Integrating Big Data to Plant AI for Digital Transformation

Big Data

There has been a consistent surge in the volume of data generated throughout the manufacturing chain as process industries become increasingly connected / Photo by: everythingpossible via 123RF


Customers realize they can achieve new outcomes by co-relating various types of data with the use of combinational analytics, according to research and consulting firm Frost & Sullivan, encouraging process industries to notably boost their digital spending.

There has been a consistent surge in the volume of data generated throughout the manufacturing chain as process industries become increasingly connected.

A whitepaper from Frost & Sullivan found that less than five percent of a plant operation's generated data is used. This low figure is due to improper data management strategy, limited application expertise, missing data, and insufficient resources.

However, high-fidelity data generated by sensors and other wireless devices—which currently produce inadequate insights for value creation—make a case for going digital on plant operations in a way that thoroughly boosts the plant's digitized sensor data.

"Already, more than 50% of customers claim they will invest two times more in analytics over the next two to three years," stated a press release on the whitepaper, which aims to assist in unraveling the value levers of digital transformation, understand the influence of integrating sense with artificial intelligence (AI) in plants, and assess application areas with high potential.

"The blurring of traditional automation boundaries is steering the development of innovative business models. Edge computing platforms are resulting in [the] democratization of analytics and near-real-time interfaces with sensing systems," Muthuraman Ramasamy, Automation & IIoT Industry Director at Frost & Sullivan, said in the press release.

"The industry understands the imperatives of digital, but the challenge resides in the 'how' of digital."

Ramasamy added that such challenges will push customers to seek help from established domain experts, who can help build a digital roadmap and also have notable AI application capacities over plant data as well as expertise over a manufacturing value chain.

The move from sensing to sensemaking enables customers to notice a closed loop from extracting data to value creation—leading to complete utilization of amassed high-fidelity plant data.

Over 90 percent of plant data currently contains noise that obscures patterns that could've been discovered in raw data. Analyzing such large volumes of raw data is not cost-effective and fails to provide new information. This is why there's a need to focus on the most relevant data and understanding plant processes, which can be done through:

• Comprehending the physics principles and AI algorithm in interpreting the data and determining the problem.
• Synaptically synthesizing various elements such as people, systems, data, services, and supply chains.
• Maintaining the relationship between OT, IT, business, and domain knowledge in achieving true digital transformation.