Inventive Step In Current EPO Practise APLF Roundtable, 17 May 2006

1
Inventive Step In CurrentEPO Practise( APLF
Roundtable, 17 May 2006 )

• ________________________________________
• Dr. Walter Maiwald

2
Inventive Step (Obviousness)
Intro
Inventive step is a difficult topic in all patent
law systems, since it requires an objective
analysis of prior art disclosures, uninfluenced
by knowledge of the invention. The greatest
dangers involved are subjectiveness and
hindsight. The EPO has developed an analytical
approach to the evaluation of inventive step,
known as the "problem and solution" approach,
which aims at avoiding these dangers.
3
Some important regulations
Intro

• Article 52 (1) EPC Article 56 EPC Rule 27
  EPC
• European patents shall An invention shall be
  considered (1) The description shall
• be granted for any inven- as involving an
  inventive step if,
• tions which are susceptible having regard to
  the state of (a) specify the technical field
  to which the invention
• of industrial application, the art, it is not
  obvious to a relates,
• which are new and which person skilled in the
  art. If the
• involve an inventive step. state of the art
  also includes (b) indicate the background art
  which, as far as known
• documents within the meaning to the
  applicant, can be regarded as useful for
  under- of Article 54, paragraph 3, these
  standing the invention, for drawing up the
  European
• documents are not to be con- search
  report and for the examination, and preferably,
• sidered in deciding whether cite the
  documents reflecting such art
• there has been an inventive step.
• (c) disclose the invention, as
  claimed, in such terms
• that the technical problem
  (even if not expressly stated
• as such) and its solution can
  be understood, and state
• any advantageous effects of
  the invention with reference
  to the background art
• (d) briefly describe the
  figures in the drawings, if any

4

Intro

• The "problem and solution" approach (derived from
  Rule 27 EPC)
• comprises four steps
• identifying the "closest prior art", i.e. a
  single prior art disclosure
• serving as the "starting point" of the analysis
• (b) determining the technical results
  (advantages) objectively
• achieved by the features as claimed, when
  compared to this
• prior art "starting point" and
• correspondingly defining the technical problem to
  be solved, i.e. to achieve these results when
  setting out from this "starting point" and
• examining whether a skilled person would have
  suggested the
• claimed features to achieve these results, in
  view of the complete
• state of the art of Article 54 (2) EPC.
• It is apparent that steps (b) and (c) are two
  aspects of the same analysis.

5

Intro
This analysis has (ideally) no subjective
elements it is fully objective (e.g. T 248/85,
OJ 1986, 261). The motivation, knowledge, skill
and all other subjective characteristics of the
inventor are irrelevant. The inventor must not
be confused with the skilled person of Article 56
EPC (T 39/93). The relevant state of the art
comprises all disclosures publicly available at
the applicable priority date of the case (e.g. T
971/95).
6

Intro
The correct use of the "problem and solution"
approach rules out a hindsight analysis, cf. e.g.
T 730/96. The "problem and solution" approach is
not obligatory (T 465/92, OJ 1996, 32) but almost
always applied (at least by the Appeal
Boards). The EPOs Case Law book actually calls
T 465/92 a "one-off decision".
7
Identifying the "closest prior art" - Step (a)
The "starting point" does not have to be a
printed reference. It can be the subject-matter
disclosed by a prior public use, e.g. the
characteristics of a prior art product (G1/92, OJ
1993, 27, in the context of anticipation).
8
Identifying the "closest prior art" - Step (a)
What is the standard for selecting the "starting
point"? - "Most features in common"? - "Some
purpose/some objective"? The first alternative
is quite often used, even at Appeal Board level.
The objective of the "starting point" reference
is ignored, and the analysis does not go beyond a
numerical comparison of items in a feature
list. It uses the invention as the search
criterion for finding the "starting point", and
this invariably favours a hindsight analysis.
9
Identifying the "closest prior art" - Step (a)
The second alternative is the only correct
approach (e.g. T 495/91 T 989/93 T 263/99). A
reference generally does not suggest a solution
to a technical problem which it fails to address
(e.g. T 686/91 T 59/96 T 644/97). It is
however appropriate to select that disclosure,
from a plurality of references addressing the
same technical problem, which has most features
in common with the claimed invention (e.g. T
606/89 T 1203/97 T 263/99).
10
Identifying the "closest prior art" - Step (a)
Setting out from a common technical problem
(objective) creates new issues, since the
specification - may fail to define any technical
problem (this is indeed permitted by Rule 27 (1)
(c) EPC) - may identify a technical problem or
objective in view of a prior art which does not
qualify as the "starting point" - may even
contain a wrong assessment of the technical
effects provided by the claimed invention,
and/or a wrong assessment of the technical
effects provided by the prior art.
11
Identifying the "closest prior art" - Step (a)
Nevertheless, the analysis normally starts from
the problem (left to be overcome in the prior
art) both as described (or implied, e.g. by the
discussion of disadvantages of the prior art) in
the specification. Only if examination show that
the problem disclosed has not been solved or if
inappropriate prior art were used to define the
problem, it is necessary to investigate which
other problem objectively existed. The
definition of artificial and technically
unrealistic problems is to be avoided (see e.g. T
246/91, T 813/93, T 68/95, T 747/97). This legal
principle is also applicable to ex parte
proceedings (T 881/92, T 882/92 and T 884/92).
12
Defining the technical problem by comparing to
the "starting point" - Steps (b) and (c)
The "technical problem" of the invention is
found by assessing the objective technical
progress made in the invention over the closest
prior art (e.g. T 20/81, OJ 1982, 217), as
understood by a skilled person at the relevant
priority date (T 268/89, OJ 1994, 50). The
technical problem must not be formulated in a
way that includes pointers to the solution (e.g.
T 229/85, OJ 1987, 237).
13
Defining the technical problem by comparing to
the "starting point" - Steps (b) and (c)
The "technical progress" must be supported by
evidence (e.g. T 20/81, OJ 1982, 217) which may
be disclosed in the case as filed, but can also
be supplied after filing (Guidelines, C IV 9.11)
and even in opposition procedures. Evidence in
support of the "technical progress" (or, in other
words, evidence for the advantages provided by
the invention) must permit a comparison with the
"closest prior art" of Step (a).
14
Defining the technical problem by comparing to
the "starting point" - Steps (b) and (c)
The closest art is often not that reference on
which the comparative examples of the original
specification were based. The closes prior art
may appear for the first time in prosecution or
even opposition. In such cases, the technical
problem may have to be reformulated, and/or new
factual evidence may have to be submitted. The
new technical problem can be less ambitious
(e.g., no improvement, only an alternative to the
subject-matter of the closest prior art). The new
technical problem must however be related to the
problem identified in the case as filed (T
184/92, OJ 1984, 261).
15
Defining the technical problem by comparing to
the "starting point" - Steps (b) and (c)
Where no evidence is available which supports an
improvement over the closest art, the EPO will
often assume that the technical problem was
merely to provide an alternative without
improvement this may be considered as obvious,
where an improvement would have been
unobvious. Where the specification reflects
embodiment examples which perform worse than the
prior art comparisons, the EPO will often
conclude that the claimed invention does not
solve the problem over the breadth of the claim,
and the claim is therefore at least partly not
based on an inventive step.
16
Defining the technical problem by comparing to
the "starting point" - Steps (b) and (c)
On the other hand The absence of any reference
in the prior art which expressly addresses the
technical problem solved by the claimed
invention, is generally a strong indicator of
inventive step.
17
Defining the technical problem by comparing to
the "starting point" - Steps (b) and (c)
The analysis of steps (a) to (c) thus often
already gives some guidance on the presence of
inventive step, even before further art is
considered New Problem solution
provided ?gt Not obvious Known Problem better
solution provided ?gt Not obvious Known
Problem alternative solution provided ?gt Often
obvious Known Problem inferior solution
provided ?gt Obvious
18
Examining the complete state of the art - Step (d)
The last step involves a study of the complete
pre-published art (Article 52 (2)
EPC). Generally, this includes art from
"neighboring fields" (T 176/84, OJ 1986, 50 T
195/84, OJ 1986, 121) and common general
knowledge such as journals, textbooks,
encyclopedias etc. (T 766/91 T 939/92, OJ 1996,
309).
19
Examining the complete state of the art - Step (d)
The analysis is simple where one reference leaves
a gap (filled by the invention) and another
reference explicitly teaches that the gap could
be thus filled. In the early days of the EPO,
this was almost the only scenario where inventive
step was denied ("one way street case"). The
analysis is also simple where the inventive
feature(s) are not found in any reference this
generally implies inventive step.
20
Examining the complete state of the art - Step (d)
The skill of the skilled person, in selecting and
evaluating the state of the art, corresponds to
that of an "ordinary practitioner" void of any
inventive or creative thinking (T 39/93 T
500/91). While able to carry out routine
experimental work (e.g. T 791/96) the skilled
person is unable to do original scientific
research (T 500/91 T 223/92).
21
Examining the complete state of the art - Step (d)
In looking at the complete state of the art, it
is inappropriate to confuse technical feasibility
(the mere availability of the technical features
used by the invention) with lack of inventive
step (the obviousness of employing such features
to solve the problem underlying the
invention). The criterion is not, whether the
invention could have been made (this only
addresses availability) but whether it would have
been made (this addresses obviousness). This
"could and would" rule is e.g. found in T 02/83
T 07/86 T 885/97.
22
Examining the complete state of the art - Step (d)
Generally, the "would" aspect requires some
prompting in the state of the art, to add those
features by which the claimed invention differs
from the "starting point" disclosure (e.g. T
219/87 T 414/98). Note that Appeal Board
practise differs widely in applying these rules.
Some Appeal Board do not even formally base their
decisions on this analysis (while they may
initially mention the "problem and
solution"-approach for the sake of political
correctness). Often, the result of the Step (a)
to (c) analysis, especially where experimental
results comparing with the "closest prior art"
show superiority of the claimed invention, is
much more important.
23
Does the EPO use "obvious to try"?
The EPO has consistently denied that "obvious to
try" is acceptable. The correct standard is
instead, whether in view of the state of the art,
a skilled person would have had a "reasonable
expectation of success" (e.g. T 60/89, OJ 1992,
268). This must not be confused with a mere
"hope to succeed", since it implies some
capability of rationally predicting the result (T
696/92, OJ 1997, 408 T 207/94, OJ 1999, 273).
24
Does the EPO use "obvious to try"?
It was held (T 187/93) that even where an
experiment was obvious to try, this did not imply
a reasonable expectation of success. Note
however, that where a feature combination was
obvious to achieve some advantage, the occurrence
of an unexpected further advantage does not
generally constitute inventive step. Such
further advantages are considered as "bonus
effects" (e.g. T 21/81, OJ 1983, 15 T 231/97).
25
Indicia for inventive step
Besides the already discussed indicia "new
problem", "new solution" and "superior results",
secondary indicia are often relied on. These
include - prejudice in the art rarely
accepted (e.g. T 69/83, OJ 1984, 357) - time
factor (old references sometimes accepted
(e.g. T 203/93) - long felt need
sometimes accepted (e.g. T 271/84, OJ 1987,
405) - commercial success generally not
accepted (e.g. T 712/92) - simple
solution often accepted (e.g. T
234/91) - surprising effect generally
accepted (e.g. T 181/82 OJ 1984, 401) (but
cf. "bonus effect"!)
26
Thank you for your attention.
Dr. Walter Maiwald maiwald_at_maiwald.de
Maiwald Patentanwalts GmbH
www.maiwald.de