Understanding How P&G Leveraged AI for Manufacturing to Yield Better Outcomes

At P&G, technology and data are at the core of corporate strategy and they work towards delivering exceptional customer experiences. Given many challenges in the old manufacturing system and the creation of new goals, P&G decided to move its manufacturing process to advanced technologies.

P&G sought to make it possible for new methods to use its exclusive historical data, composition models, and experience to generate major, game-changing ideas as well as quickly develop minor tweaks and get them to market. So, what were the challenges that P&G faced with its conventional manufacturing system?

Challenge

P&G’s traditional manufacturing method involved manual procedures including physical testing, which can drastically increase the time it takes for these items to reach the market. Formulation developers use their knowledge to construct a variety of goods that best meet the expectations of consumers. They deal with a vast number of data and factors, including product elements, restrictions, regional preferences, and consumer attitudes.

To produce a possible market-relevant product concept that meets the company’s margin targets, they must strike the proper balance. Another barrier that they had to overcome was to develop at a rate that keeps up with changing market demands. These challenges posed a great obstacle to achieving their high-set goals. Let’s learn about them as well…

Goal

By offering scalable predictive quality, predictive maintenance, targeted delivery, contactless operations, and manufacturing sustainability optimizations — which haven’t been done at this magnitude in the manufacturing arena before — P&G hopes to make manufacturing smarter.
Having known this goal, let’s see how advanced technologies such as AI in manufacturing would help P&G to make an impact.

The developments employ machine learning to automatically identify and address the reasons for line stops and rework in order to decrease manufacturing downtime, minimize waste, and slash maintenance costs.

Along with improving sustainability, the corporation will concentrate on forecasting equipment failure and deploying a digital enablement office to assist additional pilot programs. This will result in benefits such as –

• High-quality products
• Improved efficiency
• Sustainable use of resources
• Superior experiences

The 184-year-old personal care industry leader then took over utilizing AI for a variety of tasks, such as keeping track of inventory levels as goods are phased in and out to prevent waste, a strategy that P&G claims has saved the company $60 million yearly.

The four main building blocks that P&G considers when integrating AI into its applications were information, people, platforms, and trustworthiness. Every facet of the consumer packaged products business operations included artificial intelligence.

P&G wanted to discover a solution to eliminating data silos, just like many other major businesses with legacy architecture that wasn’t organized around transaction processing systems or processes. The company decided to combine all of its data into a single 10 Petabyte data lake as a result.

Automation seemed necessary for the company’s existing AI trajectory to be scaled up and accelerated, including “standardization of data; automating model development; and mechanization of risk management.” P&G thus created its own code to facilitate the automation of data centers, called internally “Turbine,” since it takes so much time to get, filter, and prepare data for use in the algorithm process.

In order to make those datasets accessible for the algorithm procedure and to increase the pace of the information integration and data-building phase, experts used this “extract, transform, and load” code. To streamline the process of developing techniques and algorithms, they have also constructed an AI factory. Come and see what were the positives of such massive-scale AI implementation in P&G.

P&G got hold of a lot of benefits by implementing AI for the manufacturing processes, the major ones being mentioned below –

• Personnel at P&G were able to assess production data and use Artificial Intelligence to make quick choices that encouraged growth and had an escalating effect.
• By utilizing ML to autonomously identify and address the major reasons for line stops and reworking, AI implementation helped decrease production downtime, minimize scrap, and save money on maintenance.
• In order to increase manufacturing efficiency, P&G was now able to gather data from sensors on the production line and employ tools like cutting-edge algorithms, machine learning, and predictive analytics.
• P&G benefitted from the deployment of AI by improving sustainability and forecasting equipment breakdown.
• P&G was able to quickly incorporate event summary data, such as production runs, outages, changeovers, and also more, together with past data using the Azure platform.
• A Digital Enablement Office (DEO), jointly established by Microsoft and P&G, was staffed by specialists from both companies with the intention to speed up technology integration and assist pilot programs. They installed the Azure platform collaboratively, and the DEO operated as an accelerator to develop high-priority corporate cases in the fields of goods production and packaging procedures that were applied across P&G.

AI for manufacturing is a vital tool for businesses in the new direct-to-consumer era to use to anticipate how customers will shop. But it cannot be accomplished with AI simply; 20% of the effort is centered on the data, analysis, and technologies required to assist the business address real-world challenges, and 80% of the task is about embracing a change in culture.

Want to deploy AI at scale in your manufacturing business?