Select Topics

Posts by Tag

See all

Latest Blog Post

22 Signs That a Robotic Cell Has Problems

Alex Owen-Hill
by Alex Owen-Hill. Last updated on Mar 28, 2018 7:00 AM
Posted on Mar 28, 2018 7:00 AM. 5 min read time

How can you tell if your robotic cell has problems? These 22 warning signs will help you get your cell running at full capacity in no time.



Sparks fly, people shout, and warning lights flash. Plumes of black smoke spew from the machine, while workers rush to the nearest fire extinguisher. A deafening alarm blares throughout the workshop as the company lawyer calmly dials the number of the insurance company.

Sometimes it's obvious when a machine breaks down. Other times it isn't.

In fact, most of the time our machines don't break down completely. If they did, that would suggest a severe lack of proper maintenance. Instead, they just slowly stop working as effectively as they could.

The same is true with robots. Most of the time, it's not immediately obvious when a robotic cell has problems.

So how can you tell when your robot has encountered an issue? What we need is a list of warning signs – signs that we can watch out for to tell when a cell has problems. Well, here you go!

1. Piles of work-in-progress (WIP) 

A robotic cell with problems quickly becomes a bottleneck. Often, the first sign of a problem is that WIP starts building up in the stages before the robotic cell.

2. Concerned team members

Employees on the shop floor are frequently the first to notice when a cell is not performing properly. For example, they may raise concerns that there's not enough (or too much) product coming into their own work station.

3. Low productivity

When there's an unexpected fall in overall productivity, you can be sure there's a problem somewhere in production. The robotic cell might be the cause of this problem, but you'll have to investigate further to be sure.

4. Product defects and reworking

Another warning sign is when an excessive number of products need to be reworked. As with the previous issue, this requires further investigation to establish the root of the problem.

5. An idle robot

A rather obvious sign that a robot has problems is when it sits doing nothing for long periods of time. However, it can take awhile for anyone to notice when a robot is idle, unless you use a monitoring software with automatic notifications.

6. Error messages

If you use monitoring software like Insights, it will notify you as soon as a robot is idle or encounters an error. You'll receive an email and/or SMS message telling you immediately that the robot has encountered a problem.

7. Emergency stop alerts

The first type of error alert you can receive from Insights is an Emergency Stop. This is usually triggered by an external input (such as a safety sensor or emergency stop button) and it leads to the robot's power being completely turned off.

8. Protective stop alerts

The second type of error alert you can receive from Insights is a Protective Stop. This also turns off the robot's power, but it is triggered by an internal trigger – usually a safety-related part of the robot's control system.

9. Safeguard stop alerts

The final error alert you can receive from Insights is a Safeguard Stop. This is similar to an Emergency Stop in that it's usually triggered by an external input (e.g. a safety sensor). However, it only pauses the robot, instead of powering it down completely.

10. Insights error flags

Along with your automatic notifications, you can view robot errors in the Insights software itself. These show up as red error flags on, and you can also see how long the error was active before the robot was reset.

11. Low utilization

When a robot is sitting idle – either due to an error or lack of work – it's not being productive. This is reflected in the Utilization metric, which is one of the most important KPIs for collaborative robots.

12. Lots of unnecessary movements

The opposite of a robot that sits idle is a robot that moves too much. Your robot should move as efficiently as possible, or else it will waste both time and energy.

13. Low efficiency

Another important cobot metric is Efficiency. This indicates the amount of time that a robot is doing productive work when it's running a program. Low Efficiency suggests that the robot is waiting too long for other processes to finish, or is encountering errors too often.

14. Everything out of sync

Production relies on smooth communication between cells. When robots are involved, inter-cell communication is especially important. There are at least five signs that robotic cell communications have broken down, which you can read about in the article How to Ensure Your Cells Are Talking to Each Other.

15. Many long disconnections

The most obvious sign that a robot is having connection problems is when it disconnects from the network. If you're using Insights, such connection problems show up in the Disconnected metric, which will be high if a robot has connection issues.

16. Connection failures

Failed communication between a robotic cell and another machine can halt production completely. Such problems can show up in several ways, but a common sign is an error message on the robot's teach pendant.

17. Teach pendant error flags

Not all robot problems will show up on Insights, which is focused on productivity metrics. Fortunately, you can get a lot of low-level information from the robot teach pendant. The pendant will display errors regarding connection, programming, software, etc. This is especially helpful when debugging.

18. Software upgrade issues

As with any software, upgrades can cause problems. Whenever you install an upgrade to your robot, be especially vigilant in the hours and days afterwards so that you can react quickly to any problems that may arise.

19. Overheating motors

Although this is more of an issue with high-volume industrial robots, overheating motors can happen to any robot. A hot robot could indicate bad programming, a faulty motor, or another hardware issue.

20. Singularities

Singularities occur when a robot tries to move to an impossible position. See our previous articles for more details: Why Singularities Can Ruin Your Day and Dealing with Singularities on Universal Robots.

21. Long wait times

Another important metric for collaborative robots is the Wait Time. This reflects how long a robot was waiting for external signals, such as for parts to arrive or other machines to finish. Excessively long wait times often indicate inefficiency.

22. Few cycles completed

Ultimately, the most important measure of a robotic cell's success is how many products it is able to process. On Insights, this is reflected in the Cycles Completed metric. If this number is unexpectedly low, it could indicate that the robot encountered some problems.

You can try out Insights for yourself within a matter of minutes by going to the Insights Demo at Try playing around with the interface and see if you can work out where the demo was experiencing some problems.

Measure your robot's performance with Insights

Leave a comment

Alex Owen-Hill
Written by Alex Owen-Hill
Alex Owen-Hill is a freelance writer and public speaker who blogs about a large range of topics, including science, presentation skills at, storytelling and (of course) robotics. He completed a PhD in Telerobotics from Universidad Politecnica de Madrid as part of the PURESAFE project, in collaboration with CERN. As a recovering academic, he maintains a firm foot in the robotics world by blogging about industrial robotics.
Connect with the writer:

Related posts

5 Issues to Solve Before Monitoring Robot Performance

There are some problems that software just can't solve. Insights can help to improve any collaborative robot application, you...

Alex Owen-Hill
By Alex Owen-Hill - January 28, 2020
Get More Out of Insights in Each Subscription Plan

Insights users can now include their custom KPIs in their email reports! They can also receive alerts for specific user defined...

David Maltais
By David Maltais - April 17, 2018
Improve your Robot's Performance with Analytics

How can you improve your cobot's performance? What are the best metrics to use? This new eBook is the essential guide!


Alex Owen-Hill
By Alex Owen-Hill - March 29, 2018