Criminal and civil liability of the management board for accidents involving machines without the CE mark

An accident involving a machine without a CE mark rarely begins with the missing marking itself. It is usually preceded by a chain of decisions in which safety is deferred: a purchase made without a clear allocation of responsibility, a modification carried out without assessing its consequences, the integration of equipment treated as a minor adjustment, or commissioning proceeding despite unresolved issues with guards, interlocks, and access to hazard zones. Only when an injury occurs do those decisions cease to be a technical matter and become a question of management liability.

From a practical perspective, the issue is not simply whether the machine should bear a CE mark. What matters is whether the organization can show who was responsible for the conformity of the specific configuration, on what basis it was released for operation, and whether the hazards were actually identified and reduced. If that cannot be established clearly today, the risk no longer concerns only downtime, redesign, and a dispute with the contractor, but also the criminal and civil liability of those who approved operation or tolerated it despite reservations.

Why this matters today

Management liability for accidents involving machines without a CE mark does not arise only after the event. It stems from earlier decisions on purchasing, relocation, modification, equipment integration, and starting production before the documentation has been completed. If an accident occurs in that context, the assessment covers not only the technical condition of the machine, but also how those in management acted: whether the risk was identified, whether responsibility was assigned, whether the release for operation had a real basis, and whether obvious deficiencies were knowingly tolerated.

In practice, the absence of a CE mark is often primarily a sign of a deeper problem. Most often, it means the machine’s legal status has not been established, safety has not been verified after changes, or it is unclear who is responsible for the solution as a whole after several devices have been integrated. That is why this is a management issue: it is not about the label itself, but about whether the organization is in control of the consequences of its own engineering and organizational decisions in the area of machine safety.

The biggest costs usually arise when the sequence of actions is wrong: installation and commissioning first, responsibility for conformity sorted out later; deadline pressure first, analysis of guards, safety-related control systems, and instructions later; pilot production first, and only after an incident is the documentation brought into order. In that model, each subsequent decision narrows the room to maneuver. Then come downtime, control system redesign, a dispute with the supplier or integrator, and, in the event of an accident, claims from the injured party and questions about the liability of those who allowed the machine to operate.

A typical flashpoint arises where a plant assumes it is dealing only with a single machine or only with a modification. In reality, adding a robot, feeder, conveyor, or additional control system often changes the function of the workstation, the operator’s way of working, and the scope of the protective measures. If such a change creates a new functional whole, the question is no longer whether it is worth adding the missing marking, but who is responsible for preparing the system for CE marking and whether a sound risk assessment and verification of the safeguards were carried out before start-up.

Only against that background does the reference to legal requirements make sense. In the event of an accident, what matters is not only whether the machine should have been CE marked, but also whether the responsible party exercised due care when putting it into use, rebuilding it, or operating it. Criminal and civil liability is therefore not a separate “post-accident” issue, but a consequence of earlier technical and organizational decisions.

Where cost or risk most often increases

The biggest increase in cost and risk does not arise at the moment of the accident itself, but when the absence of CE marking is treated as a purely formal issue. That is a management error with technical and legal consequences. If a machine is put into use without clarifying its compliance status, without defining responsibility for modification, integration, or commissioning, every subsequent stage of the project becomes more expensive. The number of reworks increases, acceptance takes longer, and after an incident it becomes harder to demonstrate due diligence.

A second area of risk is postponing key decisions until commissioning. That is when it turns out that guards interfere with the process, the control system does not provide safe stopping to the required extent, service access forces people to bypass protective measures, and the documentation describes a different machine from the one actually installed on the shop floor. These are not minor deviations, but typical sources of delays, additional costs, and liability. Any such discrepancy may require a new risk assessment, design changes, replacement of components, and re-verification of safety functions.

An underestimated source of risk is also the operating manual. If it is created at the very end merely “for the file,” it protects neither the user nor the people responsible for authorizing the machine for use. After an accident, the question will not be whether the document exists, but whether it reflects the actual hazards, intended uses, changeover activities, cleaning, and fault-clearing tasks. If the operator has to improvise because the machine requires actions not covered by the documentation, or the documentation omits operating limitations, both the likelihood of harm and the difficulty of defending the company’s position increase.

The most costly situations, however, are those in which no reliable risk assessment was carried out, or it was treated as a description of decisions already made. If the analysis does not cover the actual method of use, the changes introduced during integration, and maintenance activities, management makes investment decisions based on an incomplete picture of the hazards. Later, every failure, line modification, or workplace incident leads to the question of whether the risk was identifiable earlier and whether it could have been prevented.

This is also where the regulatory dimension comes into play. After an accident, actions by supervisory authorities and market surveillance authorities may matter, ranging from consequences related to inspections of working conditions to withdrawal of the product from the market or suspension of its availability. From the plant’s perspective, a production stoppage usually proves far more costly than resolving the machine’s status and the conditions for its safe use in advance.

How to approach this in practice

A practical approach should begin with a simple separation of three issues that are wrongly mixed together in many projects: permissibility of use, the actual level of safety, and the formal path to bringing the machine into compliance. For management, the key issue is not the abstract question of whether a machine “should have CE,” but whether the organization can demonstrate on what basis it considered the machine acceptable for use and who approved that decision despite the known limitations.

A supplier’s declaration is not enough here, nor is the argument that the machine previously operated elsewhere. A more useful criterion is practical: can the machine boundaries, safety functions, foreseeable operator and maintenance tasks, and the areas where a person comes into contact with a source of hazard be clearly identified? If the answer is incomplete, the problem stops being purely formal. In that case, the first step is to verify whether the equipment is safe in its actual configuration, and only then to determine the full compliance path.

A good example is a line consisting of a used machine and a newly added feeder. On the surface, this looks like a quick start-up. In reality, the method of material feeding changes, the operator’s work rhythm changes, access to working zones changes, and the emergency stop method changes. If, in such a setup, an accident occurs during manual clearing of a jam, the assessment will not stop at the question of a missing CE mark. The key issue will be whether, after the devices were connected, anyone checked the changed hazardous zones, the operation of guards and interlocks, and whether the new configuration now requires preparation for CE marking as a complete assembly.

From a management perspective, this means moving from “start it up and fix it as we go” to a documented decision-making process. Before authorizing operation, you should be able to answer at least three questions:

• whether the entity responsible for the conformity of the specific machine configuration is known,
• whether, after integration or modification, the boundaries of hazardous zones and the operation of protective measures can be justified,
• whether there is a document explaining why the equipment can be started now rather than only after modification.

If the answer to any of these questions is uncertain, the issue must not be treated as a minor technical detail to be resolved later. In that case, liability risk increases not because one marking is missing, but because the organization is knowingly operating without deciding whether it is dealing with a safe machine or simply a machine that has been started despite unresolved deficiencies.

Only at the end is a normative reference needed. If the machine is being placed on the market, put into service, or substantially modified, the question becomes the full compliance path and preparation for CE marking. If, however, the formal status is unclear but the equipment is already on site, the first obligation is to show that operation was not allowed without first identifying the hazards.

What to watch for during implementation

At the implementation stage, the most serious mistake is to treat the absence of the CE mark as a documentation issue that can be closed out in parallel with commissioning. In accident cases, this is often the point that is judged most harshly later. The issue is not whether the full document set was ready on a particular day, but whether management allowed the machine to operate despite an unresolved safety status. If implementation proceeds while the device status is unclear, every scheduling decision may later take on evidential significance.

In practice, commissioning must not be combined with temporary acceptance of open issues concerning guards, interlocks, service access, and work organization around moving parts. If, during implementation, it is not possible to state clearly where the operator’s permissible working area ends and where the area requiring separation begins, the risk is no longer abstract. Discovering poorly defined danger zones late usually means mechanical redesign, changes to the control system, and repeated testing, which means greater cost and delay than stopping the implementation earlier.

There is one useful assessment criterion: after start-up, can the machine’s normal, foreseeable, and service use be defended without relying on staff improvisation? If safety depends on the operator “knowing where not to go,” maintenance “just disabling one sensor for a moment,” and the integrator “finishing the guard after acceptance,” implementation enters an area of heightened liability. A typical case is a line that works from a process standpoint, but where clearing jams requires entry into the motion area without a stably designed intervention mode. In such a setup, an accident will be assessed not as bad luck in operation, but as the result of allowing the machine to run without resolving known hazards.

The second trap is drawing the line between commissioning itself and preparing the machine for CE marking. Not every correction at the implementation stage changes the device’s legal status, but the opposite assumption is just as dangerous: that because the machine is already on the shop floor, conformity analysis can be postponed. When several devices are integrated, control logic is changed, safeguards are bypassed for start-up, or a rebuild is carried out to remove production problems, the matter quickly shifts from operation to responsibility for a new technical configuration. Then the cost is not limited to documentation, but also includes the need to reassess the technical solutions, including verification of control systems in line with the principles described in EN ISO 13849-1.

Finally, the most important conclusion from management’s perspective remains the same. CE marking does not replace hazard assessment, but its absence on a machine that should have gone through the proper procedure is a clear signal that the organization must be able to demonstrate the basis on which the device was allowed to be used. In practice, it is worth requiring not a general statement that “the machine is almost ready,” but decision material showing three things: who is responsible for the conformity of the specific configuration, what hazards remain unresolved, and what condition must be met for safe start-up or for work to be stopped.

This way of proceeding does not eliminate the risk of a dispute entirely, but it limits the most dangerous situation: an implementation in which criminal and civil liability increases because the project moved faster than the organization was able to determine whether the machine was actually fit for safe operation.