Predictive Message in a Bottle: A modern approach to an age-old process

The art of making alcohol can be traced back to a time long before 7000 BCE where fruits and grains were fermented for personal consumption and pleasure. However, in this modern age this same process is industrialized for production of vast quantities and worldwide distribution making it more akin to precise chemical manufacturing than a weekend pastime.

The machinery used in fermentation, testing, aging, bottling, packaging and then distribution mean that the alcohol industry is perfectly positioned to take up the revolution of online Predictive maintenance (PdM), ensuring quality and quantity keeps up with the demands for your brand.  

Distilleries & breweries are big, complex and energy intensive. They run processes where seemingly minor issues with equipment can lead to major problems with product quality.

They also operate 24/7 and downtime can typically cost up to $40,000 an hour in lost production.  Predictive maintenance can overcome these challenges. It can reduce downtime and maintenance costs. It can also help to optimize production performance, safeguarding product quality and reducing energy bills. The potential impact on profitability is enormous.  

Predictive maintenance may seem like a niche topic when you have warehouses full of stoke aging and waiting to mature, however, distillery owners should not underestimate the big difference predictive maintenance could make to their businesses. Introducing a program of predictive maintenance can make a major contribution to profitability, as companies in many other industrial sectors are already discovering.  

In addition, distillery & brewery operators face the same demographic challenges as businesses across the aging economies of the developed world, with 70% of experienced workers set to retire in the next 15 years.

Companies need to find ways to prevent all that expertise being lost to the business. Machine learning and artificial intelligence embedded in the latest automated systems can help to counter the evolving skills shortages.  

What difference can predictive maintenance make?

Predictive maintenance relies on monitoring the condition of machinery to identify much earlier when something is going wrong. This enables engineers to fix problems before they cause a breakdown that would impact on performance or disrupt production. So how much difference could that make in a Distillery or brewery?  

Take a simple bearing failure as a common example of what can go wrong. Manufacturers give an expected useful life for every bearing, but estimates suggest that fewer than 10% of bearings reach that limit before failing. This means that more than nine out of ten bearing failures are premature. In other words, they are only avoidable with the right condition monitoring systems in place to warn when a failure is imminent.  

In an industry where there may be thousands of bearings installed across a single site and unplanned downtime from machinery failure typically costs $40,000 per hour in lost production, it is easy to see how preventing these breakdowns can impact on profitability. 

As well as reducing downtime by up to 50%, other benefits of predictive maintenance include lower labor costs, optimized management of spares and the avoidance of any secondary damage to the distillery or brewery or to product quality. 

Modern solutions promise persuasive ROI 

Predictive maintenance systems demand effective condition monitoring – watching for changing patterns of machine behavior that could be a sign of trouble brewing.

That used to mean data experts having to laboriously monitor and analyze the data coming in from individual machines. But the latest automated systems use advanced machine learning algorithms to provide condition monitoring without the need for human intervention.  

These automated condition monitoring solutions are also cloud-based and readily scalable, making it easy to test them out on a few machines to start with, before rolling them out seamlessly across the entire operation.

This cloud-based software as a service (SaaS) delivery model means that the price per machine usually drops away sharply as the number of machines covered by the predictive maintenance deployment grows.  

By slashing the investment, time and effort needed to implement condition monitoring, the new generation of smart solutions have shifted the return on investment (ROI) equation strongly in favor of predictive maintenance.

For instance, experience shows that Senseye customers can expect to recoup the cost of their subscriptions between five and ten times in the first year alone, plus:

  • 10-40% lower maintenance costs 
  • 30-50% reduction in downtime 
  • 45-55% increased productivity 
  • 85% increase in maintenance accuracy 

The Senseye PdM cloud-based predictive maintenance solution takes data from existing sensors around a site and transforms it into information about the condition of the machinery it is monitoring. Existing users include blue chip companies in manufacturing, heavy industry, automotive and FMCG, who typically enjoy a 50% reduction in unplanned downtime.  

The system is designed to begin learning from day one and starts providing useful insights in as little as 14 days. Operators can prime the system upfront with helpful information - such as the data recorded in the run up to previous failures, for example - but the algorithms are designed to start from scratch if need be.  

While most condition monitoring systems focus on abstract concepts of ‘machine health’, Senseye PdM rapidly learns to direct the operator’s attention to their most pressing maintenance priorities using an Attention Index.

Whenever Senseye PdM raises an alert, the operator can indicate at the touch of button whether that alert is useful or not. This gradually teaches the system to direct the operator’s attention towards the most important trends or events, rather than bombarding them with low level alerts from all directions. This is especially helpful in major deployments that can cover hundreds or even thousands of assets.  

While Senseye PdM starts delivering effective support immediately, the eventual goal is to reach the point where it can deliver accurate forecasts of the remaining useful life (RUL) of every asset – a technique known as prognostics. It is like having an experienced operator on hand who knows when a rattling pump needs immediate attention and when it can safely be left until the next planned shutdown.  

Is Senseye PdM right for me? 

Want to find out more about how Senseye PdM can help to boost your profitability? Book a meeting with us today.