MES Software: Vendors, Features & Costs Compared 2026
MES software compared: vendors, functions per VDI 5600, costs (cloud vs. on-premise) and implementation. Honest market overview 2026.
By Christian Fieg · Last updated: April 2026
What is Poka Yoke?
Poka Yoke (Japanese: ポカヨケ) is a method for designing production processes so that human errors are either physically impossible or immediately detected before they produce a defect. The term translates roughly as "mistake-proofing" — poka (inadvertent mistake) + yokeru (to avoid). It was developed by Shigeo Shingo, a Japanese industrial engineer who worked as an external consultant to Toyota from the 1960s onward. Shingo's core insight: quality inspection after production is too late — by the time you find the defect, you have already produced it.
Poka Yoke moves the quality check to the point of creation: the process itself prevents the error, or detects it within the same cycle. A USB-C connector that fits in only one orientation is Poka Yoke. A fixture that physically prevents a part from being loaded upside down is Poka Yoke. An MES that stops a machine when a process parameter exceeds the tolerance window is digital Poka Yoke. The principle is the same: make it impossible or immediately obvious to do the wrong thing. In Shingo's framework, Poka Yoke is not a standalone method — it is the implementation mechanism for what he called source inspection: checking the conditions that cause errors, not the results of errors. Combined with Jidoka (built-in quality / autonomation) and the broader Lean quality system, Poka Yoke is what makes zero-defect production a practical goal rather than a theoretical aspiration.
What are the 3 Poka Yoke methods?
Shingo classified Poka Yoke devices into three methods based on how they detect the error condition. Each method answers a different question about the process:
| Method | What it checks | Physical example | Digital MES equivalent | Error type prevented |
|---|---|---|---|---|
| Contact method | Physical shape, size, or feature of the workpiece. The device detects whether the part is present, correctly oriented, or the right variant — through physical contact or proximity. | A fixture with asymmetric guide pins: the part only fits if oriented correctly. If loaded upside down, the pins block it. The operator cannot start the cycle with a misloaded part. | MES reads a barcode or RFID tag on the workpiece before the cycle starts. If the scanned part number does not match the active production order, the MES blocks the machine start signal. At Carcoustics, digital changeover (Rüstprozesse) validation is a form of digital contact-method Poka Yoke. | Wrong part, wrong orientation, wrong variant, mixed parts. |
| Fixed-value method | Whether a fixed number of actions have been completed. Counts operations, parts, or movements and compares against the required number. | A parts bin with 8 compartments, each holding one screw. If one compartment is still full after the assembly step, a screw was missed. The operator cannot close the assembly with a screw remaining in the bin. | MES counts the number of press strokes, weld spots, or torque confirmations per assembly. If the count does not match the required number for the product, the MES flags the unit as incomplete and prevents it from advancing to the next station. | Missing operations, forgotten components, skipped steps. |
| Motion-step method | Whether the correct sequence of motions or steps has been followed in the correct order. Detects if a step was skipped or performed out of sequence. | A machine interlock that requires the operator to press button A before button B becomes active. Pressing B first is physically impossible — the circuit is not complete until A has been confirmed. | MES enforces the production routing: station 3 cannot start processing the unit until station 2 has confirmed completion. If the unit skips station 2 (e.g., a manual handling error), the MES blocks station 3. The SYMESTIC production control module enforces these routing sequences digitally. | Wrong sequence, skipped processes, premature advancement. |
What is the difference between prevention and detection Poka Yoke?
Shingo distinguished two levels of Poka Yoke effectiveness — both are valuable, but they are not equal:
| Level | What it does | When the defect is caught | Example | Cost of the defect |
|---|---|---|---|---|
| Prevention (Control) | Makes the error physically or logically impossible. The process cannot proceed with the error condition present. | Before the defect is created. The error condition is blocked at the source. | A fixture that physically prevents misloading. An MES that blocks the machine start when the wrong program is selected. | Zero. No defective unit is produced. |
| Detection (Warning) | Detects that an error has occurred and alerts the operator or stops the process — but the error has already happened. The goal is to catch it before it moves to the next process. | Within the same cycle or station. The defect exists but has not left the station. | A vision system that inspects the part after pressing and rejects it into a NOK bin. An MES alarm that fires when press force exceeds the upper limit — the part is pressed, but the operator is alerted immediately. | Low. One defective unit is produced but caught before it reaches the customer or the next assembly step. |
Prevention Poka Yoke is always preferred over detection Poka Yoke — because zero defects is better than immediate detection of defects. But detection Poka Yoke is vastly better than end-of-line inspection, where defective units travel through the entire production process before being caught. In practice, most production processes use a combination of both levels: prevention where the error condition can be physically or digitally blocked, detection where it cannot.
What are 6 practical Poka Yoke examples in discrete manufacturing?
| # | Error | Poka Yoke type | Physical solution | Digital MES solution | Why it matters |
|---|---|---|---|---|---|
| 1 | Part loaded upside down in press | Contact / Prevention | Asymmetric guide pins on the fixture. Part only fits in correct orientation. | Vision system confirms orientation before cycle start; MES blocks cycle if NOK. | An upside-down part in a 200-ton press destroys the tool (€10,000+) and creates unplanned downtime (hours). |
| 2 | Wrong material batch used | Contact / Prevention | Colour-coded bins — red for batch A, blue for batch B. Operator picks from the colour matching the current order card. | MES scans material batch barcode at the machine. If the batch does not match the active order's BOM, the machine is locked. Full traceability per serial number. | In automotive, a wrong material batch triggers a recall. At Meleghy, bidirectional SAP integration ensures order-to-machine-to-material traceability. |
| 3 | Screw missing in assembly | Fixed-value / Detection | Torque screwdriver with counter: 6 screws required, machine counts 6 "torque OK" signals. If only 5, the station signals NOK. | MES receives the torque counter signal from the PLC. If count ≠ 6, the unit is flagged as incomplete in the MES and cannot be booked as good output. | A missing screw in a safety-relevant assembly (e.g., brake caliper) is a potential field failure. Fixed-value Poka Yoke makes it mathematically impossible to miss. |
| 4 | Process step skipped | Motion-step / Prevention | Physical interlock: station 3's clamp only releases if station 2's "process complete" signal is active. | MES enforces routing sequence digitally. If a unit's barcode is scanned at station 3 but station 2 has no "completed" record for that serial number, station 3 does not start. | Skipped processes create latent defects — the product looks fine but has not been heat-treated, tested, or coated. These are the most expensive defects because they are invisible. |
| 5 | Press force too high (quality drift) | Contact / Detection | Mechanical overload protection: relief valve opens above 3,200 N, diverting the part. | MES process data module monitors press force per cycle. If force exceeds 3,000 N (soft limit), the MES triggers an alarm and flags the unit. At Neoperl, SPS alarm correlation identified exactly these edge-of-tolerance conditions. | Process parameter drift produces units that pass the end-of-line test but fail in the field — the classic NTF (No Trouble Found) problem. |
| 6 | Wrong machine program for product | Motion-step / Prevention | Operator selects program from a physical key lock — each product has a unique key. | MES sends the correct program number to the machine automatically when the order is dispatched. The operator does not select the program — the production control module does. At Carcoustics, digital changeover support ensures the machine runs the right program for the right order. | Running the wrong program produces scrap from the first cycle — and the operator may not notice until the end-of-line test rejects the parts, wasting an entire batch. |
How does an MES enable digital Poka Yoke?
Traditional Poka Yoke relies on mechanical fixtures, physical interlocks, and hardware sensors. Digital Poka Yoke — implemented through an MES — extends the same principle into the data layer. The error is still prevented or detected at the source, but the mechanism is software rather than hardware:
| MES capability | Poka Yoke function | Errors it prevents |
|---|---|---|
| Order-to-machine validation | MES verifies that the correct order, BOM, and machine program are active before the cycle starts. If the setup does not match the order, the machine does not start. | Wrong product on wrong machine, wrong program, wrong tool, wrong material. At Meleghy, the bidirectional SAP integration maps every machine cycle to the correct Fertigungsauftrag. |
| Routing enforcement | MES tracks each unit through the production routing and blocks processing at any station if the previous station has not confirmed completion. | Skipped processes, out-of-sequence operations, units bypassing quality gates. |
| Process parameter monitoring | MES monitors process parameters (force, temperature, time, speed) per cycle. If a parameter exceeds the defined tolerance, the MES flags the unit and/or stops the machine. | Quality drift, out-of-spec production, intermittent process deviations. At Neoperl, this capability reduced scrap by 15 %. |
| Alarm-based machine stop | MES alarms module detects PLC alarm patterns that precede defects and triggers alerts or machine stops before defective units are produced. | Repeated defects from degrading equipment. This is Poka Yoke applied to the machine condition, not just the product. |
The critical difference: physical Poka Yoke protects against one specific error at one specific station. Digital Poka Yoke through an MES protects across the entire production routing — from raw material intake to finished goods. A physical fixture cannot check whether the previous station completed its process. The MES can.
FAQ
Who invented Poka Yoke?
Shigeo Shingo (1909–1990), a Japanese industrial engineer. Shingo was not a Toyota employee — he was an external consultant who worked with Toyota from the early 1960s. He originally called the concept "Baka-Yoke" (fool-proofing), but changed the name to "Poka-Yoke" (mistake-proofing) after a worker at Arakawa Body Company objected to the term "baka" (fool) as disrespectful. Shingo documented the method in his book Zero Quality Control: Source Inspection and the Poka-Yoke System (1986, English translation). The method is part of the broader Toyota Production System (TPS) but was specifically developed and systematised by Shingo, not by Taiichi Ohno (who developed the flow and pull elements of TPS).
What is the difference between Poka Yoke and Jidoka?
They are related but distinct concepts within Lean manufacturing. Jidoka is the broader principle: "build quality into the process" — give machines and operators the ability and authority to stop production when an abnormality is detected. Poka Yoke is a specific implementation technique: design the process so that the error cannot physically or logically occur, or is detected immediately. Every Poka Yoke device is an expression of the Jidoka principle. But Jidoka is wider — it also includes Andon systems (visual management), in-station quality checks by operators, and the organisational principle that stopping the line for quality is expected, not punished.
Can Poka Yoke eliminate all defects?
No — but it can eliminate entire categories of defects. Poka Yoke is most effective against errors caused by human variability: misloading, wrong selection, forgotten steps, sequence errors. It is less effective against defects caused by material variation, tool wear, or random process variation — those require statistical process control (SPC) and process parameter monitoring, which is where the MES process data module takes over. The practical approach: use Poka Yoke for the top 5 error modes identified by your quality Pareto, and SPC/MES monitoring for the process-variation-driven defects.
How does Poka Yoke relate to quality control and ISO 9001?
ISO 9001:2015 requires organisations to "determine actions to prevent or reduce undesired effects" (clause 6.1) and to "implement planned arrangements to verify that requirements for products and services have been met" (clause 8.6). Poka Yoke is one of the most effective methods to satisfy both requirements — it prevents undesired effects at the source and verifies requirements within the process, not after it. In an ISO 9001 audit, Poka Yoke devices and digital MES-enforced routing constraints are evidence that the organisation has embedded quality into the process, not bolted it on at the end.
Related: Jidoka · Quality Control · Kaizen · Lean Management · Six Sigma · NTF Rate · OEE Explained · SYMESTIC Production Control · SYMESTIC Process Data · SYMESTIC Alarms Module · MES: Definition & Functions
About the author
Christian Fieg
Head of Sales at SYMESTIC. Six Sigma Black Belt. Has implemented Poka Yoke devices on four continents at Johnson Controls — from mechanical guide pins on headliner fixtures to digital MES-enforced routing constraints across 900+ machines. The ones that worked best were always the simplest. Author of OEE: Eine Zahl, viele Lügen. · LinkedIn
Start working with SYMESTIC today to boost your productivity, efficiency, and quality!
