SMED Lean: What Is SMED, How Does It Reduce Changeovers and Improve OEE?

SMED lean helps reduce changeover time, meaning downtime that many plants treat as a fixed part of the production plan, even though it directly reduces machine availability and lowers OEE. In this article, you will learn what SMED is, how it works in lean manufacturing, where time most often disappears during a format or tooling change, and how to get started without creating a project larger than the problem itself.

This guide is especially useful for people responsible for production, maintenance, planning, automation, or operational efficiency. It will be particularly helpful when changeovers take longer than they should, the production plan requires more and more corrections, and the team says that’s just how it works here.

SMED is a method for reducing the changeover time of machines, equipment, and production processes. The acronym stands for Single-Minute Exchange of Die, which means changing a tool or die in a single-digit number of minutes. However, the under 10 minutes target should not be treated literally in every process. The point of the method is to systematically reduce the time needed to switch an order, format, tool, recipe, or product variant.

The simplest SMED definition is this: separate activities that can be performed while the machine is running from those that require the machine to stop. Then remove unnecessary steps, organize tools, simplify adjustments, and build a standard that works even when the most experienced person is not on shift.

Descriptions of the method often return to four key actions: recording operations, separating activities, converting internal work into external work, and improving the changeover itself. This shows an important point: SMED lean does not begin with a written instruction. It begins with observing what actually happens at the machine.

SMED in lean addresses a problem that is easy to normalize on the shop floor. A machine is stopped because a tool must be changed, cleaning is required, parameters need to be adjusted, first pieces must be tested, or quality approval is pending. Once this type of downtime is included in the plan, people gradually stop noticing it. But it still takes away productive time.

OEE exposes this loss very clearly. The indicator is based on availability, performance, and quality. Changeovers primarily affect availability because they reduce the time during which the machine can actually produce. That is why SMED directly supports OEE improvement by eliminating time losses related to changeovers.

SMED helps recover time that later becomes visible in the production schedule, line availability, and the number of orders that can be completed without adding another shift.

Most lost minutes disappear into activities that have become routine over the years. A tool may be missing, someone may be searching for a fastening element, new format settings may depend on an operator’s memory because there are no reference points. Samples may be waiting for approval while the line is already stopped.

The most common sources of waste include:

In lean and production management publications, SMED is often described as a method for reducing waste related to machine changeover time. On the shop floor, it is a simple shift in thinking: you do not reduce changeover time by intuition. You check which activities stop production even though they could have happened earlier.

A good SMED project should focus on one line or machine where changeovers are frequent, repeatable, and have a real impact on the production plan. This allows the team to quickly see whether the improvements make sense.

First, measure the actual changeover time. Ideally, do this with a breakdown by activity, not only as a single value entered in a report. Video recording can help, as long as people understand why it is being done. The goal is to analyze the process, not to look for someone to blame.

The next step is to divide the work into internal and external activities. Internal activities require the machine to be stopped. External activities can be completed earlier, while the line is still running. This is where many companies find the first minutes they can recover, because preparing tools, materials, documentation, and settings does not have to wait until downtime begins.

Then you start moving work outside the machine stoppage window. If guides, parts, labels, settings, templates, or documents can be ready in advance, they should be waiting at the workstation before the changeover begins. Dedicated storage points, short checklists, labeled tools, and clear responsibility all help.

Some activities will still have to be performed while the machine is stopped. Those need to be shortened. Quick-release connectors, stops, guides, markings, saved settings, setup templates, and a better work sequence can all help. At this stage, SMED brings together operators, maintenance, process engineering, and quality, because each of these areas affects downtime.

The final step is the standard. It should be short, clear, and possible to use during a real changeover. If the standard requires long reading or nobody returns to it, it is not a work standard. It is just a document.

The value of the method is easiest to see in typical shop floor situations.

In a packaging plant, a format change may require stopping the line, replacing guides, adjusting width, testing packages, and approving the first pieces. After analysis, the team may discover that guides can be prepared earlier, settings can be saved for each format, and quality control can be initiated before the line is ready to restart. The machine will still be stopped for some time, but for less time, because part of the work no longer starts only after the machine has stopped.

The second example relates to labeling. Polish academic publications have described the use of SMED to reduce the changeover time of a labeling machine. This is a good case because changing a label or format often requires new guide settings, first-piece inspection, and corrections after startup. If preparation begins only when the line is already stopped, every small mismatch extends downtime.

These SMED examples lead to one conclusion: the biggest improvement often happens before the changeover itself. Good preparation can reduce downtime more effectively than rushing around a stopped machine.

SMED lean needs data. Without data, it is easy to assume that a changeover went well, even when the numbers show something different. At the beginning, you do not need to build a complex model. A few metrics are enough to support better decisions.

Measure:

The average alone can be misleading. One changeover may take 25 minutes, another may take 70, and in the monthly report the result may still look acceptable. This variation often reveals a lack of standardization, poor preparation, or dependence on one person.

The most common failure is starting with an instruction. Someone describes a new way of working, but nobody has first checked what actually extends the downtime. Such a standard quickly loses against everyday reality.

The second mistake is shifting the entire topic to operators. Changeover time also depends on planning, tool availability, quality, maintenance, process engineering, and data in production systems. If each department only looks at its own fragment, the whole process will still lose time.

The third mistake is quieter: no measurement after the improvement. One successful changeover does not prove anything yet. A standard has value only when it works across different shifts, with normal staffing, and without special supervision.

Choose one changeover that causes the biggest problem for production. It does not have to be the longest one. A better choice is a changeover that happens often and affects the plan, OEE, or on-time order completion.

The first step can look like this:

This kind of start does not require rebuilding the entire way of working. It gives the team specific material for discussion: what works, what needs improvement, and where the data does not match people’s impressions.

SMED lean delivers the best results when the team can see actual changeover times, causes of delays, and the impact on OEE. If data is entered manually, scattered across spreadsheets, or dependent on a few people, decisions come too late.

At explitia, we can help you look at changeovers through data: downtime duration, loss categories, machine availability, shift reports, and the connection with OEE. This makes it easier to identify which format, tool, or batch changes have the strongest negative impact on performance.

This is a good place to start if you want to see where production is losing availability instead of choosing an improvement area based on intuition. When the numbers show the source of the problem, SMED can become an improvement plan for a line, process, or group of machines.

SMED lean makes sense when changeovers are a visible time loss and you want to increase availability without buying another machine. First, you need to measure the actual course of the changeover. Then separate internal and external activities, shorten what cannot be moved, and maintain the standard across future shifts.

Start by finding out how much time your production loses to waiting, searching, corrections, and lack of preparation. That will show you where SMED can improve OEE the fastest.

SMED is a method for reducing changeover time in machines and production processes. It is based on analyzing activities, separating work performed while the machine is stopped from work that can be prepared earlier, and building a changeover standard.

The SMED definition most often refers to Single-Minute Exchange of Die, meaning a changeover completed in a single-digit number of minutes. In practice, it means systematically reducing the time needed to change a tool, format, batch, recipe, or product variant.

SMED in lean reduces planned and repeatable downtime related to changeovers. As a result, it increases machine availability, which is one of the components of OEE.

Simple SMED examples include preparing tools before the machine stops, saving settings for specific formats, using quick-release connectors, marking adjustment points, or using a changeover cart with all required elements.

It is best to start with one frequently repeated changeover. Measure the time, list the activities, separate internal work from external work, move preparation outside machine downtime, and check the result over the next several shifts.