The complexity of a busy factory shop floor is rather fascinating. It is a system of multiple moving parts, both human and machine, that is expected to turn material into products efficiently, with no trouble and zero incidents.
We all know that humans aren’t perfect, but machines aren’t, either. Because of these imperfections and other factors, such as varying environmental conditions or inconsistent material quality, the system needs constant oversight and correction to avoid machine downtime and run smoothly.
This article covers:
- how to assess your losses due to downtime
- best practices for reducing downtime and increasing efficiency (OEE)
- examples of charts, spreadsheet templates, other helpful links and materials
- what are the options and tools to track machine downtime
How to Assess Your Losses Due to Downtime
The full scale of losses is much more than the actual time lost during the stoppage. It also includes:
- loss of production that affects the process flow
- scrapped products and wasted raw materials
- overtime wages for extra shifts to catch up with the schedule
- extra maintenance cycles, cost of cleaning-up
Not to mention that the loss is not just financial: taking shortcuts to make up for time lost increases the risk of incidents.
To assess the full impact of downtime, first of all, you need an accurate log of downtime events, and to make the most out of this log, you need to capture all the important details for each event.
What is the data that you need to capture?
Here is what your log of downtime events should include:
- Date and time when the stoppage started
- Duration of the downtime event
- Location (machine, line, part of the line)
- The product that was being produced
- Operators who were in charge
- The reason why it happened (ideally – from a list of standardized categories, see a separate article on how to standardize downtime reasons)
- Additional description (for example, comments on the action taken)
Now, how exactly do you reduce downtime once you have the data?
Best Practices for Reducing Downtime and Increasing Performance (OEE)
1. Visibility of your OEE on the shop floor
As soon as you have data coming in, the first recommended step is to make it visible on the shop floor. Just doing that has the “magic” effect of reducing your downtime because people are more aware of the situation, more motivated to improve it, and now they have the information to act on downtime as it happens.
OEE consists of three components: Availability, Performance and Quality. Each of them can be visualized separately and in comparison with the previous period. Here is how our client ELBAK does it:
2. Prioritization and low-hanging fruits
Once you have a bit more data (say, a couple of weeks), then the next most effective step is to prioritize downtime reasons by total duration (or by the percentage of time planned for production). Here is an example:
Then, you focus your efforts on one of the top downtime reasons, usually the one that is the easiest or quickest to solve.
Here are some simple and effective methods to address that:
- 5 why analysis on the shop floor
- Short meetings at the line (machine) with operators to get solution ideas
- Regular standups with operators to share problems and align improvement efforts
3. Solving more complicated problems
To tackle more complicated problems, there are many proven methods to choose from. Here are a few that are popular among our clients:
- Root Cause Analysis (we have an article on how to perform RCA in 6 steps)
- A3 Problem-Solving (very similar to the 6-step RCA process, we have a chapter in the RCA article for that)
- SMED (see Wikipedia article on Single-Minute Exchange of Die)
- Six Sigma and Lean Six Sigma
- TPM (here’s an article about Total Productive Maintenance on “reliableplant” that we like)
Here is an example of our client using A3 process to reduce the second biggest downtime. They reduced the daily duration of that downtime from 1 hour 6 minutes per day to 41 minutes per day in three months.
4. Being relentless about continuous improvement
The challenge with continuous improvement is that it happens easily and in big leaps at first, but after a while, more and more effort is needed to achieve smaller and smaller gains. It becomes harder to justify improvement projects. For many, it makes sense to reach a “good enough” level and focus on monitoring the situation to keep it at that optimal level.
However, the idea of continuous improvement is to continue to improve all the time and to keep raising the targets, aiming to move from good to great.
Speaking of targets, we can use the world-class OEE, which is 85% and above, to assess how many companies achieve it. If we assume that somewhere around that number is the distinction between good and great, then we can see that only about 10% of manufacturing organizations exceed the world-class score of OEE, according to our data.
What Are the Options and Tools to Track Machine Downtime
1. Manual downtime tracking
If you are not tracking your downtime at all, you can try the simplest method: manual tracking. Its pros are: low budget solution, easy to start with, minimal training required.
Here are a couple of helpful templates:
- Excel spreadsheet template for logging downtime events manually (on paper)
- Google sheet template for analyzing downtime (make a copy to your own Google Drive to edit it)
If you are already tracking downtime manually, then you probably have already realized the cons: it’s time-consuming to enter data and analyze it, the data is not accurate, it can be manipulated by workers, etc. There are so many shortfalls of manual tracking that we have a separate article for that.
The next step is to start using automated tools to record and visualize your downtime.
2. Using a dedicated OEE software solution
A solution can be considered dedicated if it can:
- Register production in real-time as it is happening (either with sensors and dedicated IIoT devices or by connecting directly to production machines – some machines are capable of providing data about their activity)
- Visualize production data in real-time
- Automatically record data from the machines and store it
- Provide the possibility to enter additional data, such as explanations why downtime occurred
- Provide access to historical data for purposes of analysis
Pros: as you can see from the list above, data captured in this way is accurate and occurs in real-time. You can collect multiple other parameters about the process in addition to downtime.
Dedicated solutions can be on-premise or SaaS (Software as a Service). SaaS solutions have the added benefit of regular updates and new features.
Cons: you need to buy and install it separately, in addition to your other systems. This involves researching available solutions, perhaps trying out a few before you settle on the one that suits you.
3. Creating your own in-house solution
This option may be suitable for large companies with specific needs that cannot be covered by existing solutions. For example, integration with company systems is a requirement. Evocon does offer integrations, but other providers may not, or your systems may be very specific.
- full control over the solution
- in-house support
- no subscription fees
- the cost of development and maintenance has to be covered by only one company
- you need to have an existing IT department capable of building such solutions
- long time to develop the solution: between a few months and a few years
- high maintenance cost
- upkeep of the system can become an issue if a key member of the development team leaves the company
We hope this helps you make the right choice.
We have provided you with a good starting point. Now you know how to assess the impact of machine downtime. We provided some templates to help you with that and an overview of tools and methods you can apply.
To go more in-depth, there are a number of topics to pursue, according to your specific interest. Please go on and browse more of our content – we have quite a library, both on fundamentals of OEE and advice on how to increase your productivity.