IP Insights: Patentability of computer-implemented simulations

Show notes

Visit our European Software Patents Knowledge Base for more of EPO's case law.

Can a simulation as such be patented at the EPO?

Yes, it can. However, only under specific circumstances. According to G1/19, if the output of a simulation is intrinsically technical, the simulation can be patented. A simulation can be patented if it is adapted to the internal functioning of the computer on which it is executed. Other criteria, as for example mentioned in the amicus briefs, were found to be not sufficient.

In particular, it is not sufficient that the simulated system is technical. For example, in the decision "Circuit Simulation" of 2006, an algorithm for producing pink noise is applied to an electrical circuit that is affected by the noise. In that decision, the Board of Appeal found that the simulation as such is sufficiently technical. That decision, according to the new G-decision is no good law any longer.

It would not be sufficient that the simulation is simply better. For example, you come up with a new model for simulating weather conditions allows you to reliably predict the weather not only for five, but for ten days. That would not put you in a position to claim the simulation as such. What would not help either is to argue that your novel computer simulation avoids physical simulation as practiced in the prior art. For example, if you can now do without an actual model of a car that is tested in the wind channel, that might save you millions, but it wouldn’t be good enough to get a claim directed to this simulation as such at the European Patent Office .

It is not sufficient to argue that, for coming up with the novel computer-implemented simulation, technical considerations were necessary to better understand the simulated system.

03:33 If the EPO won’t give you a claim directed at the simulation as such, you as a patent attorney, what do you tell your clients?

A first option is to mention steps in the claim wording that further define how the output the simulation is used.

Think of a simulation that aims to improve a product like a rollover bar with improved stability or a steel beam with improved load capacities. In these cases, the steps of manufacturing the products have to be mentioned in the claim wording.

You may also limit the claim to the technical use of the output. Let’s assume we have a weather forecast simulation: You may claim, for example, a cat flap that is controlled by the output of such a simulation which allows a cat to leave an apartment only if rain is unlikely in the next few hours.

04:39 However, the question arises whether the presentation of the output of the simulation on a display is already sufficiently technical. What do you think, Christof?

Well, it’s unlikely that the simple presentation of the outcome of the simulation on a screen is enough to solve a technical problem. will most likely not be sufficient. Using the example of a weather forecast, if you present a weather forecast in your weather app, that is probably just considered to be presentation of information, and therefore not technical, by the EPO.

The case gets a little bit trickier if you feed this data in real time into a rain radar app which produces a film that presents to the user in a quite detailed manner how clouds and rain will move on a map in the next 48 hours.

Another example, actually addressed by the G-decision, is the simulation of a movement of a billiard ball which is then used in a computer game, presumably for presenting the movement on a screen. The G-decision seems to have doubts whether such a use produces a technical effect. But I would argue that, in many cases, we get patents on how to more realistically render a 3D object on a screen. If that is technical, why not the use of a simulation to more realistically present the movement of how the ball bounces back from the bank?

Are there any insights that the decision provides on the patentability of computer-implemented inventions in general?

Your claimed subject-matter always has to solve a technical problem.

This means that the COMVIK approach has to be used for all kinds of computer-implemented inventions for assessing inventive step. Therefore, only those features that contribute to the solution of a technical problem can be considered for this assessment. So, computer-implemented simulations do not have a special status.

Yes, but surprisingly, non-technical features can contribute to the solution of the technical problem.

The case underlying this G-decision was about the optimization of a building by simulating how pedestrians move through the building. The G-decision makes it quite clear that a simulation can contribute to the solution of a technical problem even if it reflects such non-technical aspects as human behavior.

This finding could provide applicants with ammunition for arguing that their non-technical features should be considered in many cases beyond simulations.