Comments from the EPO President on the patentability of computer-implemented simulations (G 1/19)

One of the more early awaiting referrals before the Enlarged Board of Appeal (EBA), is G 1/19, relating to the patentability of computer-simulated methods (IPKat post here). The referral has already attracted a large number of amicus curiae from interested parties, including CIPA, EPI and AIPPI. IPKat will review these observations shortly (once this Kat has had the chance to read them all). In the meantime, the EPO has recently announced that the EPO President himself, António Campinos, has taken the unusual move of submitting his own comments on the referral. 

The President's comments can be read in full here. It appears that the President is broadly in favour of the patentability of computer-implemented simulations (in contrast to his view on the patentability of products produced by essentially biological processes...IPKat post here). In summary, the President argues that the case law of the Boards of Appeal already provides that computer-implemented simulations, claimed as such, may be based on technical considerations. Furthermore, these technical considerations may confer inventiveness on the claim. Computer-simulated inventions may therefore be inventive, and thereby patentable. 

Case Background

Full details of the background to the referral can be found in the previous IPKat post here. In summary, the referral to the EBA was from the Technical Board of Appeal (TBA) in case T 0489/14. The appeal related to the decision by the Examining Division to refuse application EP 03793825, relating to a computer simulated method, on the grounds of lack of inventive step (Article 56 EPC).

Considering the issue of
computer implementation

The application in question describes a mathematical model for simulating the movement of individual pedestrians through an environment. The simulation can be used to aid the design of busy environments such as railway stations. The Board of Appeal considered the claimed simulation method to lack technical features, other than the fact that it was implemented on a computer (hence "computer-simulated invention"). Given that computers are known, the method was therefore found to lack inventive step. Furthermore, given that the assessment of inventive step should be based only on the technical features of the claim, the claim was rejected for lacking inventive step over a computer (Article 56 EPC). 

However, the Board of Appeal accepted that their opinion was at odds with other Board of Appeal decisions, namely T 1227/05 and T 1227/05. The Board of Appeal also believed that the question to be determined, namely the interpretation of Article 56 EPC could not, in fact, be answered directly and unambiguously by reference to the EPC. The Board of Appeal therefore referred the question of how the inventive step of computer-simulated inventions should be assessed to the EBA. 

Comments from the President

Taking the President's answer to each of the three referred questions in turn: 

Question 1. In the assessment of inventive step, can the computer-implemented simulation of a technical system or process solve a technical problem by producing a technical effect which goes beyond the simulation's implementation on a computer, if the computer-implemented simulation is claimed as such?

The first referred question asks whether a claim that relates only to a method that is implemented on a computer can ever be said to solve a technical problem beyond the simple matter of implementing the method on the computer. Question 1 arises in cases of computer-implemented simulations because the claimed invention is only considered to relate to technical subject-matter in so far as the claimed method is implemented on a computer (the computer being the "technical subject matter"). The inclusion of the technical means of a computer shifts the question to one of whether the method implemented on a computer is inventive (Article 56 EPC). 

In proposing his answer to this question, the President first referred to previous Boards of Appeal decision on claims including non-technical and technical features. According to T 641/00 and T 154/04,  non-technical features that do not interact with the technical features of a claim in order to solve a technical problem should be ignored for the purpose of assessing inventive step (T 641/00 and T 154/04). According to the same principle, "non-technical features which do interact with the technical subject-matter of the claim for solving a technical problem must be taken into account" (para. 20).

Applying the problem solution approach, the President considered the closest prior art to the claim as being a general purpose computer. In order to formulate the technical problem to be solved, the President noted that, when a non-technical method is carried out for a technical purpose and thereby involves "technical considerations", these technical considerations can only originate from a person skilled in a technical field"(para. 24). The objective technical problem to be solved in such a case is therefore not simple be “how to implement the non-technical method steps on a computer?”. Such a formulation, the President's view, would involve an ex facto analysis that would only be possible with hindsight knowledge of the technical considerations.

Running a simulation on a computer in order to determine a technical parameter, the President argued, is also not equivalent to a mental act. The simulation is, on the contrary, "an experiment relying on technical principles: replicating on a computer the operation of the technical system or process under certain conditions implied by the model and its parameters". The simulation also relies "on technical considerations pertaining to the technical field relating to the simulated technical system or process" (and not merely to considerations of how to implement the simulation on a computer). Therefore, in the President's view "the provision of the simulation results represents a technical effect going beyond the computer-implementation at least to the extent that it is based on technical principles reflected in the simulation steps" (para. 32).

The President also noted, according to EPO case law, that a "technical effect" does not require the presence of a direct link to physical reality. A technical effect may be found, for example, in the processing of computer-generated image (T 208/84). A model designed by a computer-implemented simulation, the President argued, may also be considered technical e.g. because the model relies on real-world measurements "attributing technical character on the simulation using this model".

The President's answer to the first question was therefore "yes". In the President's view, a computer-implemented simulation of a system of process claimed as such can solve a technical problem. In particular, the computer-implemented simulation can produce a technical effect (which goes beyond the simulation’s implementation on a computer) when the claimed simulation method reflects technical principles underlying the simulated system or process. 

Question 2. If the answer to the first question is yes, what are the relevant criteria for assessing whether a computer-implemented simulation claimed as such solves a technical problem? In particular, is it a sufficient condition that the simulation is based, at least in part, on technical principles underlying the simulated system or process?

The second question asked how one may assess whether a computer-implemented simulation solves a technical problem? Is it sufficient that the simulation is based on technical principles? The President answered this question fairly succinctly, with reference to T 1227/05. According to the President, it is sufficient for a claimed computer-implemented simulation of a technical system or process to solve a technical problem, if the simulation method reflects technical principles underlying the simulated system or process (para. 45). 

Question 3. What are the answers to the first and second questions if the computer-implemented simulation is claimed as part of a design process, in particular for verifying a design?

The third question was dealt with even more succinctly. The President did not believe the third question required an answer, in view of his answer to the first and second questions. In particular, "if a simulation reflects, at least in part, technical principles underlying the simulated system or process, it solves a technical problem beyond the simulation’s implementation on a computer, irrespective of whether it is claimed as part of a design process".

Will the EBA agree with the President? The EBA is independent of the President and is therefore not obliged to follow his opinion. Furthermore, as mentioned above, there have already been a large number of observations from third parties submitted to the EBA, some in favour and some against the patentability of computer-simulated inventions (Article 10 RPEBA). Individuals with strong views on this issue [Merpel: such as certain hyperbolic bloggers...], are encouraged to submit their own!