A recent decision from the EPO Board of Appeals, T 66/12, deals with, amongst other things, numerical ranges within a claim.

The present decision relates to an appeal filed against the interlocutory decision of the Opposition Division, posted on 27 December 2011, to maintain European patent EP1340276 in an amended form. The parties involved during the appeal were the patent proprietor, Toyota Jidosha, and the appellant, Linde Material Handling.

Claim 1 of EP1340274 relates to a direct current power source, comprising, in part, a fuel cell and an electric power storage device, and where a ratio of a maximum output of the fuel cell to a maximum total output of the fuel cell and the electric power storage device is in a range of equal to or greater than 0.65 and equal to or less than 0.8.

The Opposition Division had found that the range of the ratio of equal to or greater than 0.65 and equal to or less than 0.8 was novel and inventive over the prior art documents, E1 and E2, a point contested by the Appellant.

It is established European case law (see T 198/84 and T 279/89) that for a sub range of the prior art to be novel, it must satisfy the following three criteria:

  1. the selected sub-range is narrow compared to the known range;
  2. the selected sub-range is sufficiently far removed from any specific examples disclosed in the prior art and from the end-points of the known range;
  3. the selected range is not an arbitrary specimen of the prior art, i.e. not a mere embodiment of the prior art, but another invention (purposive selection, new technical teaching).

In the appeal, the appellant argued that the selection of the range specified in claim 1 amounted to an arbitrary selection, and as such the third criterion was not met. Additionally, the appellant argued that there was no known normal range, and as such, the second criterion could not be met, since before deciding if a range was sufficiently removed from a known range, a normal range should be determined.

In response to the summons to oral proceedings relating to the appeal, the patent proprietor provided late filed supplementary evidence in support of the claimed range. The supplementary evidence was test results showing plots of fuel cell to battery output ratio against efficiency obtained from a variety of vehicles. The proprietor argued that the test results showed that greater efficiency was obtained when the ratio was within the claimed range of 0.65 to 0.8, and as such, did not result from an arbitrary selection.

In its decision, the Board argued that, while the supplementary evidence shows that the fuel efficiency is high for a certain range of fuel cell to battery output ratio, the weight of the vehicles specified in the supplementary evidence varied with the ratio. As such, the Board argued that there is doubt as to whether the values reported in the supplementary evidence relate to vehicles of the same type and weight. Additionally, the Board noted that, in particular, the supplementary evidence did not clearly show that the end points of the range (0.65 and 0.8) were significant, since the plots showed that the fuel efficiency varied little over a much broader range of output ratio. The Board further argued that the results do not show that this range would apply generally, for instance when using different test cycles. Further still, the Board argued that the test results seemed to relate to current vehicles, and so it could not be excluded that the current vehicles make use of technologies which were not available at the priority date of the application underlying the patent.

Given the late filing of the test results, as well as the lack of sufficient information derivable from the test results, the Board did not admit the supplementary evidence into proceedings.

The question then turned to novelty of the claimed range. The proprietor had argued that the claimed range of 0.65 to 0.80 was determined with the aim of improving fuel efficiency. Additionally, the proprietor had argued that "The lower limit of 0.65 for the output ratio of the present invention is set in accordance with passage [0037] of the patent specification in order to fulfil the requirement of outputting the required electric power for continuous cruising by the fuel cell".

The Board referring to paragraph [0037], noted that it is described that a vehicle weighing 2000kg with a motor efficiency of 80 per cent cruising at 120km/h on a gradient of 4.5 per cent requires approximately 65kW, whereby if the maximum required electric power is assumed to be 100kW, the lower limit of the ratio should be 0.65. Therefore, the Board, agreeing with the appellant, argued that the lower limit of the claimed range is influenced by both the driving conditions and the vehicle type, and that paragraph [0037] can also be interpreted as defining a power minimum for a fuel cell independently of the fuel efficiency, the battery or a proportion of fuel cell power to battery power. Additionally, the Board, referring to paragraph [0038] of the patent specification, raised similar comments regarding the upper range of the ratio of 0.8.

Furthermore, the Board noted that paragraph [0041] of the patent described that the fuel efficiency becomes significantly higher in an output range of 40 to 80 per cent, indicating that the value 0.65 is based on different assumptions and requirements. Given the above comments regarding the disclosure of the specification, as well as the fact that claim 1 is not limited to any particular vehicle type or particular driving conditions, the Board argued that the proprietor has not demonstrated that fuel efficiency was improved with the claimed ratio in any circumstances and that no purpose behind the selection could be seen. As such, the Board ruled that the third criterion relating to a range being an arbitrary selection was not met and that the subject matter of claim 1 lacked novelty over E1.

We can learn from this decision that it is important to clearly define in an application why a claimed numerical range is significant and not an arbitrary selection. This is particularly important to keep in mind in cases where a numerical range is a secondary consideration and is not initially claimed, since should cited prior art force the applicant to claim the range, a good description may be required in order to help provide novelty for that range. Additionally, we can learn that if evidence is to be submitted in order to back up a claim that a particular range leads to an advantage, particularly if it is late filed, the technology used in creating that evidence should ideally be technology that was available at the priority date of the application.