EP 3 756 153 A1 has been published as WO 2019/164758, and relates to a pump chemical compatibility management system.
To prevent damage to the pump, an operator scans an RFID tag attached to a reservoir containing the chemical and an RFID tag attached to the selected pump. Based on the scanned data, a chemical identifier and a pump identifier, a data server determines whether the chemical composition is compatible with the pump. The result is then displayed to the operator.
Brief outline of the procedure
The EPO established the ISR and the IPER which was negative for IS in view of the PA quoted in the ISR. D5=US 2010/0024915 was introduced by the ED in the regional procedure before the EPO.
The ED refused the application as lacking IS over D5.
The applicant appealed.
In appeal, the board cited D6=A BRIEF GUIDE TO CHOOSE THE SUITABLE PUMP MATERIALS FOR CHEMICAL APPLICATIONS; 24 October 2013; available at: https://www.gemmecotti.com/suitable-chemical-pump-materials
The board held that D5 was not a suitable starting point for assessing IS, and remitted the case to the ED for a further search.
The applicant’s point of view
D5 did not address the problem underlying the invention, namely checking the compatibility of a chemical composition with each component of a chemical pump.
The applicant argued that D5 was concerned with a different purpose than the claimed invention, namely that of ensuring that the correct chemical was dispensed in the right container, which was a labelling issue rather than a compatibility issue.
Nothing in D5 suggested analysing the chemical compatibility between the material of a pump or container and the chemical composition being in contact with such a material. Thus, the ED erred in its conclusion that D5 disclosed a chemical compatibility check between the chemical solution and the container.
The applicant furthermore argued that, starting from D5, the skilled person would not have recognised the problem of chemical damage to the pump, and would therefore not have sought a solution to such a problem.
The applicant contested that this problem was well known and argued that it had not been established that D6 was published before the priority date.
Although the pump and the receiving container in D5 would eventually show signs of wear, this would not necessarily be recognised as being caused by the diluted chemical. In such a case the skilled person would simply replace the damaged item and there would be no need to check chemical compatibility between the components and the chemical compositions coming into contact with them.
The board’s decision
The board agreed with the applicant that D5 does not disclose a compatibility check between a pump or container and the chemical composition that comes into contact with this component.
In D5, the check ensures that the correct chemical is dispensed into the correct container so as to avoid a mislabelling.
The board went further and considers that D5 did not disclose a pump at all.
The chemical dispenser in D5 comprises an “eductor for drawing chemical from that source into a diluent, such as water, for dilution and discharge into a receiving container”. Although an eductor moves fluids in a manner similar to a pump, it operates according to a different principle. In particular, unlike a pump, it has no mechanical components driven by an electric motor or the like, but is instead fluid-driven.
Accordingly, in order to arrive at the claimed subject-matter starting from D5, the skilled person would first have to add a chemical pump, then address the issue of chemical compatibility, and finally apply the automated compatibility check proposed in D5 for this purpose.
In the board’s view, even if each of these steps were known or individually trivial, a motivation would still be required for the skilled person to modify the prior art in the claimed manner. The board cannot identify such a motivation when starting from D5 as the CPA. The board thus held that D5 was not an appropriate starting point for the assessment of IS.
While it is not mandatory to start from the strictly closest prior art when arguing a lack of inventive step, see CLBA, 11th edition, I.D.3.3, it is nevertheless recommended to start from prior art having a similar purpose and overall technical effect.
Starting from a more remote document often leads to an ex post facto analysis, in particular to the formulation of a “hindsight problem” – a problem that the skilled person would not have realistically conceived.
In the present case, D5 neither pursues the same purpose nor discloses the structurally most significant component – the pump – which is a prerequisite for the technical problem addressed by the invention to arise at all.
In the board’s view, a more suitable starting point would be a system comprising a chemical pump and addressing issues of chemical compatibility with the components of that pump. Such a starting point would require fewer structural modifications and would allow for the formulation of a more realistic technical problem, such as the automation of the compatibility check.
D6 deals with chemical compatibility checks for pump components, but the board agreed that D6’s publication date was uncertain and therefore remitted the case to the ED for a further search.
The board further noted that the IPC class “G06Q” relates to information and communication technology (ICT) adapted for administrative, commercial, financial, managerial, or supervisory purposes, and does not encompass liquid pumps, let alone issues of chemical compatibility of pumps.
Comments
Whilst it is clear that patents should not be granted if they lack IS, the way the ED handled the present case was not correct. The IPER was negative as far as IS was concerned on the basis of the documents in the ISR. In particular D1 dealt with a pump. Why was it necessary to bring in the procedure a document which was much further away?
The present decision is a reminder that the CPA is prior art having a similar purpose and overall technical effect.
It is also a reminder to examiners and opponents, that not every document having some links with the claimed subject matter can be taken as CPA. Comparing RFI data to ensure that the correct chemical is dispensed into the correct container so as to avoid a mislabelling has nothing to do with chemical compatibility between pump and liquids going through pumps.
It is interesting to note that the some of CPC classes for this application are F04B 15/04, F04B 49/10, where F04B deals with pumps. A further search in pumps appears appropriate. My guess is that the compatibility between pumps and liquid going through pumps is not a recent problem. A pump moving liquids for human consumption should not move dangerous chemicals for humans.
Comments
9 replies on “T 0610/24-On the correct choice of the CPA-No “hindsight” OTP”
In my experience, the “correct” choice of the CPA seems to be a controversial topic. A colleague recently introduced me to T2759/17, where the board discusses two approaches to identify the CPA:
In a first approach, the CPA is selected by the deciding body, and the skilled person is then chosen on the basis of the objective technical problem identified, meaning that the skilled person is not necessarily the person working in the field of the patent application or the closest prior art.
This approach may be supported by the fact that the objectivity of the PSA is preserved as far as possible: if the skilled person selects the CPA themselves (based on which distinguishing features, etc. are then determined), they would effectively set themselves the objective technical problem.
In a second approach, the skilled person selects the closest prior art as the most promising springboard from the get-go.
This approach may be supported by the fact that it reflects a technically realistic situation: the skilled person is the relevant point of reference from the outset and selects the starting point in a purpose-oriented manner. This is supposed to avoid an ex post selection of the structurally closest prior art.
The decision discussed in this blog post seems to align with the second approach (focussing on the formulation of a realistic technical problem). This is also what is suggested by the EPO GL chapter G-VII, 5.1:
“The closest prior art must be assessed from the skilled person’s point of view on the day before the filing or priority date valid for the claimed invention. The examiner must not interpret the closest prior art artificially on the basis of prior knowledge of the application”
It will be interesting to see whether technical boards will converge to one of those approaches or not.
@ Dan B,
Thanks for your comment.
The selection of the CPA has often be one of the points of critic of the PSA.
The wrong choice of the CPA can lead either to a positive or a negative assessment of IS.
This is why the choice of the CPA is to be made carefully.
In the present case, it lead to a negative assessment of IS, but in my opinion, the ED did not apply the PSA correctly. For a start, the ED held that D5 disclosed a pump, which is not the case as shown by the board. To me the ED appears to have chosen D5 as CPA as it appeared to have, allegedly, the most features in common with the claim. This is not a criterion to be used when applying the PSA.
A tube with holes can be a drainage pipe, a cheese grater, a piece of exhaust, a washing machine drum, a hair curler or even a stent. If the claim relates to a drainage pipe, the skilled person would never start as CPA with a hair curler or a stent.
The argumentation of the ED was thus flawed from the beginning.
You are right when you say that in T 2759/17, the board distinguished two approaches for selecting the CPA. In a first approach, it is the deciding body which selects the CPA and in the second approach it is the skilled person.
I confess that I do not agree that it is the deciding body which has to select the CPA. As the claim has to be read and interpreted by the skilled person, it follows necessarily that the CPA has to be selected by the skilled person and not by the deciding body.
I personally consider that the decisions quoted in Reasons 5.3.1 of T 2759/17 are at odds with most of the decisions concerning the choice of the CPA. Reasons 5.3.2 of T 2759/17 give examples of decision according to this second line. The established case law corresponds to the second line and certainly not to the first one, cf. Reasons 5.4 of T 2759/17.
I presently consider that the few decisions at odds with the established case law should not be followed.
I furthermore do not like the notion of “most promising springboard” as such. To me the most promising springboard is prior art having a similar purpose and overall technical effect. As a summary one can then use the wording “most promising springboard”, but not as selection criterion as such.
The prior art having a similar purpose and overall technical effect, is in general the one needing the least modifications when solving the OTP and leading to the claimed subject-matter, and it is only in this respect that the latter has the most features in common with the claim. Selecting the CPA because it has the most features in common with the claim is not a reasonable choice. Here again, it can come as summary, but nothing more.
Thanks for your comments. I am also inclined to prefer the second approach, as it appears more “natural” to me. This is probably what the board meant with “technically realistic situation”.
I fully agree with your thoughts on the most promising springboard. In my carreer so far, focusing on a similar purpose to choose the CPA has always proven to be the most efficient (and effective) way of persuading the deciding body of the presence/absence of inventive step.
@ DanB
I can only but agree that starting with a realistic piece of prior art is best way to correctly assess IS.
For a proprietor it is always tempting to start with a further away piece of prior art as “CPA”, since it is then easier to show IS.
For an opponent, it is always tempting to start with a a piece of prior art which has the most features in common with the claim as “CPA”, since it is then easier to show a lack of IS.
Both approaches have the problem of not being a realistic starting point.
My preference for fixing the prior art disclosure with which one begins the enquiry into obviousnes is to call it the “starting point” and accept as such only a disclosure that would have been “realistic” for the notional person, skilled in the correct technical field, to consider as a starting point for solving the specific technical problem. At the outset, that problem is going to be the one identified in the WO or A publication, the “subjective problem” so to speak. Now suppose that applicant narrows the claim (from digging implement to spade, or from slotted tube to stent). That will likely have a knock-on effect, reducing the number of available prior art starting points because not all of them that were “realistic” given the subjective problem are still realistic relative to the narrowed claim. EPO-PSA, correctly performed, eliminates impermissible ex post facto analysis and is fair to both the public and the inventor.
Dear Max Drei,
I can only but agree with your statement.
To me the discussion whether the deciding body or the skilled person should decide on the CPA, is rather academic.
Any “starting point” has to be realistic or reasonable, that is, it has to do something, preferably directly, with the claimed subject-matter. Anything else, would lead to an ex-post facto assessment of IS.
I concur with DanB and I understand Mr Thomas also agrees.
I had touched on this topic in my comment of 26 December on the thread of Mr Thomas’ post of 18 December relating to T 2044/23. I pointed out in this comment the addition to the CLR in the July 2025 version of sections I.D.3.2 « Approaches to selecting the closest prior art » and I.D.8.1.2 « Point at which the skilled person enters the stage3, signalling a major divide between the BOAs over this issue.
If the CPA is to be chosen by the skilled person, the implication is that the definition of the « art » of the skilled person should be the first step of the PSA, and this definition can only be based on the content of the application as filed, including the technical field and the problem as exposed in the description.
The whole point of the PSA is to mitigate hindsight-guided reasoning, keeping in mind as explained in the guidelines that the search of prior art can only rely on the foreknowledge of the claimed subject matter and is inevitably hindsight-guided. The north star in the assessment of IS must be to minimise hindsight.
If the selection of the CPA is based on prior art found after the filing of the application and of which the applicant was not aware, it is guided by hindsight. The risk is then that the selection will be based on common features between the claimed subject matter and the prior art document, even if the document does not pertain to the art of the skilled person and/or does not address or mention the problem exposed in the application.
@ Mr Hagel,
Thanks for reminding us on your comments on the post about T 2044/23.
Whilst I understand Mr Hagel’s position, I have my reservations about it.
As explained in my comment to Max Drei, whether the CPA is chosen by the deciding body or the skilled person, is rather academic. Any “starting point”=CPA has to be “realistic” or reasonable, that is, it has something to do, preferably directly, with the claimed subject-matter.
One important aspect is that we are in presence of a fiction. It is often forgotten that the applicant started with PA known to him. However, the PA to be taken into account is any piece of prior art published anywhere before the effective date of the claim. This PA is that revealed by the documentary search or the opponent when it opposes a patent.
The chances to find PA coming closer to the claimed subject-matter is thus very high. In order to be chosen as CPA a document not know from the applicant has necessarily to have a technical relation with the claimed subject-matter.
Deciding on a piece of prior art having the most features in common with the claimed subject-matter can indeed be considered tainted by hindsight. This is why this selection criterion is not the right one.
The selection criterion is that the CPA has to have the same purpose or effect or presents a similarity with the technical problem solved by the claimed subject-matter. To me such PA is to be selected as CPA as it represents a “realistic” starting point.
I am not in favour of other criteria like most promising springboard towards the invention, most promising starting point, easiest route to arrive at the claimed invention or needing a minimum of changes. Those can reasonably be considered tainted by hindsight.
Interesting, the question whether spring board imports impermissible hindsight. I had always thought not but I know otherwise now. As to the”same effect” criterion, when the argument for patentability includes the notion of a problem invention then I presume that the number of starting points that are “realistic” is correspondingly large (or else vanishingly small).