How to Track and Monitor Machine Downtime

As the employees who operate the machines are closest to production, they have extensive visibility into the shop floor. On another side, other authorities including shop floor managers as well as manufacturing engineers sometimes lack thorough visibility into the production status, how a team is working against production goals as well as overall machine utilization.

This an essential part of manufacturing. It is important to know when the machines are in run mode and when they’re not – and if they are in run mode – are they working the way they are designed to perform? As per Analyst firm Aberdeen Research, 82% of companies have noticed unplanned downtime over the past three years and that unscheduled downtime can cost a company about $260,000 an hour.

If you’re facing a loss of productivity from machine downtime and you’re looking for a solution, here is a brief about how to track and monitor precise downtime that will improve overall throughput.

The Concern with Manual Tracking

The problem here is generally when trying to conduct a downtime analysis, is that manually gathered data is mostly imperfect and delayed. This not only prevents certain authorities to take quick action on the data but results in an inclined view of the actual production performance of the plant.

Further, manual data collection is time-consuming and needs unnecessary data assembling. So to conduct the analysis with such data requires far more effort to interpret and utilize effectively.

Automating Machine Data Collection

• Automating the accumulation of data and assuring its accuracy is not only required to conduct a successful audit but also gives a foundation for efficient decision-making that goes far beyond downtime analysis alone.
• Utilizing a solution machine condition monitoring which gives manufacturers the capability to automatically capture precise machine data in real-time. This plug-and-play solution connects directly to the machine control and also gathers correspondent data from operators with the help of tablets placed at each machine.

In this way, manufacturers not only have precise downtime data directly from the machine but also data that describes the “why” behind these downtimes. And then out-of-the-box reports, as well as dashboards, can easily be utilized by operators, managers to better interpret production performance and react to data in real-time.

There are several types of data you will need to collect from both your machines and operators in order to run an efficient downtime analysis, including:

• Downtime category (What is the certain cause of the downtime?)
• Whether it was Planned or Unplanned Duration of downtime (Was the downtime scheduled?)
• On which machine the downtime raised on
• The time the downtime occurred (How many minutes was the machine down?)
• The shift or operator running the machine?

Additional factors such as the Cost of machine downtime are also there which are helpful in the long term.

How To Start Analyzing Machine Downtime

Begin with a 24-hour period from your most recent downtime report. This will concentrate your analysis on a manageable time period and give you sufficient data to get insights and prepare for the next day’s starting shift. Initiate with unplanned downtime, since planned time can still be minimized, it is the unplanned downtime that can generate cascading concerns to performance and affect productivity unexpectedly.

Look hour by hour to interpret when and for how much time downtime occurred. This will give you an understanding of what might have caused or amplified downtime. Was it due to the confusion of a shift change or an unplanned operator break?

Why Use a Pareto Chart to Analyze Downtime?

 Gathering data, whether it be manual or automated, is not sufficient. That data must be assembled into reports in order to be monitored. You will require a simple way to access as well as query this data, here is an automated machine monitoring solution that will make this simpler as it automatically drags the data to run analytics, populates reports and enables you to develop custom reports and dashboards

One of the most useful reports to interpret downtime reasons is the Downtime Pareto, which will combine all the logged downtime reasons. By mapping your downtime in a Pareto chart you can find out the 20% of concerns that cause 80% of downtime.

Using this report, you can simply recognize the downtime reasons. Further, you can separate the data amongst several shifts, machines cells, or even individual machines for a thorough look at where problems may lie.

Common types of downtime contain extreme tool changeover, job changeover, lack of operator and unplanned machine maintenance.

Planned versus Unplanned Downtime

Until factories are completely autonomous, there will always more chances of downtime. Planned downtime is when you plan the down periods at a time that is suitable to the company and reduces any negative impact for the users.  Whereas unplanned downtime is when an error in operations occurs due to an unplanned machine or server error.

With machine monitoring, you can better interpret what % of downtime is unplanned and notice where the data shows differences from expectations.

With the help of a machine condition monitoring system, manufacturers can follow continuous improvement initiatives knowing they have instant access to exact production data enabling them to take actionable steps to minimize downtime.

The main aim here is to constantly, precisely and effectively track downtime, so we can finish it when it happens, notice its impact on performance, interpret what causes it and avoid it in the future.

Manufacturing analytics applications like IoT-based machine monitoring, offer significant ROI for manufacturers that face such types of issues.