With the advent of AI-driven technologies, especially predictive analytics tools that use machine learning, Lean has joined with Industry 4.0 to reach new levels of efficiency and value.
Lean principles have played an important role in manufacturing for decades. Fuji Cho, the CEO of Toyota, is generally credited with having led the Lean manufacturing concept in the 1950s, when he spearheaded principles to eliminate waste, but lean principles have continued to evolve and improve ever since, generating approaches like Six Sigma and the Four Principles.
The process manufacturing industry hasn’t been slow to learn from batch manufacturers, implementing Lean approaches that help drive value and boost profit.
What are Lean principles?
The primary definition of Lean is to eliminate waste from every stage and process within manufacturing. Cho defined waste as "Anything other than the minimum amount of equipment, materials, parts, space, and workers time, which are absolutely essential to add value to the product.” Lean is usually associated with the continuous improvement/continuous development approach, which is a commitment to ongoing change to drive value and cut waste.
Hand in hand with the goal of eliminating waste comes the mission to increase value. They are two sides of the same coin, because as Cho noted, waste is that which doesn’t add value to the product. Process plants that aspire to Lean practices need to remove every action that doesn’t drive value. This rests on 3 pillars:
Reduce wasted materials. Every part failure that could have been prevented is a waste of equipment. Every batch of product that’s below standards is a waste of raw materials; overproduction is a waste of money for storage; and any energy loss is also a form of waste.
Save time. Lost time is considered another form of waste, whether that’s worker time that was wasted preparing sub-par product, managerial time wasted before starting or completing a project, or a preventable equipment failure that causes a factory shutdown.
Cut total costs. In a way, this is another aspect of reducing waste. Lean demands that you cut all unnecessary costs, including extra personnel costs, materials costs, and the cost of more work hours.
How does SAM GUARD drive Lean practices?
You can’t achieve Lean principles without careful and accurate observation and analysis, otherwise it’s just guesswork. You need to track batch quality, part failure rates, worker hours, and more in order to spot waste and increase efficiencies, and that’s precisely where Machine Learning (ML) tools like SAM GUARD come in.
Improve efficiency in production to cut waste
SAM GUARD uses ML models combined with human intelligence (ML+HI) to quickly map new plants and learn how to distinguish between “normal” operation and anomalies. Clustering anomalies allows SAM GUARD to produce accurate early alerts about potential part failures.
By spotting failures before they occur, SAM GUARD enables plant engineers to fix equipment and parts instead of wasting them by “running to fail.” Additionally, equipment that is in top condition is more efficient, wasting less energy and producing more consistent product.
For example, a complex SAM GUARD alert showed that the heat transfer in a gas economizer in a steam process plant had fallen by 3℃, which causes a serious decrease in the efficiency of the economizer. When efficiency drops, the plant has to pay more for gas consumption, which is a waste of energy and of money.
Elsewhere, in a pesticide plant, SAM GUARD generated an alert about temperature anomalies in the reactor. When the process engineer investigated, they found that the valve controlling the reactor floor was leaking material, making it impossible to close the valve fully. The faulty valve was causing a waste of raw materials, as well as resulting in damaged product which had to be thrown away. Without SAM GUARD, the total waste would have been far greater.
Save time in the investigative process
It frequently takes a lot of time and effort for plant managers and process engineers to investigate alerts and identify the cause of the problem. Noticing part failure or a drop in product quality is only the first step in a time- and resource-consuming journey to find out what went wrong and how to correct it.
Here too, SAM GUARD can help. SAM GUARD’s alerts are accurate enough to speed up the CI/CD process of iterative improvement by directing engineers to the cause of the problem.
Reduce costs by preventing expensive failures
Because SAM GUARD can spot impending failures well in advance, it can significantly cut overall costs for process plants. Plant engineers can make small fixes and repairs well in advance, instead of waiting for an expensive and important item of machinery to fail entirely.
What generally happens is that a small and relatively inexpensive part begins to fail, but that goes unnoticed by the human eye. However, that small failure triggers a domino effect which eventually causes damage to the entire large, costly piece of equipment. On top of the cost of buying replacement parts and the added employee hours that are needed to carry out the repair, it’s usually necessary to shut down part or all of the plant to allow the maintenance team access.
This wastes employee time and production time, on top of the waste of raw materials that may have been damaged by the faulty equipment, and the potential impact on customer trust if an order has to be delayed to allow for repair.
ML helps process plants go Lean
By bringing ML models into process plants and applying its unique combination of “ML + HI,”, SAM GUARD helps advance Lean principles of reducing waste, saving time, and cutting overall costs, thereby helping drive value in the process manufacturing industry.