Lean Production: Lean Filling Processes for Powders and Granules

A bagging process for bulk materials consists of a defined sequence of process steps—bag positioning, dosing, filling, sealing, palletizing—and each of these steps has its own bottlenecks and sources of loss. During bag positioning, positioning accuracy determines the error rate. During dosing, the switch between coarse and fine flow determines the cycle time. During sealing, the quality of the weld seam determines whether the bag reaches the customer or bursts during transport.

Lean Production makes these losses visible and eliminates them methodically—step by step, directly at the packing machine. In contrast to the strategic Lean principles behind bagging optimization, which ask where waste arises throughout the entire value chain, Lean Production focuses on the shop floor: on what physically happens between the silo outlet and the finished pallet.

Hänggi, Fimpel, and Siegenthaler sum it up in a formula: produce without waste and without detours—the right product, in the right place, at the right time, in the right quantity, and with the right quality. In the bagging of powders and granulates, this means: every bag filled to the exact weight, cleanly sealed, and stably palletized—without waiting times, without rework, and without unnecessary steps in between. This article highlights the most effective Lean methods for every step of the bagging process.

The material flow in a bagging system follows a fixed sequence. Every bulk material—whether pyrogenic silica at 30 g/l or potato starch at 600 g/l—goes through the same five core steps before leaving the system as a container ready for shipment. The differences lie in the technology (pneumatic, turbine, vacuum), the speed, and the error-prone nature of each individual step.

Hänggi, Fimpel, and Siegenthaler emphasize: Lean Production optimizes each step individually AND the flow between steps. A fast packer is of little use if palletizing cannot keep up. Precise dosing is worthless if the sealing process subsequently damages every tenth bag.

Kletti and Schumacher describe the same principle from the perspective of key performance indicators: Each individual process step has its own OEE value—availability, performance, and quality. The overall efficiency of the line is the product of all these individual values. A step with 80 percent availability drags the entire line down to 80 percent—even if all other steps are at 99 percent. That is why Lean Production starts where the weakest step is, not where the largest machine stands.

The most common cause of process deviations in a bagging system isn’t the machine—it’s the variation in procedures. After a product change, Shift A cleans the equipment in a different order than Shift B. The experienced operator adjusts the dosing parameters based on intuition, while the new operator follows a table that hasn’t been updated in three years. On Mondays, the bag magazine is to the left of the nozzle; on Fridays, it’s to the right because someone parked the pallet jack in between. Each of these deviations causes errors, wasted time, or both—not as isolated incidents, but as statistical background noise that doesn’t show up in any error reports.

Lean Production addresses this variation with two tools: standardization of the work environment (5S) and technical error prevention (Poka-Yoke). Both address different aspects—5S focuses on workplace organization, Poka-Yoke on process design—but they pursue the same goal: to prevent deviations from the target in the first place.

Hänggi, Fimpel, and Siegenthaler view 5S as the first step toward real improvement—even before any process optimization. The logic: Stable processes are built on order. As long as the workplace is disorganized, deviations cannot be detected because there is no visible target state.

At the filling station, the five steps specifically mean:

https://www.greif-velox.com/en/knowledge/5s-method is therefore not a one-time cleanup effort, but a permanent operating mode. Its value is not reflected in a single metric, but in the fact that all subsequent methods—SMED, Poka-Yoke, Andon—only work in an organized workstation. Anyone who tries to optimize setup times while the operator spends three minutes looking for the right screwdriver is optimizing the wrong thing.

Poka-Yoke – literally: “preventing unfortunate mistakes” – is based on a premise that Hänggi, Fimpel, and Siegenthaler state unequivocally: People make mistakes. The method does not aim to detect and correct errors, but rather to make them constructively impossible.

At the bagging line, there are two points where Poka-Yoke has the greatest impact:

At the bag inlet, the correct valve opening determines whether the bag can be filled cleanly and subsequently sealed without defects. A bag inserted at an angle or a valve that is not fully open leads either to dust escaping during filling or to a defective weld seam. Visual bag detection—a sensor that checks the valve position before filling—prevents this error before it occurs: The filling process only starts when the position is correct. Not checking after the error, but preventing the error.

During sealing, the positioning of the valve relative to the welding sonotrode is critical. If the angle deviates, an incomplete or asymmetrical weld seam results—a bag that tears open in storage or during transport. Sensor-based position control prior to the welding process ensures that the sonotrode only activates once the valve is fixed in the correct position. This is precisely the Poka-Yoke logic described by Hänggi, Fimpel, and Siegenthaler: A form-fitting or sensor-based solution that completely prevents the error from occurring—more effective than any checklist, warning sign, or inspection instruction. The technical article on dust generation explains in detail how error prevention during bag filling reduces dust emissions, and provides further insights.

A filling plant that bags ten different bulk materials on the same line faces the same decision every day: run large batches and change setups infrequently—or run small batches and change setups frequently. The first option minimizes setup time but leads to overproduction, inventory, and tied-up capital. The second option keeps inventory low but eats into production capacity due to frequent downtime. As long as a product changeover takes 30 or 45 minutes, there is no way out of this dilemma.

SMED—Single Minute Exchange of Die—resolves this by tackling setup time itself. Hänggi, Fimpel, and Siegenthaler translate the name directly: tool change in under ten minutes. Kletti and Schumacher document that in most companies, setup time reductions of over 50 percent are achievable, provided that systematic optimization has not already been implemented.

The key lies in a distinction that seems trivial at first glance but, in practice, transforms the entire setup process: internal versus external setup operations.

Hänggi, Fimpel, and Siegenthaler illustrate this effect with a concrete example: Simply by reorganizing external setup steps into the preparation phase—without any technical changes—the company in their example gained over an hour of production time per setup operation. The operator’s working time remained the same, but the machine was idle for a shorter period.

For the bagging line, this means in practice: While the last batch of the old product is still running, the next bag type is already ready at the magazine, the new dosing parameters are stored in the system, and the cleaning tool is within reach at the station. When the line stops, only the internal part begins: bag type change, cleaning, calibration. Everything else has already been taken care of.

The impact extends beyond the individual setup process: How shorter lead times achieved through flow production at the filling line reduce overall lead time becomes apparent only at the order level. Those who can run four small batches of 500 bags each instead of one large batch of 2,000 bags, without setup time eating into profits, reduce the wait time for the next order, lower intermediate inventories, and shorten the time from order receipt to the pallet ready for shipment. Kletti and Schumacher summarize it this way: Halving setup time enables smaller batch sizes and thus shorter lead times—the relationship is not linear, but multiplicative. SMED for quick product changeovers at the bagging line will be explored in depth in a separate technical article as soon as it is published.

In lean logic, automation is neither an end in itself nor the opposite of lean production—it is a consequence of it. Hänggi, Fimpel, and Siegenthaler explicitly warn against the reflex to automate waste instead of eliminating it: Anyone who buys a robot to carry the cup from the cupboard to the coffee machine has not eliminated the waste, but merely performed it mechanically. In the context of a bagging line, this means: First make the process lean, then automate the remaining manual steps.

Viewed in this order, three levels of automation emerge, which simultaneously represent the lean maturity levels of a bagging system:

The key point is the third column: What changes with each level of automation is not just the speed, but the operator’s role. In manual operation, the human operator sets the pace. In semi-automated operation, they share the rhythm with the machine—and it is precisely at this interface that new coordination losses arise. In fully automated operation, the operator becomes a process observer who intervenes when the system deviates from the target. Bertagnolli describes precisely this shift as a prerequisite for the eighth type of waste—unused employee knowledge—to become productive: Only when the operator no longer handles every bag manually does he have the capacity to observe the process, identify deviations, and implement improvements.

Automation is thus not a substitute for Lean, but its logical continuation: First 5S, then standardization, then Poka-Yoke—and once the process is stable enough, automation as the next step. Levels of automation as indicators of Lean maturity are explored in greater depth in a separate article.

Sources

Hänggi, Roman / Fimpel, André / Siegenthaler, Roland: LEAN Production – einfach und umfassend. Ein praxisorientierter Leitfaden zu schlanken Prozessen mit Bildern erklärt.

Kletti, Jürgen / Schumacher, Jochen: Die perfekte Produktion. Manufacturing Excellence durch Short Interval Technology (SIT). 2. Auflage, Springer Vieweg, Berlin Heidelberg 2014.  

Bertagnolli, Frank: Lean Management. Einführung und Vertiefung in die japanische Management-Philosophie. Springer Gabler, Wiesbaden 2018.

Dennis, Pascal: Lean Production Simplified. 3rd Edition, CRC Press, Boca Raton 2015.