TL;DR:
- Jeff Kent, VP of Smart Platforms Technology & Innovation at Procter & Gamble, shares insights on the application of AI and machine learning.
- P&G aims to cut manufacturing operating expenses and enhance operations through staffing efficiency improvement, maintenance cost reduction, and QA cost reduction.
- The company’s digitalization journey spans five years, focusing on the practical application of Industry 4.0 concepts.
- P&G emphasizes the importance of the control system and IT/OT convergence.
- The WISE initiative drives internal collaboration and introduces AI and machine learning models across operations.
- P&G utilizes the SmartBox edge device to interface and collect data from various controls and equipment.
- OPC UA plays a vital role in enabling seamless data access and communication between OT and IT.
- P&G’s architecture supports the Manufacturing ML Lifecycle and aims to blur the boundaries of traditional system models.
Main AI News:
Procter & Gamble’s Vice President of Smart Platforms Technology & Innovation, Jeff Kent, recently shared insights on the application of AI and machine learning during the ARC Forum 2023. In his discussion, Kent outlined the company’s objectives, which include reducing manufacturing operating expenses and enhancing overall operations. These goals encompass various targets, such as a 5-10% increase in staffing efficiency, a 50% reduction in maintenance and repair costs, and a 50% decrease in quality assurance expenses. With over 25 years of expertise at P&G, coupled with experience in the U.S. Air Force working with enterprise systems and networking, Kent emphasized the significance of AI and machine learning in the digitalization of production processes.
Over the past five years, Kent has been actively involved in P&G’s digitalization journey, spearheading a corporate group that was initiated seven years ago. This group, now comprising 25 members, is dedicated to practically implementing Industry 4.0 concepts. With engineering innovation centers situated in Cincinnati, Ohio, and Kronberg, Germany, near Frankfurt, the team’s efforts have been nothing short of exhilarating, as Kent himself expressed.
Kent underscored the importance of the control system as a potent asset within P&G. The company has devised a robust operational intelligence program with the objective of implementing thousands of machine learning algorithms at the equipment level across 120 sites in over 40 countries. He emphasized that the control system should not be overlooked, as adding intelligence at the edge level plays a pivotal role in achieving IT/OT convergence.
Under the WISE initiative, P&G has established internal branding to support its efforts. This initiative encompasses the utilization of the SmartBox, an edge control and computing device that the company collaborates on with industry leaders to deliver in a practical and cost-effective manner. P&G remains committed to incorporating AI and machine learning models throughout its operations, with a particular focus on the equipment level and collaborative work systems such as quality assurance, maintenance, and material utilization. WISE serves as an overarching internal service at P&G, facilitating the DevOps of the SmartBox device and encompassing the entire Machine Learning (ML) lifecycle.
To facilitate their objectives, P&G employs an edge device known as the “SmartBox,” which interfaces and collects data from existing controls, new controls, and OEM equipment, including interfaces with industry leaders such as Mitsubishi, Rockwell, and Siemens. Kent emphasized the significance of computing at the control edge, as many machine learning algorithms crucial to P&G’s core work processes necessitate real-time integration with the control system to enable functions like adaptive control.
Data is communicated to the OT stack above and below the firewall using OPC UA, which also enables collaboration with Microsoft’s cloud applications. This data becomes available throughout the organization, accelerating digitalization and enhancing efficiency, quality, and profitability. Kent elucidated the comprehensive data cycle implemented at P&G, which encompasses data acquisition, contextualization, model development, deployment, and ultimately delivering information to operators.
P&G’s architecture is specifically designed to support the Manufacturing Machine Learning (ML) Lifecycle as outlined by Kent. This lifecycle includes data capture, historicization, contextualization, data exploration, machine learning model development, testing and validation, deployment, monitoring, and maintenance of machine learning models.
Kent unequivocally stressed the vital role of OPC UA in achieving seamless data access and establishing a common language between OT and IT. He commended the strong relationship P&G has cultivated with OPC UA, enabling reliable and scalable data access across the organization.
In a departure from the traditional Purdue Model, Kent highlighted the need for structural architectural changes in systems. While acknowledging the value of the Purdue model over the years, he emphasized the importance of blurring the boundaries between its levels to foster a more communicative and inclusive architecture. Kent expressed the necessity of introducing a network of things that transcends specific platforms, enabling the delivery of Industry 4.0 and the power of intelligence. He lauded the OPC Foundation for spearheading the definition of this agnostic network and emphasized the need to move away from the limitations imposed by the Purdue model.
Kent concluded by stating that P&G intends to roll out these architectural changes to all 120 factories within a year. He reiterated the significance of OPC UA in establishing a common language between OT and IT, as it allows for direct communication where needed, bypassing unnecessary layers of information transfer.
Conclusion:
Procter & Gamble’s strategic focus on AI, machine learning, and OPC UA is poised to revolutionize its operations and drive operational excellence. By leveraging advanced technologies and blurring the boundaries of traditional models, P&G aims to optimize manufacturing processes, reduce costs, and improve efficiency.
The company’s commitment to digitalization, coupled with its emphasis on IT/OT convergence and the practical application of Industry 4.0 concepts, positions them at the forefront of innovation in the market. P&G’s efforts are expected to set new standards for operational intelligence and pave the way for other companies to follow suit in embracing transformative technologies for business success.
