Improving Setup Times with Industry 4.0

Introduction

Reducing setup times has always been a core focus for lean manufacturing for many reasons. In addition to simply improving the amount of time available for production, reducing setup time also increases the agility of the manufacturing process.?The end goal for setup reduction is to enable single-piece flow where the economic lot size can be as small as one. There are many approaches to reducing the amount of setup time required. Setup reduction techniques such as “SMED” (Single-Minute Exchange of Dies) facilitate reducing the amount of time for each setup. In this article, we’ll cover how to supercharge these efforts with Industry 4.0 / Smart Manufacturing solutions.

The Importance of Setup Time

OEE’s “Six Big Losses”

Overall Equipment Effectiveness (OEE) is a common metric in manufacturing used to capture what percent of capacity is being utilized effectively. The capacity losses are typically categorized into six categories – one of which is defined around Setup and Adjustments. This is considered one of the “Big Six” losses because it represents a substantial loss of capacity across many industries, particularly for companies in high-mix environments.

The loss category is described as “Setups and Adjustments” because of the time lost not only to the initial setup efforts, but also any adjustments to the equipment that are required. For companies that have not effectively eliminated sources of variation during the setup process, the adjustments can sometimes take as long as the initial setup. This variation during the setup process can also lead to quality losses from startup rejects at the beginning of each batch / mfg order.

Agility vs Efficiency

Another consideration that has become much more important in the past couple of years, is the idea of agility versus efficiency. With efficiency, we are trying to minimize the cost of production and distribution, carry minimal amounts of inventory while having minimal amounts of excess capacity, and running things as close to the bone as we possibly can while still getting the results that we're looking for.

Agility allows us to maximize the responsiveness to market dynamics and unexpected events. In the past couple years, we have seen plant shutdowns due to the pandemic, wild swings in demand profiles and supply chain issues. There have been labor issues - people not showing up for work due to illness and personnel churn due to people retiring or leaving their jobs. How can a company maximize not just agility, but robustness and resilience to these different disruptions?

Some of that is keeping more inventory. Some of that is having excess capacity in different areas, so that if this type of product doesn't show up, we can utilize capacity to do that product instead. There is a trade-off between the efficiency we crave and the agility we require. We need to balance keeping the company agile while still pushing for efficiency as much as possible.

Then there are activities such as reducing setup times that can improve both efficiency and agility.

Shift the Tradeoff Curve

As explained above, reducing setup times will improve efficiency by eliminating some machine downtime and quality losses during changeovers.

In addition, reducing setup times significantly improves the agility of the organization by eliminating friction from moving between production of different products. This is important for many reasons.

First, reducing setup times helps the plant track closer to customer demand. If there was no time to change between products (and if the materials were on hand), it might be possible for the plant to simply produce what customers buy each day. This could be in the form of make to order, or it could simply be replenishing finished goods inventory based on daily consumption.

Even though this type of production is not possible in many environments due to supply chain constraints or highly variable daily demand quantities, being able to reduce setup times will still help companies track more closely to customer demand and reduce finished goods inventory.

Benefits of Setup Reduction

Let’s summarize the benefits of setup reduction by looking at the graphic below. The starting point here is the “Normal Setups” where going through the typical product cycle on this line of A->B->C->D->E->A takes a significant amount of time when economic lot sizes are used.

When the shift is made to reduced setup times, there are two compounding impacts on being able to cycle through those products much more quickly. The first is that the time to produce a lot of each item is reduced due to the elimination of a portion of the setup time between products. In addition, smaller economic lot sizes of each product can be run because of the gains in efficiency.

If it previously took two weeks to cycle from one run of product A to the next run of product A, the safety stock in finished goods inventory had to account for those two weeks. If it now only takes one week to run through that cycle, finished goods inventory can be reduced by a full week.

A more subtle impact is a change in mindset from production. When changeovers between products take a lot of time, there are a lot of comments like the ones in the graphic.

Manufacturing will push for large batch sizes. They’re often measured on production totals and on efficiency. If they must do too many long changeovers, their numbers will suffer. So, they push against changeovers during Sales and Operations Planning (S&OP) meetings. They also push back against plan changes and try to freeze as much of the production plan as possible.

Here are some of the benefits of smaller lot sizes and shorter setup times:

• Shorter Product Cycles and Product Lead Time
• Greater Flexibility
• Improved Efficiency
• Improved Planning and Scheduling
• Reduced Variability
• Reduced WIP
• Reduced Finished Goods Inventory
• Fewer Defects
• Less Obsolescence
• Less Firefighting

Improving Setup Reduction with Industry 4.0

Reducing setup times is obviously an important objective in manufacturing. This section will take a deeper look at how this can be done.

Where Can Setups Be Most Improved?

This chart represents areas where setups can be most improved. The vertical axis is the typical impact, and the horizontal axis is the time to implement the changes.

In general, the fastest and highest impact changes involve standardization, quick process improvements, and clear definitions of roles and responsibilities. Often, simply looking at who is currently doing the best job setting up the process and standardizing on that sequence of steps will yield the biggest gains. Once everyone is performing the setup according to the standard work, then additional changes can be made as a next step.

Next are changes to the process such as retrieving required parts before the changeover starts. This seems basic, but any parts or tooling required for the change should be delivered to the work area prior to the end of the current production run. Standard settings should be marked on the equipment to make it easier to adhere to the standard work. Any waiting or unnecessary motion should be identified and eliminated.

These activities do not require much or any changes to the physical equipment and can be very fast to implement. Once these activities have been completed, then we can look at projects such as installing quick release mechanisms, modularizing equipment, and other larger scale projects.

The next sections will take a look at these improvements in greater detail.

Example Setup Process

First, let’s define some terms by going through an example. (As an aside, I’d like to cite where I got this graphic, but I pulled it off the internet years ago and can’t find the link).

In this example, the timeline is divided into three main sections: production time for the previous run; production is stopped for the change; production begins on the next product. The boundaries are defined by the last good part for “Part A” and the first good part for “Part B”.

Any tasks that are performed while the process is running are said to be “External”. Tasks that are performed while the process is stopped are said to be “Internal”.

The details on the timeline show example activities that often take place during a changeover such as getting tools and material, removing old fixtures and placing new ones, modifying dimensions, performing trial runs and making adjustments.

As we will discuss in a minute, a major part of setup reduction is to identify these tasks in detail and make as many of them “External” as possible.

A Typical Setup Reduction Process

The graphic to the right outlines the steps in a typical setup reduction process.

Prior to the beginning of the reduction process, the tasks need to be identified in detail and timed. This may need to be done many times if setups vary depending on many factors. If standard work is not in place, then the tasks and timings may vary significantly from worker to worker. The tasks may also be highly dependent on which products are ending production and beginning production. Some changes may be very minor if form factors are similar, but very intricate and extensive if the form factor is different.

It is also important during this process to understand the settings on the machines that work best, any sequencing requirements or dependencies between the steps, and more. In general, it is important to detail as much of the manufacturing process as possible before beginning the optimization process.

There are many excellent resources for the setup reduction process itself. We will just list the steps here with a brief explanation of each. The graphic above shows the impact of the steps on the setup process.

1.?????Separate Internal Activities from External: When the process begins, the internal and external tasks are often interweaved. In this step, those tasks are re-sequenced so that the internal tasks are all completed back to back in the middle of the process

2.?????Standardize External Activities: Create standard work around each of these tasks so that they are performed the same way each time, by everyone

3.?????Convert Internal to External: In this step, internal tasks are adapted so that they can be performed while the manufacturing process is running

4.?????Improve Internal Activities: Reduce the amount of time the internal tasks require

5.?????Improve External Activities: Reduce the amount of time the external tasks require

6.?????Automate Where Possible: Eliminate variation from standard work by automating any of the tasks where automation is possible and economical

7.?????Eliminate Where Possible: When possible, eliminate tasks entirely through process change or modifying the equipment

The final step is to standardize the new process and maintain it over time. Using this improvement process, I have seen the “Internal” setup times reduced by 80-90% in some cases.

Improving Setup Reduction with Industry 4.0

Now let’s look at how Industry 4.0 solutions can improve the setup time reduction process. In general, these solutions are going to help identify the tasks, capture the best (and worst) practices, document the variation in the process, help with the analysis, and then help standardize and maintain the new practices in the future.

Impacts During the Reduction Process

The impacts of these technologies can be split two ways. The first is the impact on the setup reduction process itself. We’ll explore that first in this section by looking at some of the technologies of Industry 4.0 and the impact they can provide. When we’ve completed this review, we will look at how to sustain the improvement and standardize the new procedures.

Augmented Reality / Expert Capture

One method of making it easier to capture the current process is to use augmented reality solutions to perform “Expert Capture”. This process is facilitated by the operator wearing AR glasses while performing the operation using their current procedures. As they are recording video and taking pictures of the process, they can also give commentary and denote which steps they are performing. This can be done repeatedly for different products or with different operators to capture all the variations in the setup process.

Capturing the process in this manner allows the true steps in the process to be captured. When the process is documented through interview processes, there are often gaps in the documentation or the process is documented the way it is “supposed” to happen instead of the actual process. Even when time studies are performed by the team, it is very expensive to have the additional resources at the machine to capture all the variations of the process across all the different products and operators. So, a narrower picture of the process often emerges.

In addition to capturing the steps in the process more effectively using the AR glasses, these can also help document the machine settings that are being used by each different operator for each product combination.

Finally, the AR solutions are fantastic for documenting the to-be process for training. As the setup reduction process is completed, the new procedure can be performed by the team and captured with the Expert Capture solution. This augmented video can then be used to train the existing operators on the new process, as well as any operators who are new to that process. The video can be viewed on a regular tablet with detailed process instructions, on-screen pointers and more. Using the AR glasses again, the instructions can even be shown on top of the machine itself with detailed guidance on how to perform the setup as efficiently as possible.

Spatial Computing / RFID

Spatial computing uses video analysis to automatically capture steps in a process and how everything relates based on position in 3D space. This can be used in a very similar fashion to what was covered in the Augmented Reality section above. RFID can also be used for spatial analysis, though not in nearly as much detail as through spatial computing.

In addition to what was covered earlier, these technologies can be used to do spaghetti charting for the initial setup reduction analysis. Spaghetti charting tracks the movement of all the people and materials involved in the process to identify wasted movements.

Once again, a big benefit to this approach is the ability to record and analyze a large number of setups by different operators and across different product combinations. This will give a much truer picture of the current process than the traditional methods of getting data through interviews or manual time studies.

Another advantage of the spatial computing over those techniques (or even AR) is that when the video is used over time to record the process, people will tend to drift back to the way they normally run the process. When a manual time study is being performed or when the operator is using the glasses to capture the process, they may not behave the way they normally would. This can introduce bias into the study and diminish the results.

Industrial Internet of Things (I-IoT)

I-IoT is where most every device in the plant is connected to a system backbone that can then persist the data, pass it to other applications, or analyze that data to take various actions. For setup reductions, this type of platform has many purposes.

The first of these is to capture the timing of each of the individual steps. One basic way to think about how this could work is to put an “easy button” by each station to record when each step in the process is completed. So even without the AR glasses, spatial computing or anything too sophisticated in place, we can begin to capture the times using the easy button and some basic tracking software.

Naturally, I-IoT software can be used for a much more comprehensive data capture process, as well. For any smart devices being used during the setup, the I-IoT system can connect to those devices to capture the start and end times, along with the settings that were input to the device, and any data tags or process variables that can be captured during the process. These settings can then be used as part of the analysis and eventual documentation.

In some cases, different operators use different settings for each production run. There can be some difficulty in determining which of those settings are the true ideal that should be used as the standard work going forward. It can sometimes take significant trial and error to see what the settings should be to produce parts at the highest quality at the fastest pace. Those ideal settings can also be different based on the product combination for the setup. They can also be influenced by environmental conditions such as temperature or humidity.

However, if data is being captured 24/7 through an I-IoT system, then the analysis can be greatly simplified. The system will be able to show quality and production rates with different input settings and operating conditions. Sophisticated platforms can even automate the analysis to determine the ideal settings under different conditions.

One of the keys to this is to use the I-IoT system to capture all the contextual information as the process is being performed. This is information such as who is running the process, the product being produced, environmental conditions and anything else that may be a potential factor in the analysis. If this information is captured, then it can be used in the analysis.

Sustaining Short and Standardized Setups

Once the setup reduction process has been completed and the new process is fully documented, the next step is to implement and train people on the new methods. Then there must be a focus to standardize and sustain the new process. Without this focus what will often happen is that the process performance will degrade over time. New people will come into the process, the products being manufactured in that area might change, and other changes may come about that will lead to that performance erosion.

In this section, we will cover how Industry 4.0 solutions can help prevent that performance degradation.

Augmented Reality and Expert Capture

The first tool for documenting the process is also the first tool for implementing and sustaining the new process. Earlier, we talked about documenting the new process using the AR and Expert Capture solutions. With these solutions ready to go with the new process, we can now train all the workers using the tools. Augmented Reality is a tremendous solution for people to learn the process by having visual cues laid over the top of the actual assets in their field of vision. This type of immersive learning both accelerates how fast the workers can get up to speed and helps them retain those lessons for longer.

Additionally, since these are available through the software, the work instructions can be used on an ongoing basis for two other purposes. The first is to train workers who may be new to the process – whether they are new employees, someone shifting to a new department, or people who are cross-training. The same benefits of accelerated training and better retention will be available here.

The second additional benefit is that the AR solutions can be used within the process itself when someone who has not yet been trained has to perform the task. This can help to cover the task during vacations, absences, or other situations where someone completely new has to step in. Because of the visual nature of the available training, the solution will guide the operator through each step of the process while giving visual cues on how each step should be performed.

Spatial Computing

As mentioned earlier, these solutions offer a way to use video to automatically identify when tasks are being performed and how much movement is taking place. These solutions are still maturing in their capabilities, but they currently offer intriguing capabilities that are rapidly increasing. They can be used to detect when individual tasks are being performed as defined in the standard work, when there are variances, and capture the video to reinforce proper methods with the workers.

This has obvious implications for avoiding performance degradation. The work can be tracked in near real-time on an ongoing basis to see when standards are not being followed. This can lead to dramatic increases in standardization and in sustaining the new process. This may only be applicable in limited domains today, but this will be a tremendous tool as the solutions in the space gain maturity and new capabilities.

Industrial Internet of Things (I-IoT)

I-IoT provides another tool for tracking adherence to standards and creating alerts when those standards are not followed. While these tools are currently more focused on the devices than the humans, they can give a lot of insight into what steps are being performed, how long they are taking, the machine settings being utilized and much more. As with Spatial Computing, this gives the ability to track deviations from the standards and reinforce proper procedures to the workforce.

These solutions also provide the capability to provide feedback to the workers on the shop floor. Something as simple as a setup countdown clock can dramatically reduce the amount of variation in setup completion time. Additionally, these solutions can also deliver the work instructions to the shop floor and be used for coordinating the required activities across many people. For example, if tooling or other materials will be needed at the machine for the setup, these platforms can be used to create a workflow and notify the necessary people long enough in advance for the materials to be gathered and delivered to the line prior to the setup.

This type of workflow and coordination can help standardize the process and ensure adherence over time.

Planning and Scheduling

Another benefit from the I-IoT system is the ability to connect and coordinate with other systems. Scheduling, in particular, is very sensitive to a number of variables in the process. For example, being able to pass the initial conditions of the factory to the scheduling system allows the first shift of the schedule to accurately reflect the capabilities in manufacturing based on how the machines are currently set up, how far each work order is from changeover at the current machine, etc.

The scheduling system also needs accurate run times and expected set up times to be able to put everything into the proper place to minimize changeover times while maximizing overall output. These times can be quite complex in certain environments when there are a number of different setup types based on differing requirements to go from product A to product B or product B to product C, or product C to product A. For example, in some environments it is much easier to go from light colors to darker colors. Or to run a setup to a similarly sized product. Being able to provide accurate setup estimates for each of the combinations in this hierarchy or ring is vital to the optimization process.

Conclusion and Summary

In summary, reducing setup times across the organization is critical to improving both manufacturing capacity and agility. Lean manufacturing and other operational excellence practices provide a fantastic foundation for making improvements. Enhancing those techniques using Industry 4.0 solutions such as Augmented Reality, I-IoT and Spatial Computing can help make the setup reduction process much more effective. They can also provide great assistance with standardizing and sustaining the gains over time.