Unplanned Downtime? I Have A Cunning Plan, My Lord

Let’s talk about downtime. It halts production, can require expensive repairs and replacement of vital machinery, and costs manufacturers thousands. Wouldn’t it be wonderful to know, in advance, when repairs or maintenance will be needed, and build them into your schedule at a time that suits your production? The good news is you don’t need the gift of foresight or a crystal ball to make this a reality, just the right technology. FANUC’s Steve Capon explains.

Unplanned downtime is a nightmare for the manufacturing industry, yet it’s one of the few certainties that we have. Almost every business will have something go wrong at some point, whether that’s due to equipment failure, human error, or simply an act of god. This will, more often than not, halt production until the issue is resolved.

The problem is that, when downtime is unplanned, it interrupts operations in all parts of the business and can even trickle down the supply chain. In the most serious cases, downtime can impact not just the balance books, but even the brand’s reputation.

Loss of production is also problematic because it’s unquantifiable. It’s not always possible to know how long a piece of equipment will be down for, which means you won’t know how much that broken gear or worn joint will cost you until the damage is already done.

Is it really revolutionary?

The solution is essentially turning unplanned downtime into downtime that is planned through the use of condition monitoring technology. This means identifying problems before they occur by analysing data and trends. Further developments, such as those happening with Industry 4.0, allow this data to be shared, providing an open platform to collaborate and compare like never before. The truth is, however, that condition monitoring was a reality long before the phrase “Industry 4.0” was first uttered.

In its simplest form, planned downtime is scheduled maintenance for certain parts, which is often a pre-requisite in order for them to function properly. Manufacturers are usually made aware of this upon purchase, and good manufacturers will actively build this into their maintenance routine without causing disruption to production.

Modern condition monitoring, aided by diagnostic tools, moves a step beyond this. It predicts the unpredictable.

Knowledge is power

How, then, do we plan repairs and maintenance for problems that we don’t even know exist? Why not pip potential problems to the post by operating an over-zealous maintenance schedule? This is hardly the best solution. After all, manufacturers will need to find a home for all of the spare parts, and a salary to pay all of the maintenance engineers, who will mostly be sat around, waiting for something to go wrong.

You can’t have every solution on stand-by for every problem all the time, but you can gain insight into what might be needed, and when. Diagnostics tools don’t make your machine bullet-proof, but they can provide a warning about any potential problem, weeks before it actually happens. 

Some modern diagnostic tools work by connecting to a robot in order to conduct a mixture of maintenance scheduling and process health monitoring. They are pre-programmed to know how a machine should work, and use a simple traffic-light system (green, amber, red) to communicate the severity of a potential problem.

This isn’t to say that diagnostics tools are as sensitive as a kitchen smoke alarm. They won’t flash red at the first sign of change. Instead, they track it, and are designed to see patterns in machine behaviour, which can then be interpreted by the program. As the warning level increases, so too does the likelihood of the machine stopping. 

Diagnostics tools are already making their mark in the automotive industry, which has historically been the birthplace of technological innovations within manufacturing. Current iterations of diagnostic tools are best suited to this kind of environment, which has high-volume production and multiple robots.

That isn’t to say that smaller-scale production cannot benefit from diagnostic technology, but the hardware behind the technology will need to become more cost-effective in order to provide a tangible benefit. The integration of diagnostics tools into Industry 4.0-based monitoring systems will facilitate this.

What to look for

The best diagnostics tools on the market will be simple to use and easy to interpret. Many types of software can be loaded directly into the robot, and will automatically generate self-test programs. One-time self-test programs can also be executed through an external PLC.

EDGE-heavy devices are used within the environment in which they are needed, rather than via an external server. These are also useful, because they connect directly devices on the factory floor to provide real-time data. There’s also an added benefit of security, because you are using a local connection rather than exporting your data to an external data management system.