The condition monitoring tool Watchdog - detecting machine changes at an early stage

Machine downtimes and stoppages are not only frustrating, they also incur high costs in production companies and can have a significant impact on the entire company. When a machine is down, various types of costs are incurred, such as costs to identify the fault, contain the problem, or recover. Arguably the biggest loss, however, is caused by the lost time when the machine is not running and not generating output. We want to save our customers in the beverage industry from unplanned downtime and the impact it has on overall production efficiency and output.

How can unplanned downtime be avoided?

This is where our condition monitoring service, called Watchdog, comes in. Watchdog detects changes in machine condition that, in the worst case, can negatively impact the production process as an unplanned shutdown. Condition monitoring is a process based on state-of-the-art IoT technologies that is used in production to monitor and analyze the condition of production equipment. By permanently collecting data and interpreting condition information about machines, equipment and their components, deviations can be detected and corrected in time before serious problems, malfunctions and costly production downtimes occur. Condition monitoring helps to better understand the machines, notice changes more quickly and better coordinate machine maintenance. Predictive and intelligent maintenance is of great importance in terms of optimizing production processes.

Our edge device ReadyKit, with which every new Krones line is equipped as standard, creates the perfect conditions for being able to monitor machine data. By reading raw data, forming information from it, and sending it to the cloud, ReadyKit provides the interface for collecting and processing standardized machine data.

How Watchdog works

The digital service accesses process-relevant machine information to reliably monitor defined limit values. If the corresponding limit values are exceeded or not reached, the tool immediately sends information to the responsible maintenance personnel - and thus helps to avoid problems before they arise. In summary, Watchdog can be used to transform unplanned downtime into plannable maintenance activities.

In a first step, the underlying machine and consumption data can be visualized. Necessary information such as the type of product which is produced or the status of the machine help the user with detailed data analysis. The ability to store machine data and make it accessible for long-term analysis means that trends can be identified even in the case of slow wear processes. In a second step, watches for abnormal data behavior can be created to detect anomalies in the line. The maintenance staff is then immediately informed of the need for action via SMS. In the future, more channels such as email or Microsoft Teams will also be added and linked to other services such as Tasks.

There are two ways to monitor machine data with the help of Watchdog. For customer-specific use cases, logic can be created with the help of visual logic modules. Watchdog provides a number of common statistical methods to generate a trend analysis of consumption data. Depending on the condition of the machine, leaks or operator errors can be detected in the data. For more complex algorithms, ready-made templates are available which only notify the end user and issue a recommendation for action. Based on Krones know-how, predefined rules should already be created and configured in the Watchdog customer system. A concrete use case is, for example, the monitoring of servo drives. Here, the current is calculated using meaningful statistical measures to form an index that can be easily monitored. Faults and damage to the servo drives can thus be avoided, saving the customer storage costs and unplanned downtimes.

The interaction of Watchdog with other Krones digital services such as Performance, Tasks and Assistance also enables monitoring of other key figures from the maintenance or the reporting service.

Where do we stand?

Basically, we are still at the very beginning with Watchdog. We are talking about a condition monitoring tool. This means that the more data we collect, the more statements we will be able to make in the future. It remains exciting to see whether the focus will be on monitoring human error (operator error) or mechanical wear and process-related faults. Since the critical machine information can be used effectively in all areas of production, the potential that lies in the service is very large in my eyes. In order to fully exploit this, the functions are constantly being optimized and expanded in cooperation with our customers. The great challenge lies in the enormous complexity: a large number of parameters have to be taken into account, which in turn depend on very many other parameters, such as the machine operating status, startup and shutdown behavior, the bottle produced, or the air humidity. Added to this is the difficulty that we are confronted with a broad customer base. Our customers range from small bottlers to huge corporations that have their own monitoring systems. Watchdog must therefore deliver added value that covers the use cases of as many customers as possible. In addition, the users of the products also vary - from experts to operators who have no specific training or industry knowledge. We want to meet these challenges and offer a service that actively and reliably supports our customers in avoiding production downtimes.