind sectors such as financial services in AI adoption.

“AI can improve safety, enhance employee and customer experiences, develop new business models and unlock efficiency and growth,” he adds, noting that financial services – for example – has utilised AI for fraud detection for over a decade.

Brandon, who helps manufacturers transform operations through technology at GFT, believes manufacturing stands to gain more from AI implementation than many other industries, but notes adoption has progressed slowly due to inadequate digital infrastructure.

“Now that AI is taking hold in manufacturing, its value is evident in improved product quality, more efficient production and fewer resources required,” he says. “The real excitement comes from connecting the entire manufacturing process, from raw materials to the final product.”

He views AI as a connector for manufacturing workflows, enabling real-time data analysis to optimise processes throughout production.

“We’ve implemented point solutions, such as computer vision AI for visual inspection, to detect defects in production,” he says. “This can then be extended to root cause analysis by adding traditional manufacturing data. Once we identify root causes, we can create alerts or automation to prevent defects.”

This automation capability will transform manufacturing operations in the near term, according to Brandon.

“These solutions don't replace humans, they augment their roles by providing new upskilling and enabling data-driven decisions,” he notes. “This is already happening, and the industry is starting to see the benefits.”

Despite the promise, AI implementation faces obstacles – most notably the persistent issue of data siloing.

“Manufacturing generates vast amounts of data that's crucial for process automation, but connecting and ‘right-sizing’ that data for AI use cases has always been challenging,” Brandon explains.

To overcome these barriers, Fabien argues that integrating AI with existing systems – CRMs, ERPs, PLMs and MES – provides a solution. AI can enhance these systems with flexibility and speed, enabling manufacturers to make informed decisions.

GFT partnership with Google Cloud enables factory-wide AI implementation

The collaboration between GFT and Google Cloud addresses these data integration issues head-on through a platform solution designed specifically for manufacturing environments.

“We’ve worked together to provide manufacturers with the digital infrastructure to centralise their data, making it accessible for AI applications across the production lifecycle,” Fabien says. “One example is the Manufacturing Data Engine (MDE), a cloud platform that stores and analyses data from factory machines, enabling faster and more efficient operations.”

First launched in 2022, MDE functions as a centralised data repository, aggregating information from factory sources and processing it for analysis. The platform integrates with Google Cloud’s AI and machine learning tools to support applications like predictive maintenance and quality control.

“Manufacturers can now visualise data trends, identify bottlenecks, and make informed decisions more quickly,” he continues. “MDE allows companies to use AI models for predictive maintenance, quality control and other applications to optimise production.”

The partnership has facilitated MDE implementation for clients including Ford Motor Co. GFT contributes architectural design expertise and deployment of tools for predictive maintenance, visual inspection and robotics on the data platform.

“Our work with Google Cloud provides manufacturers with the infrastructure to house their data and make it actionable,” says Brandon. “We’ve built tools that help manufacturers improve efficiency and product quality using this data.”

MDE capabilities have been demonstrated at the MXD Experience Centre, showcasing how manufacturers can improve production processes through AI-driven insights.

Beyond MDE, GFT has worked with manufacturers like Fehrer to deliver AI-powered visual inspection solutions.

“AI helps scale visual inspection by automating the process of detecting defects in assembly lines, where manual inspection can be time-consuming and prone to error,” Brandon says. “With Google Cloud’s help, we implemented a solution for Fehrer in just three months, improving production quality and efficiency.”

Despite solutions like MDE, AI adoption remains challenging for many manufacturers. Both executives offer practical advice for organisations embarking on this journey.

“Don’t rush implementation,” Brandon suggests. “Start small, take an incremental approach and showcase value step by step. It’s an evolutionary process, not a revolution.”

Fabien concurs: “Introduce AI gradually to minimise disruption. Allow employees time to adapt and provide support to help them optimise new tools.”

Flexibility during implementation is essential, according to Fabien: “Be ready to pivot based on feedback, and don't try to change everything at once. Start with technologies that integrate well with existing systems.”

Brandon emphasises that AI should complement human workers rather than replace them: “AI should create a safer work environment and enhance workflows,” he says.

Both leaders encourage manufacturers to recognise AI’s capabilities, focusing on possibilities rather than constraints.

“Be bold, experiment, and know that the only limit to AI is your imagination,” Brandon concludes.

To read the full article in the magazine, click HERE.

Explore the latest edition of Technology Magazine and be part of the conversation at our global conference series, Tech & AI LIVE.

Discover all our upcoming events and secure your tickets today.

Technology Magazine is a BizClik brand