14-15 November 2018
|European Case Law Identifier:||ECLI:EP:BA:2012:T120612.20121214|
|Date of decision:||14 December 2012|
|Case number:||T 1206/12|
|IPC class:||G06N 5/00|
|Language of proceedings:||EN|
|Download and more information:||
|Title of application:||Optimization of decisions regarding multiple assets in the presence of various underlying uncertainties|
|Applicant name:||LANDMARK GRAPHICS CORPORATION|
|Relevant legal provisions:||
|Keywords:||Decision sufficiently reasoned - no
Substantial procedural violation - yes
Remittal for further prosecution
The mere reference to jurisprudence of the boards of appeal does not, by itself, constitute or re place an argument in a first instance decision. If a de ciding body, in a decision, wants to rely on an ar gu ment put forward in a decision of the boards it is insufficient merely to refer to it or to recite it. The deciding body must also make clear that it adopts the argu ment and explain why, in what respect and to what extent this argument applies to the case at hand.
Summary of Facts and Submissions
I. The appeal lies against the decision of the examining division, with written reasons dated 30 December 2011, to refuse European application no. 06751882.9 for lack of an inventive step only in view of common knowledge and without reference to any specific document.
II. Notice of appeal was filed on 7 March 2012, the appeal fee being paid on the same day. A statement of grounds of appeal was filed on 9 May 2012. The appellant re ques ted that the decision be reversed and a patent be granted, based on claims 1-28 as re-filed with the grounds of appeal in combination with, as the board un derstands the request, the description and the draw ings as originally filed. Failing this, the appellant re ques ted that oral proceedings be held.
III. In response to the board's communication that it inten ded to remit the case to the first instance for further prosecution, the appellant withdrew its request for oral proceedings.
IV. Pending claim 1 reads as follows.
"A method for determining how many wells to drill in a plurality of oil and gas field assets, comprising:
(a) receiving, on a first computer, information specifying decision variables and uncertainty variables for the plurality of oil and gas field assets and specifying for each oil and gas field asset a corresponding set of one or more algorithms, wherein the set of one or more algorithms for at least one of the oil and gas field assets includes an algorithm to estimate oil and gas production over time, and wherein the decision variables represent the possible outcomes for the number of wells to be drilled in the given oil and gas field asset;
(b) generating, by the first computer, a decision vector, wherein the decision vector includes a value for each of the decision variables;
(c) executing, by the first computer, an evaluation process on the decision vector to determine at least a value of a global objective for the decision vector, wherein said executing the evaluation process includes:
(c1) generating an uncertainty vector, wherein the uncertainty vector includes a value for each of the uncertainty variables;
(c2) generating for each oil and gas field asset a data set for the corresponding set of algorithms using a corresponding subset of one or more values in the decision vector and a corresponding subset of one or more values in the uncertainty vector;
(c3) invoking execution of the set of algorithms on one or more second computers for each oil and gas field asset on the corresponding data set to obtain output arguments corresponding to the oil and gas field asset;
(c4) computing for each oil and gas field asset one or more values of one or more respective quantities form the output arguments corresponding to the oil and gas field asset;
(c5) performing operations including (c1) through (c4) a number of times, thereby generating a population of values of each quantity for each oil and gas field asset;
(c6) computing one or more statistics for each oil and gas field asset based on corresponding ones of the population; and
(c7) combining a first subset of the statistics to determine the global objective value corresponding to the decision vector;
(d) executing, by the first computer, an optimizer, wherein said executing the optimizer includes performing (b) and (c) repeatedly, thereby generating a plurality of decision vectors and corresponding global objective values, and, wherein the optimizer uses the plurality of decision vectors and the corresponding global objective values to generate and repeatedly update a reference set of decision vectors in order to optimize the global objective;
(e) storing, by the first computer, data including the plurality of decision vectors and their corresponding global objective values in a memory."
Reasons for the Decision
1. Article 113 (1) EPC 1973 provides that decisions of the European Patent Office may only be based on grounds or evidence on which the parties concerned have had an oppor tunity to present their comments. Rule 68 (2) EPC 1973 further provides that decisions of the Eu ro pean Patent Office which are open to appeal must be rea soned.
1.1 According to established jurisprudence of the boards of appeal (see e.g. T 763/04, not published; reasons 4.3), Article 113 (1) EPC 1973 is contravened where facts and argu ments, which from the appellant's sub missions are clear ly cen tral to his case and which may speak against the de ci sion taken, are completely dis re ga r ded in the decision in question.
1.2 Further, established jurisprudence of the boards of appeal explains Rule 68 (2) EPC 1973 as requiring that a decision must contain, in logical sequence, those ar gu ments which justify the te nor. The grounds upon which a decision is based and all decisive consi de rations in res pect of the factual and legal aspects of the case must be discussed in detail in the decision (see e.g. T 278/00, OJ 2003, 546; reasons 2) so as to enable the par ties and, in case of an appeal, the board of appeal to examine whether the de ci sion was jus tified or not.
1.3 In the board's judgment the decision under appeal vio lates both fundamental requirements for the following reasons.
2. Claim 1 defines a method "for determining how many wells to drill in a plurality of oil and gas field assets" which is defined by a number of algorithmic, mostly mathematical steps operating on variables re la ting to these assets, in particular on decision vari ables which represent the desired number of how many wells to drill (see step a). Moreover, as part of the method, algorithms making asset-specific predictions are executed, too (see steps a and c3).
2.1 That a major part of the claimed method is a mathe ma tical one was uncontroversial between the examining di vision and the applicant. It was however contro ver sial whether the mathematical steps according to claim 1 must or need not be taken into account when assessing inventive step.
2.2 The applicant filed the claims subject to the appealed decision in response to the examining divi sion's ob ject ion (see communication dated 23 June 2009) that the claims essentially con sti tu ted the imple men tation of a mathematical method as such on a computer. In the accom panying letter, the applicant argued that and why it considered that the method according to the amended claims was not a mathematical method as such, had tech nical character and provided the tech ni cal so lu tion to a technical problem (see submission of 29 De cember 2009, points II.1-3). In its response to the summons to oral proceedings, the applicant further de tailed its posi tion that the claimed invention "em ploy[ed] techni cal means" and was "limited to a par ti cu lar technical appli cation" (see submission of 11 No vember 2011, point III.1). It further invoked T 1227/05 to justify its opin ion that the claimed method cannot be denied a tech nical effect even though it does not in corporate the physical end product (in this case, the actual drilling of wells) since it "serve[s] a techni cal pur pose" and is also "functionally limited to that tech ni cal pur pose" (ibid., point III.2). A similar argu ment was based on T 49/99 (ibid., point III.3). Therefore, so the applicant's argument, all method steps, the ma the matical ones in clu ded, had to be taken into account when assessing inventive step (loc. cit.).
2.3 Apparently, this issue was the main con ten tious point when the examining division took its decision.
2.4 In its decision, the examining division provides a rather terse summary of claim 1 (see point 10) and then states that "all ... features defining the algorithm" except those relating to a "computer pro cessing data" neither have technical character nor "confer technical conside ra tions" on the claim (points 11-12).
2.5 The examining division does not, however, provide rea sons for this allegation. In particular, no reasons are provided as to why neither the claimed fact that "the assets are wells to drill" (decision, point 10) nor the purpose of the claim 1 according to its preamble, name ly to "determin[e] how many wells to drill" confer tech ni cal character on the claimed method.
3. Points 13-15 of the decision cite T 258/03, T 641/00 and T 1173/97 as established jurisprudence and suggest that the examining division means to follow the prin ciples laid out in these decisions. This could not have sur prised the applicant given that the applicant itself referred to pertinent jurisprudence of the boards of appeal and to a Practice and Pro cedure No tice of the EPO (PPN 04/08; see applicant's sub mission of 11 No vem ber 2011, point II.1 and 2).
3.1 However, the mere reference to jurisprudence of the boards of appeal does not, by itself, constitute or re place an argument in a first instance decision. If a de ciding body, in a decision, wants to rely on an ar gu ment put forward in a decision of the boards, be it part of its ratio decidendi or an obiter dictum, it is insufficient merely to refer to it or to recite it. The deciding bo dy must also make clear, by explicit state ment or by un ambiguous implication, that it adopts the argument and explain why, in what respect and to what extent this argument applies to the case at hand.
3.2 The decision under appeal however does not explain how the cited decisions apply to the case at hand. Specifi cally, it does not ex plain why the examining division considers, as it appears to, that the fact that "assets are wells to drill" and the aim to "calculate the num ber of wells to drill" do not contribute to the tech ni cal character of the claim and thus cannot, accor ding to T 641/00 (OJ 2003, 352; catchword 1), con tribute to in ventive step. The decision also does not formulate the technical prob lem in accordance with T 641/00 (catch word 2) which, if any, the claimed inven tion is considered to solve, nor argues why it does not con si der that the preamble of claim 1 ("how many wells to drill in a plurality of oil and gas field assets") con sti tutes this technical problem as the applicant pro posed (see submission of 29 December 2009, point II.3, esp. 2nd and 3rd dash, and submission of 11 Novem ber 2011, point III.1).
4. With reference to T 1227/05 the examining division only argues (decision, point 16) that "the specific purpose of the procedural steps of claim 1 is not reflected in the aim of the claim, namely to calculate the number of wells to drill". This sentence is the only one in the decision under appeal which responds directly to the applicant's argument based on T 1227/05.
4.1 In the board's view, the precise meaning of this state ment is unclear as it stands, but it suggests at least that the "steps of claim 1" are, in the opinion of the examining division, somehow detached from the "purpose of the claim". This allegation however appears at least not to be completely accurate, for in stance insofar as step a spe cifies the method to ope rate on "decision va ri ables represent[ing] the possible outcomes for the num ber of wells to be drilled".
4.2 Even assuming, arguendo, that the "purpose of the ... steps of claim 1" is indeed not "reflected in the aim of the claim", it remains unclear from the decision why this is an objectionable deficiency and, in particular, what bearing this deficiency has on the overall finding of lack of inventive step. As it stands, the statement in point 16 might also be read as a clarity objection.
5. In point 17 of the decision under appeal, the examining division argues that "point 7 of T 1029/06 (Toshiba) ve ry well reflects the line of reasoning for the current application", and cites it: "the majority of the steps are typical tools of operational research that normally have not been considered technical in the established jurisprudence, in particular if only infor mation is pro cessed for the purpose of management decisions".
5.1 The board's view, point 17 is ambiguous as to whether the examining meant the reference to T 1029/06 to be a constitutive part of its "line of reasoning" or a mere illustration of it. In the former case, it would appear arguable that, in conflict with Article 113 (1) EPC 1973, the applicant had hardly any opportunity to pre sent its comments on this part of the grounds for the refusal, since that the examining division had intro duced T 1029/06 during a telephone conversation a mere two weeks before the oral pro ceedings and, according to the mi nu tes of this conversation, ex plained its per ti nence with the same terse words as used in point 17 of the decision.
5.2 Moreover, the appealed decision does not explain in what sense the cited passage of T 1029/06 has any bea ring on the inventive step of claim 1 (see point 3.1 above). First, it would seem from the applicant's argu ments that it would probably oppose the view that the steps of claim 1 are "typical tools of opera tio nal research" or are meant to process information only "for the purpose of management deci sions", and this position would appear to be at least prima facie plausible and therefore cannot be silently dismissed by the examining division. Second, the cited passage of T 1029/06 refers only to "the majority of steps" which leaves open the possibility that the remaining steps may be technical and sufficient to establish an inven tive step.
6. The board further notes that the applicant had argued in favour of inventive step that the claimed invention would "provide a massive amount of parallelism for the overall calculation" "by dispatching the execution of the set of algorithms from the first computer to one or more second computers" (see submission of 29 December 2009, p. 5, last par.), but that neither the summons to oral proceedings nor the decision under appeal addressed this argument.
7. In the board's judgment, the reasons given in the de ci sion under appeal do not conform with Rule 68 (2) EPC 1973 because they are insufficient for the applicant to understand why the examining division considered the claimed method to lack an inventive step. Moreover, the reasons do not conform with Article 113 (1) because they are insufficient for the appli cant to under stand why the examining division dismissed the appli cant's argument provided in the submissions dated 11 No vember 2011 and 29 December 2009.
8. The violation of a party's right to be heard and in sufficiency of reasoning are both substantial pro ce du ral violations which require the decision under appeal to be set aside and the case to be remitted to the first instance without the board assessing the sub stan tive merits of the case, see Article 11 RPBA. The appeal is thus to be allowed and the Board con siders it equitable by reason of the sub stan tial procedural vio la tions to reimburse the appeal fee (Rule 67 EPC 1973).
For these reasons it is decided that:
1. The decision under appeal is set aside.
2. The case is remitted to the department of first instance for further prosecution.
3. The appeal fee is reimbursed.