Georgetown Rail Equipment Company

Patent Trials and Appeals Board

Aug 25, 2020

IPR2019-00620 (P.T.A.B. Aug. 25, 2020)

Trials@uspto.gov Paper No. 26
571-272-7822 Date: August 25, 2020

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

TETRA TECH CANADA INC.,
Petitioner,
v.
GEORGETOWN RAIL EQUIPMENT COMPANY,
Patent Owner.
____________

IPR2019-00620
Patent 9,441,956 B2
____________

Before HUBERT C. LORIN, BENJAMIN D. M. WOOD, and
KRISTINA M. KALAN, Administrative Patent Judges.

KALAN, Administrative Patent Judge.

JUDGMENT
Final Written Decision

Determining All Challenged Claims Unpatentable
35 U.S.C. § 318(a)

Denying Patent Owner’s Contingent Motion to Amend
35 U.S.C. § 316(d)

IPR2019-00620
Patent 9,441,956 B2

2

I. INTRODUCTION
Tetra Tech Canada Inc. (“Petitioner”) filed a Petition requesting inter
partes review of claims 1–3, 7–13, and 16–20 of U.S. Patent No. 9,441,956
B2 (Ex. 1001, “the ’956 patent”). Paper 2 (“Pet.”). Georgetown Rail
Equipment Company (“Patent Owner”) filed a Preliminary Response to the
Petition. Paper 7 (“Prelim. Resp.”). We instituted an inter partes review of
claims 1–3, 7–13, and 16–20 of the ’956 patent on all grounds of
unpatentability alleged in the Petition. Paper 8 (“Institution Decision” or
“Dec.”).
After institution of trial, Patent Owner filed a Patent Owner Response.
Paper 10 (“PO Resp.”). Patent Owner also filed a Contingent Motion to
Amend. Paper 11 (“Motion”). The Motion requests the Board cancel
original claims 1–3, 7–13, and 16–20, if those claims are found
unpatentable, and replace each unpatentable claim with corresponding
proposed substitute claims 30–44. Id. at 1. Subsequently, Petitioner filed an
Opposition to the Motion to Amend. Paper 16 (“Opp.”). Petitioner also
filed a Reply. Paper 15 (“Reply”). A consolidated oral hearing with related
cases IPR2019-00619 and IPR2019-00662 was held on May 18, 2020, and a
transcript of the hearing is included in the record. Paper 25 (“Tr.”).
This Final Written Decision is issued pursuant to 35 U.S.C. § 318(a).
For the reasons that follow, we determine that Petitioner has established by a
preponderance of the evidence that claims 1–3, 7–13, and 16–20 of the ’956
patent are unpatentable.
We also determine that Petitioner has established by a preponderance
of the evidence that Patent Owner’s proposed substitute claims 30–44 are
IPR2019-00620
Patent 9,441,956 B2

3

unpatentable, and, therefore, we deny Patent Owner’s Contingent Motion to
Amend.
A. Related Proceedings
The parties identify the following proceeding as involving the ’956
patent: Georgetown Rail Equipment Company v. Tetra Tech Canada Inc.,
No. 6:18-cv-377 (E.D. Tex.). Pet. 4; Paper 4, 2. The parties also state that
the ’956 patent is the subject of an additional Board proceeding, IPR2019-
00662. Pet. 2; Paper 4, 2.
B. The ’956 Patent
The ’956 patent, titled “System and Method for Inspecting Railroad
Ties,” discloses “systems and methods for inspection and grading of wood
railroad ties.” Ex. 1001, code (54), 1:28–29. In particular, the ’956 patent
describes a “computerized system and method for inspecting railroad tracks,
which can both identify and locate railroad ties, tie plates, and appurtenant
structures, and also determine condition metrics” of these, and further “can
provide a condition grade or score for railroad ties and other features.” Id.
at 2:15–22. The system can scan and grade the condition of the railroad ties
by capturing images of a railroad track and then analyzing those captured
images. Id. at 14:24–27, 14:42–46. Figures 1 and 2 illustrate an inspection
system used to capture images of the railroad track, and are reproduced
below.
IPR2019-00620
Patent 9,441,956 B2

4

Figures 1 and 2 show systems for inspecting railroad tracks. Id.
at 3:24–28. As shown in both figures, “inspection system 30 includes a light
generator such as a laser 40” which “projects a beam 42 of laser light at the
track bed.” Id. at 4:24–25, 4:39–40. As shown in Figure 2, “beam 42
produces a projected line L.” Id. at 4:40–41. Projected line L “follows the
contours of the surfaces and components of the track bed,” e.g., the
“crossties 10, rails 12, tie plates 14, spikes 16, and ballast 18” (id. at 4:38–
43) and, further, projected line L “is substantially straight and extends
substantially across the track bed” (id. at 5:11–12, 18–20). Then, “a device
for receiving light reflected from the area to be inspected such as a camera
50” captures “an image of the line L of laser light 42 projected on the track
bed.” Id. at 4:26–27, 4:43–45. “The camera 50 sends the captured image to
the processing device 60 for processing and analysis.” Id. at 4:45–46.
Processing and analysis of the captured image allows the inspection
system “to determine various condition metrics of the ties.” Id. at 18:10–12.
The condition metrics “can include a variety of factors that relate to the
IPR2019-00620
Patent 9,441,956 B2

5

function and longevity of railroad ties,” e.g., “tie warpage . . . tie height and
tie surface roughness,” (id. at 18:14–21) and “tie plate cut (i.e. a
measurement of the depth that a tie plate has worn, pressed or cut into a tie’s
surface)” (id. at 19:4–7). Further, once the “various tie condition metrics
have been determined, the system can be configured to generate or compute
a condition grade or score for each tie. This can be done by applying a
weight factor to each of the condition metrics.” Id. at 21:4–8.
C. Illustrative Claim
Of challenged claims 1–3, 7–13, and 16–20, claims 1, 8, and 17 are
independent. Claim 1 is reproduced below:
1. [pre] A system for inspecting railroad ties, comprising:
[a] a. a rail vehicle, configured for moving along a railroad
track in a travel direction;
[b] b. a light generator, attached to the rail vehicle, oriented to
project a beam of light across the railroad track while moving
thereon;
[c] c. an optical receiver, attached to the rail vehicle, oriented to
receive at least a portion of the beam of light reflected from the
railroad track and configured to generate image data
representative of a profile of at least a portion of the railroad
track; and
[d] d. a processor, configured to analyze the image data by
applying one or more algorithms configured to:
[e] i. find boundaries of at least one railroad tie in the railroad
track; and
[f] ii. determine the condition of the at least one railroad tie by:
a. determining a series of condition metrics of the at least
one railroad tie;
b. applying a weight factor to each condition metric; and
c. computing a condition grade of the at least one railroad
tie based on the weighted condition metrics.

Ex. 1001, 23:36–58 (bracketed annotations added).
IPR2019-00620
Patent 9,441,956 B2

6

D. Asserted Grounds of Unpatentability
We instituted inter partes review of claims 1–3, 7–13, and 16–20 of
the ’956 patent on the following grounds. Dec. 5, 24.
Reference(s)/Basis 35 U.S.C. §1 Claims
Villar,2 Haas,3 Uzarski4 103 1–3, 7, 8, 11–13, and 16–20
Villar, Haas, Uzarski,
Velten,5 Kanade6
103 9, 10
Petitioner relies on the Declarations of Dr. Nikos Papanikolopoulos.
Exs. 1004, 1050. Patent Owner relies on the Declaration of Dr. Alan Conrad
Bovik. Ex. 2001.
II. ANALYSIS
A. Legal Standards
A claim is unpatentable under 35 U.S.C. § 103(a) if the differences
between the subject matter sought to be patented and the prior art are such
that the subject matter as a whole would have been obvious to a person of
ordinary skill in the art at the time the invention was made. KSR Int’l Co. v.
Teleflex Inc., 550 U.S. 398, 406 (2007). Obviousness is resolved based on

1 The Leahy-Smith America Invents Act (“AIA”), Pub. L. No. 112-29, 125
Stat. 284, 287–88 (2011), amended 35 U.S.C. § 103, effective March 16,
2013. Because the application from which the ’956 patent issued was filed
before this date, the pre-AIA version of § 103 applies.
2 U.S. Pat. No. 7,616,329 B2, issued November 10, 2009 (Ex. 1036).
3 U.S. Pub. No. 2012/0263342 A1, published October 18, 2012 (Ex. 1037).
4 Uzarski, D. R., Development of Condition Indexes for Low Volume
Railroad Track (1993) (Ex. 1038).
5 Velten, J. et al., Application of a Brightness-Adapted Edge Detector for
Real-Time Railroad Tie Detection in Video Images (1999) (Ex. 1041).
6 Kanade, Takeo, Three-Dimensional Machine Vision (1987) (Ex. 1010).
IPR2019-00620
Patent 9,441,956 B2

7

underlying factual determinations, including: (1) the scope and content of
the prior art; (2) differences between the prior art and the claims at issue;
(3) the level of ordinary skill in the art; and (4) objective evidence of
nonobviousness, i.e., secondary considerations. Graham v. John Deere Co.,
383 U.S. 1, 17–18 (1966). Secondary considerations may include the
following: “commercial success, long felt but unsolved needs, failure of
others, etc.”7 Id. The totality of the evidence submitted may show that the
challenged claims would not have been obvious to one of ordinary skill in
the art. In re Piasecki, 745 F.2d 1468, 1471–72 (Fed. Cir. 1984).
Petitioner bears the burden of proving unpatentability of the
challenged claims, and the burden of persuasion never shifts to Patent
Owner. Dynamic Drinkware, LLC v. Nat’l Graphics, Inc., 800 F.3d 1375,
1378 (Fed. Cir. 2015). Petitioner must demonstrate unpatentability by a
preponderance of the evidence. 35 U.S.C. § 316(e); 37 C.F.R. § 42.1(d); see
also Harmonic Inc. v. Avid Tech., Inc., 815 F.3d 1356, 1363 (Fed. Cir. 2016)
(citing 35 U.S.C. § 312(a)(3) (requiring inter partes review petitions to
identify “with particularity . . . the evidence that supports the grounds for the
challenge to each claim”)). A party that petitions the Board for a
determination of obviousness must show that “a skilled artisan would have
been motivated to combine the teachings of the prior art references to
achieve the claimed invention, and that the skilled artisan would have had a
reasonable expectation of success in doing so.” Procter & Gamble Co. v.

7 We note that Patent Owner did not present any evidence or arguments
directed to secondary considerations during this proceeding.
IPR2019-00620
Patent 9,441,956 B2

8

Teva Pharms. USA, Inc., 566 F.3d 989, 994 (Fed. Cir. 2009) (quoting Pfizer,
Inc. v. Apotex, Inc., 408 F.3d 1348, 1361 (Fed. Cir. 2007)).
B. Level of Ordinary Skill in the Art
Petitioner contends that a person of ordinary skill in the art “would
have a bachelor’s degree in electrical engineering, computer engineering,
mechanical engineering, computer science, physics, or a related field, and at
least four years of experience (or the academic equivalent) in the field of
computer or machine vision.” Pet. 21 (citing Ex. 1004 ¶ 42). Patent Owner
states that, for purposes of its Response, “Patent Owner does not contest the
level of ordinary skill in the art asserted by Petitioner.” PO Resp. 3.
Neither party argues that the outcome of this case would differ based
on our adoption of any particular definition of one of ordinary skill in the art.
In light of the record now before us, we find that a person of ordinary skill in
the art would be an individual with at least a bachelor’s degree in electrical
engineering, computer engineering, mechanical engineering, computer
science, physics, or a related field, and at least four years of experience (or
the academic equivalent) in the field of computer or machine vision.
Ex. 1004 ¶ 42. The level of ordinary skill in the art is also reflected by the
references themselves. Okajima v. Bourdeau, 261 F.3d 1350, 1355 (Fed.
Cir. 2001) (“[T]he absence of specific findings on the level of skill in the art
does not give rise to reversible error ‘where the prior art itself reflects an
appropriate level and a need for testimony is not shown.’”); In re GPAC
Inc., 57 F.3d 1573, 1579 (Fed. Cir. 1995) (finding that the Board of Patent
Appeals and Interferences did not err in concluding that the level of ordinary
skill in the art was best determined by the references of record).
IPR2019-00620
Patent 9,441,956 B2

9

C. Claim Construction
We apply the claim construction standard articulated in Phillips v.
AWH Corp., 415 F.3d 1303 (Fed. Cir. 2005) (en banc). See Changes to the
Claim Construction Standard for Interpreting Claims in Trial Proceedings
Before the Patent Trial and Appeal Board, 83 Fed. Reg. 51,340, 51,358
(Oct. 11, 2018) (amending 37 C.F.R. § 42.100(b) effective November 13,
2018) (now codified at 37 C.F.R. § 42.100(b) (2019)). Under Phillips, claim
terms are afforded “their ordinary and customary meaning.” Phillips, 415
F.3d at 1312. “[T]he ordinary and customary meaning of a claim term is the
meaning that the term would have to a person of ordinary skill in the art in
question at the time of the invention.” Id. at 1313. Only terms that are in
controversy need to be construed, and only to the extent necessary to resolve
the controversy. Vivid Techs., Inc. v. Am. Sci. & Eng’g, Inc., 200 F.3d 795,
803 (Fed. Cir. 1999).
Petitioner proposes that the “terms of claims 1–3, 7–13, and 16–20
should be given their plain and ordinary meaning,” and does not propose any
claim constructions for any particular claim terms. Pet. 21.
Patent Owner argues that, for purposes of its Response, “Patent
Owner believes that the terms of the claims at issue should be given their
plain and ordinary meaning, including the term ‘applying a weight factor to
each condition metric’ of claim 1, which Patent Owner requests that the
Board construe.” PO Resp. 3. Patent Owner argues that “one weight factor
is applied to the first condition metric, a second weight factor is applied to
the second condition metric, and so on.” Id. (citing Ex. 1001, 21:4–20).
Patent Owner argues that Petitioner “ignores the claim word ‘each’ in its
analysis by effectively construing it to mean that a single weight factor can
IPR2019-00620
Patent 9,441,956 B2

10

be applied to all of the condition metrics as a whole” and in essence
“construes the term to mean ‘applying a weight factor to the condition
metrics.’” Id. at 3–4 (citing Pet. 35–36).
Petitioner replies that Patent Owner’s proposed construction
“essentially rewrites claim 1 to recite ‘determining a series of condition
metrics of [each of] the at least one railroad tie.’” Reply 17. Petitioner
argues that the plain language of claim 1 “does not preclude determining a
series of condition metrics for multiple ties (e.g., a series of condition
metrics for a set of railroad ties).” Id.
On the full record now before us, we determine it is not necessary to
construe any claim term expressly to resolve the parties’ dispute. Vivid
Techs., Inc., 200 F.3d at 803 (“Only those terms need be construed that are
in controversy, and only to the extent necessary to resolve the
controversy.”); Nidec Motor Corp. v. Zhongshan Broad Ocean Motor
Co., 868 F.3d 1013, 1017 (Fed. Cir. 2017) (applying Vivid Techs. in the
context of an inter partes review).
D. Prior Art
i. Villar (Ex. 1036)
Villar is titled “System and Method for Inspecting Railroad Track.”
Ex. 1036, code (54). Figures 1 and 2 of Villar are reproduced below:
IPR2019-00620
Patent 9,441,956 B2

11

Figures 1 and 2 show systems for inspecting railroad tracks. Id. at 3:24–28.
Villar discloses an “inspection system 30 [that] is used to survey the track
bed of a railroad track” that “includes a light generator such as a laser 40
[and] a device for receiving light reflected from the area to be inspected such
as a camera 50.” Id. at 3:29–34. The inspection system captures the
reflected light as an image and “sends the captured image to the processing
device 60 for processing and analysis.” Id. at 3:48–51. The processing and
analysis of the captured image allows the inspection system to identify
various crosstie features, e.g., “the spacing between crossties” (id. at 7:55–
56) and “edges of the crosstie contour” to calculate the distance between
edges “to estimate the width W2 of the crosstie 10” (id. at 9:57–61).
Furthermore, “defects in the crossties 10 can be determined from the image
data,” e.g., “cracks, splits, or breaks in the ties.” Id. at 9:30–31, 9:38–39.
ii. Haas (Ex. 1037)
Haas is a published patent application titled “Method and System of
Rail Component Detection Using Vision Technology.” Ex. 1037, code (54).
IPR2019-00620
Patent 9,441,956 B2

12

Haas describes “using machine vision technology for more efficient,
effective and objective inspections” which “helps lower maintenance cost
and improve track capacity.” Id. ¶ 25. In particular, Haas describes a
“system of automatic railroad track inspection” that captures “videos 110 [of
a track] using multiple cameras” and uses “image and video analytics 114”
to detect important rail components “including tie plate 116, spike 118, spike
hole 118, anchor 120, joint bar and joint bar bolt 122.” Id. ¶ 27. Haas’s
system involves “comparing the configuration of the rail components to
safety requirements for the rail components” (id. ¶ 3) and “applying
knowledge from federal regulations or domain experts” (id. ¶ 37). Haas also
provides for:
1) monitoring of spiking and anchor patterns; 2) detection of
spikes whose heads are raised above the tie plate by more than
one inch, as well as spikes that are deadheads; 3) detection of
displaced anchors that have moved more than a half inch away
from the tie; and 4) detection of any missing bolts, as well as
missing nuts and washers, from rail joint bars.
Id. ¶ 25.
iii. Uzarski (Ex. 1038)
Uzarski is titled “Development of Condition Indexes for Low Volume
Railroad Track.” Ex. 1038, 4.8 Uzarski describes condition indexes that
“were designed to provide an objective and quantitative means for facility
condition assessment” (id. at 24), e.g., for “component groups of rail, joints,
and fastenings; wooden cross ties and switch ties” (id. at 17). The condition
indexes have a numerical scale that “ranges from 0 to 100” and “is divided

8 We refer to the page numbers appended to Exhibit 1038, rather than the
page numbers included in certain portions of the original text of the Exhibit.
IPR2019-00620
Patent 9,441,956 B2

13

into seven condition categories,” e.g., the range of 85–100 corresponds to
“Excellent” condition, and the range of 70–85 corresponds to “Very Good”
condition. Id. at 24.
To calculate the condition indexes, a “weighted deduct-density
model” is used to “translate track inspection information . . . to condition
indexes.” Id. at 33. In particular, the weighted deduct-density model
calculates an index “by summing the appropriate individual component
group distress types over their applicable severity and density levels through
the use of appropriate weighting factors.” Id. at 34; see id. Fig. 7.
iv. Velten (Ex. 1041)
Velten is titled “Application of a Brightness-Adapted Edge Detector
for Real-Time Railroad Tie Detection in Video Frames.” Ex. 1041, 6.9
Velten describes “real-time image processing” for railroad “tie detection for
automated railway inspection.” Id. In particular, Velten describes “the
application of an edge detection algorithm” for “[a]utomated fault
detection . . . limited to the detection of splits or full breaks which look like
wider splits” on railroad ties. Id.
v. Kanade (Ex. 1010)
Kanade is titled “Three-Dimensional Machine Vision.” Ex. 1010, 2.10
Kanade describes vision systems “developed for an automatic inspection
task,” in particular, “a 3D assembly robot vision system.” Id. at 546, 557.
Kanade describes that “two-dimensional shape-recognition techniques can

9 We refer to the page numbers appended to Exhibit 1041, rather than the
page numbers included in certain portions of the original text of the Exhibit.
10 We refer to the page numbers appended to Exhibit 1010, rather than the
page numbers included in certain portions of the original text of the Exhibit.
IPR2019-00620
Patent 9,441,956 B2

14

be applied to recognize [a] part and to detect its rotation angle” in 3D space.
Id. at 562; see id. at 558.
E. Prior Art Status of the References
A challenge in an inter partes review can be raised “only on the basis
of prior art consisting patents or printed publications.” 35 U.S.C. § 311(b).
To qualify as a printed publication, “a reference ‘must have been sufficiently
accessible to the public interested in the art.’” Blue Calypso, LLC v.
Groupon, Inc., 815 F.3d 1331, 1348 (Fed. Cir. 2016) (quoting In re Cronyn,
890 F.2d 1158, 1160 (Fed. Cir. 1989)). “Because there are many ways in
which a reference may be disseminated to the interested public, ‘public
accessibility’ has been called the touchstone in determining whether a
reference constitutes a ‘printed publication’” under 35 U.S.C. § 102. Id.
(quoting In re Hall, 781 F.2d 897, 898–99 (Fed. Cir. 1986)). “A reference
will be considered publicly accessible if it was ‘disseminated or otherwise
made available to the extent that persons interested and ordinarily skilled in
the subject matter or art exercising reasonable diligence, can locate it.’” Id.
(quoting Kyocera Wireless Corp. v. Int’l Trade Comm’n, 545 F.3d 1340,
1350 (Fed. Cir. 2008)).
The Board recently clarified in a precedential opinion that “the
petition must identify, with particularity, evidence sufficient to establish a
reasonable likelihood that the reference was publicly accessible before the
critical date of the challenged patent and therefore that there is a reasonable
likelihood that it qualifies as a printed publication,” at the institution stage.
Hulu, LLC v. Sound View Innovations, LLC, IPR2018-01039, Paper 29 at 13
(PTAB December 20, 2019) (precedential). The Petition asserted that “the
earliest purported effective filing date of challenged claims 1–3, 7–13,
IPR2019-00620
Patent 9,441,956 B2

15

and 16–20 is March 12, 2013” and that all the relied-upon references are
prior art. Pet. 20, 5–7. We determined, at the institution stage, that
Petitioner had made the requisite showing. Dec. 10–14. The same evidence
from Petitioner regarding public accessibility is before us on this complete
record. See, e.g., Exs. 1010, 1038, 1041, 1045.
Patent Owner contends that “Petitioner fails to show that Uzarski,
Velten, and Kanade are prior art,” and, accordingly, Petitioner has failed to
“establish that its publications are prior art printed publications.” PO
Resp. 9. Patent Owner addresses each of these references in turn in its
arguments (id. at 9–17), and we do the same, below.
i. Uzarski
Uzarski, titled “Development of Condition Indexes for Low Volume
Railroad Track,” describes the work of the author and the U.S. Army Corps
of Engineers Construction Engineering Research Laboratories in developing
these indexes. Ex. 1038, 4. Petitioner submits, with Uzarski, a “Declaration
of Rachel J. Watters on Authentication of Publication.” Id. at 1. Ms.
Watters, a librarian at the University of Wisconsin library system, provides
an affidavit stating that, based on her knowledge, Uzarski “was received by
Wendt Library, University of Wisconsin-Madison Libraries on February 11,
1994,” and “catalogued and available to library patrons within a few days or
at most 2 to 3 weeks after February 11, 1994.” Id. at 2.
Patent Owner argues that the declaration of Ms. Watters “does not
provide sufficient evidence that (1) Uzarski has been disseminated or (2) that
Uzarski has been otherwise made available” (PO Resp. 10–11) because (1) it
contains “mere speculation, not evidence, regarding library procedures
during the relevant period Uzarski was allegedly published” (id. at 11); (2) it
IPR2019-00620
Patent 9,441,956 B2

16

“is devoid of any discussion of how a POSITA could locate Uzarski prior to
the critical date by exercising reasonable diligence” (id.); and (3) it “fails to
address whether any member of the public ever accessed [] Uzarski before
the critical date” (id. at 13).
Petitioner replies that Ms. Watters’s statements concerning library
procedures are not speculative, but “based on her personal experience and
knowledge as a University of Wisconsin librarian.” Reply 1. Petitioner
further replies that Patent Owner’s cited cases, such as In re Cronyn, 890
F.2d 1158 (Fed. Cir. 1989), fail to support Patent Owner’s assertion that
Uzarski is not prior art, because unlike Uzarski, the relevant prior art in
Cronyn was an “undergraduate thesis organized by student names and not
generally indexed or cataloged.” Reply 2–3.
As far as indicia on its face, Uzarski has a date of “July 1993”
(Ex. 1038, 3, 4), a date stamp of December 7, 1993 (id. at 4), an ISBN
Number of 89075655969A (id. at 3), a library date stamp stating “UW-
Madison Wendt Library Feb 11 1994 U.S. Depository Copy Do Not
Discard” (id.), and a Report Documentation Page, Form OMB No. 0704-
0188, with a July 1993 “Report Date” (id. at 8). The Report Documentation
Page further provides in its “Supplementary Notes” that “Copies are
available from the National Technical Information Service, 5285 Port Royal
Road, Springfield, VA 22161,” and in its “Distribution/Availability
Statement” that it is “Approved for public release; distribution is unlimited.”
Id. These statements emphasize the lack of restrictions on accessibility of
Uzarski. Id. Our findings regarding Uzarski’s facial indicia align with the
guidance provided in Hulu, in which a date stamp and a declaration
contributed to the totality of evidence relevant to public availability. See
IPR2019-00620
Patent 9,441,956 B2

17

Hulu, Paper 29 at 9–11, 17–19 (discussing the case-by-case analysis turning
on particular circumstances, including the nature of the reference
(distinguishing a thesis from a journal article) and the indicia of public
accessibility appearing on the face of the reference). Significantly, the facts
here include “the indicia on the face of a reference, such as printed dates and
stamps,” that are considered as part of the totality of the evidence. Id. at 17.
Additionally, by providing the declaration of Ms. Watters, Petitioner
has taken further steps to provide testimony that Uzarski was catalogued,
shelved, and indexed before the critical date, and made available such that
persons of ordinary skill in the art, exercising reasonable diligence, could
have located it. Id. at 1–2. Ms. Watters, a librarian with the University of
Wisconsin-Madison since 1998, provides details regarding her professional
background leading to her current position as Director of Wisconsin
TechSearch within the University of Wisconsin-Madison. Ex. 1038, 1. She
testifies that, based on her knowledge, Uzarski “was received by Wendt
Library, University of Wisconsin-Madison Libraries on February 11, 1994.”
Id. at 2. She further testifies that, based on the information in Uzarski, “it is
clear that the microfiche was received by the library on or before
February 11, 1994, catalogued and available to library patrons within a few
days or at most 2 to 3 weeks after February 11, 1994.” Id. The declaration
concludes with an affidavit that Ms. Watters is aware that false statements
are punishable by fine or imprisonment or both. Id.; see 37 C.F.R. § 1.68;
28 U.S.C. § 1746. We have reviewed the testimony of Ms. Watters
supporting Uzarski’s public accessibility as of the critical date, and find that
Ms. Watters provides credible testimony regarding the receipt of Uzarski by
the University of Wisconsin-Madison Library and the cataloging and
IPR2019-00620
Patent 9,441,956 B2

18

availability of Uzarski. On this complete record, we are persuaded that the
indicia on Uzarski and the support provided by the Watters declaration
demonstrate sufficiently that Uzarski was publicly accessible as of the
critical date.
ii. Velten
Velten is an excerpt from the proceedings of the 1999 IEEE
International Conference on Acoustics, Speech, and Signal Processing.
Ex. 1041, 4. Petitioner submits, with Velten, a Declaration of Pamela
Stansbury, an employee of the Cornell University Library since 1982. Id.
at 1. Ms. Stansbury states that Velten was “cataloged and publicly available
at the Engineering Library, in its public stacks, shelved by its Library of
Congress classification system number no later than May 26, 1999” based
on “the date stamp inside the volume, and my knowledge of the Library’s
standard procedures relevant to date stamping, cataloging, and shelving
items.” Id.
Patent Owner argues that the “only evidence in the record that Velten
was publicly accessible prior to the critical date” is the declaration of
Ms. Stansbury, which Patent Owner criticizes as “containing conclusory
opinions.” PO Resp. 13–14. Patent Owner also argues that Ms. Stansbury
never explains the standard procedures referenced in her declaration, nor
how they support her opinions. Id. at 14. Patent Owner further argues that
Petitioner “failed to establish sufficient details as to how one of skill in the
art could have located Velten with a reasonable search” and that
“Ms. Stansbury’s declaration does not address whether any member of the
public ever accessed the book before the critical date.” Id. at 16.
IPR2019-00620
Patent 9,441,956 B2

19

As far as indicia on its face, Velten has a date stamp stating
“Engineering Library May 26 1999 Cornell University” (Ex. 1041, 3), a
legend stating that the conference took place on March 15–19, 1999 (id.
at 4), an ISBN Number of 3 1924 085 251 076 (id. at 3), a Library Catalog
Number (id. at 5), extensive publication data including an IEEE Catalog
Number and several additional ISBN Numbers (id.), a statement regarding
“Copyright and Reprint Permission: Abstracting is permitted with credit to
the source,” along with information about obtaining other copying, reprint,
or republication from the IEEE Copyrights Manager (id.), and a copyright
date of 1999 (id. at 5, 6). See Hulu, Paper 29 at 9–11, 17–19.
Additionally, Petitioner has taken further steps to provide testimony
that Velten was catalogued, shelved, and indexed before the critical date,
and made available such that persons of ordinary skill in the art, exercising
reasonable diligence, could have located it. Id. at 1–2. Ms. Stansbury, a
librarian since 1982 with the Cornell University Library, provides details
regarding her professional background leading to her current position as
Administrative Supervisor in the Original Cataloging Unit. Ex. 1041, 1.
She testifies that, based on her knowledge, Velten was “cataloged and
publicly available at the Engineering Library, in its public stacks, shelved by
its Library of Congress classification system number no later than
May 26, 1999.” Id. She further testifies that “based on the date stamp inside
the volume, and my knowledge of the Library’s standard procedures relevant
to date stamping, cataloging, and shelving items,” she has not seen any
indication that Velten would have been an exception to the Library’s
standard practice, and that “any member of the public could have located
this item by searching our catalog by subject matter, author, or title.” Id.
IPR2019-00620
Patent 9,441,956 B2

20

The declaration concludes with an affidavit that Ms. Stansbury is aware that
false statements are punishable by fine or imprisonment, or both. Id.;
37 C.F.R. § 1.68; 28 U.S.C. § 1746. In our evaluation, Ms. Stansbury
provides credible testimony regarding the receipt of Velten by the Cornell
University Library and the cataloging and availability of Velten.
Patent Owner relies on a number of cases to support its argument that
multiple PTAB panels “have rejected declarations by Ms. Stansbury with
substantially similar wording to Exhibit 1041 for containing conclusory
opinions.” PO Resp. 14. We have reviewed those cases as well as Ms.
Stansbury’s declaration in this proceeding, and find this proceeding’s
declaration more thorough than the other declarations and sufficient in its
own right. See Reply 3 (“Ms. Stansbury’s declaration includes additional
statements not present in the referenced declarations” in Patent Owner’s
cited cases). Moreover, we agree that the standard is public accessibility,
not access, and thus Petitioner is not required to show that any member of
the public actually accessed the reference, only that a member of the public
could have accessed it. Reply 4–5; Samsung Elecs. Co. v. Infobridge Pte.
Ltd., 929 F.3d 1363, 1374 (Fed. Cir. 2019). We, therefore, credit the
testimony of Ms. Stansbury supporting Velten’s public accessibility as of the
critical date. On this complete record, we are persuaded that the indicia on
Velten, and the support provided by the Stansbury declaration, demonstrate
sufficiently that Velten was publicly accessible as of the critical date.
iii. Kanade
Kanade is a book titled “Three-Dimensional Machine Vision,”
published by Kluwer Academic Publishers. Ex. 1010, 2. Petitioner submits,
with Kanade, a Declaration of Rosalina Delgado-Jones, Assistant Business
IPR2019-00620
Patent 9,441,956 B2

21

Enterprises Officer in the Office of Business Enterprises of the Library of
Congress. Id. at 1. Ms. Delgado-Jones certifies that Kanade is in the
collections of the Library of Congress, and further certifies that Kanade “is
marked with a Library of Congress Cataloging-In-Publication stamp that
bears the date March 23, 1987.” Id. The seal of the Library of Congress is
affixed to Ms. Delgado-Jones’s declaration. Id. Petitioner also submits a
declaration of Ms. Pamela Stansbury, substantially similar to Ms.
Stansbury’s declaration for Velten in this proceeding, stating that Kanade
was “cataloged and publicly available at the Engineering Library, in its
public stacks, shelved by its Library of Congress classification system
number no later than November 11, 1987” based on “the date stamp inside
the volume, and my knowledge of the Library’s standard procedures relevant
to date stamping, cataloging, and shelving items.” Ex. 1045, 1.
Patent Owner argues that the declaration of Ms. Stansbury contains
“the same fatal errors as Stansbury Ex. 1041 discussed above.” PO
Resp. 16. Patent Owner faults the declaration of Ms. Delgado-Jones for not
discussing “the entire Kanade article.” Id.
Kanade, on its face, has a copyright date of “1987 by Kluwer
Academic Publishers” (Ex. 1010, 7), a stamp stating “Library of Congress 5
Mar 23 1987 Copy __ CIP” (id.), publication data (id.), a Library of
Congress library stamp (id. at 3), a Library Catalog Number (id. at 7), and
additional cataloging notations with the date 1987 (id.). There is also a
check-out card indicating that Kanade was in fact checked out on multiple
occasions between 1987 and 2002 (Ex. 1045, 8). See Hulu, Paper 29 at 9–
11, 17–19.
IPR2019-00620
Patent 9,441,956 B2

22

Additionally, Petitioner has taken further steps to provide testimony
that Kanade was catalogued, shelved, and indexed before the critical date,
and made available such that persons of ordinary skill in the art, exercising
reasonable diligence, could have located it. Ex. 1010, 1; Ex. 1045. The
Stansbury declaration for Kanade is substantially similar to the Stansbury
declaration for Velten. Compare Ex. 1045, with Ex. 1041, 1–2. For the
same reasons given above, which address Patent Owner’s objections to the
Stansbury declarations regarding Velten, we determine that Ms. Stansbury
provides credible testimony regarding the receipt of Kanade by the Cornell
University Library and the cataloging and availability of Kanade. Ms.
Delgado-Jones certifies that the Library of Congress contains a copy of
Kanade, and the reproduced pages in the record “are a true representation
from that work.” Ex. 1010, 1. Ms. Delgado-Jones further testifies that the
work is marked with a Library of Congress stamp bearing the date
March 23, 1987. Id. The testimony of Ms. Stansbury and of Ms. Delgado-
Jones credibly supports Petitioner’s assertions of Kanade’s public
accessibility as of the critical date. Id. We are persuaded that the indicia on
Kanade itself, and the support provided by the Stansbury and Delgado-Jones
declarations, demonstrate sufficiently that Kanade was publicly accessible as
of the critical date.
Therefore, on this complete record, Petitioner provides evidence
establishing the public accessibility of Uzarski, Velten, and Kanade as of a
date prior to the critical date.
IPR2019-00620
Patent 9,441,956 B2

23

F. Asserted Obviousness Over Villar, Haas, and Uzarski (Ground 1)
Petitioner asserts that claims 1–3, 7, 8, 11–13, and 16–20 of the ’956
patent are unpatentable as obvious over Villar, Haas, and Uzarski under 35
U.S.C. § 103. Pet. 22–56.
i. Claim 1 - Petitioner’s Arguments
Petitioner alleges that the combination of Villar, Haas, and Uzarski
teaches:
1[pre] “A system for inspecting railroad ties, comprising:” (id. at 22
(citing Ex. 1036, 3:22–28, 10:34–36, 11:34–36, 12:18–20, Figs. 1, 2));
[a] “a. a rail vehicle, configured for moving along a railroad track in a
travel direction;” (id. at 23 (citing Ex. 1036, 3:53–61, 11:40–42));
[b] “b. a light generator, attached to the rail vehicle, oriented to
project a beam of light across the railroad track while moving thereon;” (id.
(citing Ex. 1036, 3:29–30, 3:44–49, 3:53–57, 4:9–32, 10:37–39, 11:44–46));
[c] “c. an optical receiver, attached to the rail vehicle, oriented to
receive at least a portion of the beam of light reflected from the railroad
track and configured to generate image data representative of a profile of at
least a portion of the railroad track; and” (id. at 23–24 (citing Ex. 1036,
3:30–32, 10:40–44, 3:53–57, 3:29–32, 4:9–14, 4:43–52, 5:36–39, 4:55–58,
5:36–39));
[d] “d. a processor, configured to analyze the image data by applying
one or more algorithms configured to:” (id. at 25 (citing Ex. 1036, 3:33,
4:65–67, 5:7–14, 8:13–24, 8:61–9:4, 5:15–31));
[e] “i. find boundaries of at least one railroad tie in the railroad track;
and” (id. at 25–29 (citing Ex. 1036, passim)); and
IPR2019-00620
Patent 9,441,956 B2

24

[f] “ii. determine the condition of the at least one railroad tie by:
a. determining a series of condition metrics of the at least one railroad tie;
b. applying a weight factor to each condition metric; and c. computing a
condition grade of the at least one railroad tie based on the weighted
condition metrics” (id. at 29–38 (citing Ex. 1037 ¶¶ 2, 25, 27, 29, 36, 37, 41,
Figs. 1–7; Ex. 1038, 16, 18–19, 31–36, 40–41, 56–57)).
Petitioner argues that Villar discloses every limitation except
“determining the condition of the railroad tie by: (1) determining a series of
condition metrics of the at least one railroad tie; (2) applying a weight factor
to each condition metric; and (3) computing a condition grade of the at least
one railroad tie based on the weighted condition metrics.” Pet. 30.
However, according to Dr. Papanikolopoulos, “it was well-known to one of
ordinary skill in the art at the relevant time to determine the condition of
railroad ties in this manner, as evidenced, e.g., by Haas (Ex-1037) and
Uzarski (Ex-1038).” Id. (citing Ex. 1004 ¶ 58).
Petitioner states that a person having ordinary skill in the art would
have been motivated to combine Villar, Haas, and Uzarski
because Uzarski provides an ‘unbiased and reputable’ TCI [tie
condition index] assessment method to help railroad track
maintenance managers routinely (and uniformly) perform tasks,
such as assessing current track conditions, predicting future track
conditions, establishing track deterioration rates, determining
and prioritizing current and long range M&R [maintenance and
repair] reports, formulating budgets, and measuring the
effectiveness of M&R, as taught by Uzarski, 16. And, one of
ordinary skill in the art would have known to look to Uzarski’s
TCI assessment method because it is a specific example of
government regulations referenced by Haas as providing
pertinent information related to assessing the condition of rail
components.
IPR2019-00620
Patent 9,441,956 B2

25

Pet. 37 (citing Ex. 1004 ¶ 70).
ii. Claim 1 - Patent Owner’s Arguments
a. Manual and Machine Vision
Patent Owner argues that one of ordinary skill in the art would not
have combined a reference concerning a manual method with a reference
concerning machine vision. PO Resp. 17 (citing Ex. 2001 ¶¶ 37–39). More
particularly, Patent Owner argues, one of ordinary skill in the art would not
have combined Uzarski’s manual method with Villar’s and/or Haas’s
machine vision systems. Id. at 19–25 (citing Ex. 2001 ¶¶ 40–46).
According to Dr. Bovik, there is a “vast chasm between a manual method
and an automated method especially where, as here, the automated method
uses machine vision,” and “it is not trivial or obvious to write an algorithm
to implement a manual method.” Id. at 22 (citing Ex. 2001 ¶ 43). Patent
Owner further argues that one of ordinary skill in the art “would not have
combined the TCI indexes and weighted model of Uzarski with the
automated machine vision inspection system of Villar because the
underlying basis is the subjective judgement—the grading—by human
inspectors.” Id. at 23 (citing Ex. 2001 ¶45). The Petition also “provides no
proof that Uzarski’s manual inspection methodology is capable of being
combined with the automated inspection systems in Villar and/or Haas,”
according to Patent Owner. PO Resp. 24. More particularly, Patent Owner
argues that the “Petition fails to show how the visual inspection method
using track inspectors disclosed in Uzarski is capable of being physically
integrated into the automated inspection systems of Villar and/or Haas.” Id.
Petitioner disputes that it would be generally “difficult, time-
consuming, or require special expertise” to implement the method of Uzarski
IPR2019-00620
Patent 9,441,956 B2

26

into the system of Villar. Reply 8. Petitioner also argues that the Board has
discretion to “ignore conclusory opinions without scrutiny.” Id. Petitioner
asserts that Dr. Bovik’s testimony is not only conclusory but also incorrect
in view of evidence demonstrating that automating manual inspection tasks
was known in the art even before 2004. Id. (citing, inter alia, Ex. 1010,
545–546, 73, 107, 248–250, 361, 406; Ex. 1007, 5; Ex. 1012, 15; Ex. 1014,
18; Ex. 1018, 9; Ex. 1026, 2; Ex. 1019, 6; Ex. 1021, 3; Ex. 1047, 4; Ex.
1036, 1:66–2:6).
Having reviewed the evidence and testimony before us, we determine
that it would have been within the level of ordinary skill in the art to
automate the task of inspecting tie condition. For one, Villar itself
“automates manual inspection tasks.” Id. (citing Ex. 1036, 1:66–2:6).
Petitioner provides evidence that Villar and others were using known
software and other tools to automate defect-measuring tasks in the art. Id.
at 9–10 (citing Pet. 14–15; Ex. 1036, 1:66–2:6; Ex. 1019, 3, 6; Ex. 1023, 16–
19; Ex. 1012, 19–20, Ex. 1004 ¶¶ 32–33). Petitioner’s expert also attests,
with supporting evidence, that the challenged patent as well as “other sheet-
of-light range imaging sensor technology would have been well-known and
readily available to a person of ordinary skill in the art by 2004.” Ex. 1004
¶¶ 32–33; Pet. 14–15.
Even taking into account Patent Owner’s allegations of difficulty,
amount of time required, or special expertise required, which are based
largely on Dr. Bovik’s assertions but little other evidence (see, e.g., Ex. 2001
¶¶ 37–46), Petitioner has provided sufficient evidence and testimony to
demonstrate that it would not be beyond the skill of one in the art to combine
the art as proposed and accomplish the tasks described in the challenged
IPR2019-00620
Patent 9,441,956 B2

27

claims. Reply 9–10; see also Medichem, S.A. v. Rolabo, S.L., 437 F.3d
1157, 1165 (Fed. Cir. 2006) (a “given course of action often has
simultaneous advantages and disadvantages, and this does not necessarily
obviate motivation to combine”).
Petitioner also argues that Patent Owner “provides no credible reason
why a POSITA would have been unable to implement Villar’s automated
vision system with Uzarski’s tie condition index (‘TCI’ []) weighted deduct-
density model” in Uzarski’s Equation 2, “including the underlying
subjective ratings used to define and determine the distress types (T),
severity (S), and density (D) levels, and deduct weighting values.”
Reply 10. As interpreted by Petitioner, Uzarski’s TCI index incorporates
“subjective rating opinions of various track problems” to provide a
meaningful “unbiased and reputable (i.e. objective)” crosstie condition
assessment model. Id. at 11. The language of Uzarski relied upon by
Petitioner indicates that, once in Equation 2 form, Uzarski’s model is able to
be applied objectively to assess railroad track conditions. Id. at 15
(“Uzarski’s approach is like GREX’s Aurora system, which implements
objective tie grading models based on the subjective tie grading attributes
and concerns provided by its customers” (citing Ex. 1054, 18; Ex. 1053, 54–
55; Ex. 1052, 38–39; Ex. 1051, 2–3; Ex. 1048, 44:5–19, 53:7–54:10, 43:22–
44:2, 79:17–82:6)). Based on Petitioner’s arguments and evidence, we find
it is both reasonable and persuasive that one of ordinary skill would have
been able to program a system such as Villar’s to apply an objective model
like Uzarski’s.
Petitioner also replies that the “Petition does not propose bodily
incorporating Uzarski’s visual manual inspection of crossties to Villar,” but
IPR2019-00620
Patent 9,441,956 B2

28

rather, that “the combined teachings of Villar and Uzarski would have
suggested to a POSITA to implement Villar’s automated inspection system
with a tie condition assessment method, as taught by Uzarski.” Reply 13
(citing Pet. 37–38 (citing Ex. 1038, 16, 18–19)).
Bodily incorporation is not the standard for an obviousness analysis.
See In re Mouttet, 686 F.3d 1322, 1332–33 (Fed. Cir. 2012); see also MCM
Portfolio LLC v. Hewlett-Packard Co., 812 F.3d 1284, 1294 (Fed. Cir. 2015)
(the “test for obviousness is not whether the features of a secondary
reference may be bodily incorporated into the structure of the primary
reference” (citation omitted)). Petitioner’s proposed combination relies on
Uzarski for “government regulations related to the development of condition
indexes for railroad track components, including crossties,” and for its
assessment method including “(1) determining a series of condition metrics
for crossties; (2) applying a weight factor to each condition metric; and (3)
computing a condition grade for the railroad tie based on the weighted
condition metrics.” Pet. 33–34. Petitioner’s proposed combination does not,
in our view, require the ordinarily skilled artisan to bodily incorporate the
various systems and processes of Uzarski into Villar.
b. Teaching Away
Patent Owner argues that Villar teaches away from combination with
Uzarski. PO Resp. 25–28. Villar, Patent Owner argues, considers manual
inspection problematic, because the “process of visual inspection is very
time consuming” (id. at 25 (citing Ex. 1036, 1:45–46), and to “overcome the
problem of visual inspection, Villar discloses a laser/camera line optical
scanning and analysis system that can automatically identify tie defects and
grade the ties” (id. at 26 (citing Ex. 1036, 2:14–31)). This contrasts with
IPR2019-00620
Patent 9,441,956 B2

29

Uzarski’s “manual visual inspection method” (id. (citing Ex. 1038, 4, 17, 42,
88). Patent Owner argues that the automatic inspection system of Villar was
invented “to eliminate the need for manual visual inspection, like that
employed in Uzarski.” Id. at 26 (citing Ex. 2001 ¶ 49).
Petitioner replies that “Villar does not teach away from implementing
a tie condition assessment method,” but “rather encourages it.” Reply 12–
13. Regarding Patent Owner’s argument that Villar was intended to
“eliminate the need for manual visual inspection” (id. at 13 (citing PO
Resp. 26)), Petitioner argues that, instead, “Villar appreciates the need to
regularly inspect and grade track components,” and that “Villar concedes
that automating manual inspection tasks in 2004 was well-known.” Id.
at 13–14 (citing Ex. 1036, 1:45–52, 1:66–2:9). Petitioner also notes that
Uzarski itself “appreciates using both manual and automatic inspection
methods and states that its indexes ‘may work very well for trackage
subjected to high traffic volumes and speed,’ where automated systems were
commonly used.” Id. at 14 (citing Ex. 1038, 18–19, 30).
A reference does not teach away “if it merely expresses a general
preference for an alternative invention but does not ‘criticize, discredit, or
otherwise discourage’ investigation into the invention claimed.’” Polaris
Indus., Inc. v. Artic Cat, Inc., 882 F.3d 1056, 1069 (Fed. Cir. 2018) (quoting
In re Fulton, 391 F.3d 1195, 1201 (Fed. Cir. 2004) (a reference teaches
away if it criticizes, discredits, or otherwise discourages the claimed
solution)). We are not directed to any portion of Villar that discredits or
discourages incorporation of a stepwise assessment method such as the one
disclosed in Uzarski. As Petitioner notes, Villar encourages implementation
of tie condition assessment methods and automating manual inspection
IPR2019-00620
Patent 9,441,956 B2

30

tasks, and Uzarski itself acknowledges the existence of both manual and
automatic inspection methods, stating that its indexes “may work very well
for trackage subjected to high traffic volumes and speed,” where automated
systems were commonly used. Reply 14 (citing Ex. 1038, 18–19, 30). We
agree that using Villar’s inspection system to grade the inspected crossties’
condition, as taught by Uzarski, would “simply be an extension of Villar’s
own teachings,” and that Villar does not teach away. Id.
c. Motivation to Combine
Patent Owner also argues that the “Petition fails to demonstrate that a
POSITA would have a sufficient reason to combine Villar with the other
references to achieve the claimed inventions with a reasonable expectation
of success.” PO Resp. 28–33. More particularly, Patent Owner faults
Petitioner for not providing a motivation to combine Haas with Villar (id. at
28–30) and for not providing a reason to combine Villar with Haas and
Uzarski (id. at 30–33).
Regarding Patent Owner’s arguments directed to motivation to
combine Haas and Villar, we understand that Petitioner’s position is based
on a combination of three references—Villar, Haas, and Uzarski—rather
than just two references—Haas and Villar. Thus, Patent Owner’s arguments
directed to motivation to combine only two of these three references do not
consider the proposed combination of all three references. Because
Petitioner does not, as Patent Owner argues, “suggest[] that the combination
of Villar and Haas alone could possibly create one of the claimed
combinations” (PO Resp. 29), the alleged absence of reasons to combine
Haas alone with Villar alone is not critical to our analysis.
IPR2019-00620
Patent 9,441,956 B2

31

Regarding Patent Owner’s arguments directed to the combination of
Villar, Haas, and Uzarski, Patent Owner argues that there is no teaching by
Uzarski that its “condition indexes would be ‘unbiased and repeatable’ for
automated machine vision,” and thus, one of ordinary skill in the art “would
not have expected that Uzarski’s methodology was capable of being
implemented into Villar’s machine vision system.” PO Resp. 32 (citing
Ex. 2001 ¶ 46). Patent Owner also argues that one of ordinary skill in the art
would not combine Villar with Uzarski and Haas because Uzarski does not
teach or suggest “that any of its contents are ‘government regulations’” and,
thus, Uzarski is not “a specific example of government regulations
referenced by Haas” as argued by Petitioner. Id. at 32–33 (citing Ex. 2001
¶ 55).
Petitioner reiterates that “it would have been obvious to a POSITA at
the relevant time to implement Villar’s automated track inspection machine
vision system’s processing device to determine the condition grade of ties
via a tie condition assessment method, as taught by Uzarski” because
“Uzarski’s ‘unbiased and reputable’ tie condition assessment method would
help track maintenance managers routinely (and uniformly) assess current
and predicting future track conditions, establish track deterioration rates, and
determine and prioritize current and long-range maintenance and repair
reports.” Reply 6–7 (citing Pet. 37–38). Petitioner also reiterates that
Uzarski “provides a specific example of government regulations referenced
by Haas,” and Uzarski “further discusses visual and automated track
inspection systems used by the railroad industry—like Villar’s and Haas’s
systems.” Id. at 7 (citing Pet. 37–38).
IPR2019-00620
Patent 9,441,956 B2

32

Regarding Patent Owner’s argument that Uzarski is not itself a
“government regulation” (PO Resp. 32–33), Petitioner argues that
“Uzarski’s tie condition assessment method” is a “specific example of
government regulations referenced by Haas.” Reply 21. Petitioner also
argues that both Uzarski and Haas recognize that their “assessment methods
are suitable for application to both mission readiness (military) track
assessment tasks and for profit (commercial) by individual railroad
companies to maintain track health.” Id. at 21, 23. We are persuaded that
Uzarski’s disclosure of government-specific indexes, and its discussion of
the development of these indexes by U.S. Army Construction Engineering
Research Laboratories in cooperation with the University of Illinois
(Ex. 1038, 9), together with Haas’s reference to “government regulations,”
sufficiently underpins Petitioner’s reasons to combine the two. Pet. 33–38
(citing Ex. 1038, 9, 16–19, 31–36, 40–41, 56–57; Ex. 1037 ¶¶ 2, 29, 37;
Ex. 1004 ¶ 70); see also Ex. 1038, 8, Abstract (stating that Uzarski describes
development of the indexes and a “separate, follow-on report (USA ERL
Technical Report FM-93/14, July 1993) provides track inspectors with a
standard reference for identifying track dis-tresses and calculating condition
indexes”). Whether or not Uzarski is a government regulation per se is not
fatal to the proposed combination, when the government regulations
discussed and disclosed therein are known to one of ordinary skill in the art,
and when Uzarski provides other reasons to combine with Villar and Haas.
See, e.g., Tr. 15:7–11 (stating that “even if Uzarski wasn’t a codified
regulation it’s still the type of report and guidelines that if you’re looking
through Haas and Haas – that you would consider looking at if you were
concerned about inspecting rail components.”). Even so, Villar, Haas, and
IPR2019-00620
Patent 9,441,956 B2

33

Uzarski are ultimately combinable as part of Petitioner’s three-reference
combination for reasons not having to do solely with “government
regulations.” Pet. 37–38; Tr. 16:5–8 (“without even getting into whether it’s
a regulation or not, one would have looked to Uzarski because he is
providing information about why you would want to adopt his tie condition
grading index”).
Petitioner and its declarant provide ample reasons that one of ordinary
skill in the art would have been motivated to employ Villar’s processor to
use Uzarski’s method, and ample reasons that one of ordinary skill in the art
also would have known to look to Haas to determine what was “well-known
to one of ordinary skill in the art at the relevant time to determine the
condition of railroad ties,” and to identify and use the particular regulations
referenced in Uzarski. Pet. 30 (citing Ex. 1004 ¶ 58), 37–38 (citing
Ex. 1004 ¶ 70); see also Reply 6–12; Ex. 1004 ¶¶ 70–71; Ex. 1038, 18–20.
On this complete record, and for the reasons provided by Petitioner and
discussed above, we determine that Petitioner provides adequate reasoning
to combine the three references in the manner proposed in the Petition with a
reasonable expectation of success.
d. Elements [d] and [f]
Patent Owner argues that neither Uzarski, nor Haas, nor the
combination of Uzarski and Haas disclose elements [d] and [f] of claim 1.
PO Resp. 33–48. More particularly, Patent Owner argues that Uzarski “does
not disclose a processor configured in the manner mandated by elements [d]
and [f]” of claim 1, for three reasons: (1) Uzarski does not disclose
“determining a series of condition metrics of the at least one railroad tie” as
recited in element 1[f] (id. at 34–35); (2) Uzarski does not employ a
IPR2019-00620
Patent 9,441,956 B2

34

“processor” as recited in element 1[d] to implement its system of grading
ties (id. at 38); and (3) Uzarski does not disclose at least “applying a weight
factor to each condition metric” recited in element 1[f] (id. at 40). Finally,
Patent Owner argues that Haas does not “disclose a processor configured in
the manner mandated by elements [d] and [f]” (id. at 41), and that the
combination of Haas and Uzarski does not disclose or suggest elements [d]
and [f] (id. at 42–48).
In response to Patent Owner’s first argument, Petitioner replies that
Uzarski does disclose “determining a series of condition metrics of the at
least one railroad tie” as recited in element 1[f], because claim 1 is directed
to “the at least one railroad tie,” and Uzarski discloses determining a
condition metric for at least one railroad tie. Reply 17–23. First, Petitioner
argues, Uzarski discloses claim 1’s “determining a series of condition
metrics of the at least one railroad tie,” because the plain language of claim 1
“does not preclude determining a series of condition metrics for multiple ties
(e.g., a series of condition metrics for a set of railroad ties).” Id. at 17.
Petitioner also argues that even if the claim were read to require “more than
one condition metric for each crosstie, Uzarski still discloses it” with its
determination of “a series of condition metrics—distress type, severity level,
and distress amount—for its tie condition index assessment method.” Id.
at 18 (citing Pet. 34–35; Ex. 1038, 33, 56–57).
Uzarski’s Equation 2 is reproduced below:
IPR2019-00620
Patent 9,441,956 B2

35

Equation 2 accounts for a distress type (T), a severity level (S), and distress
density (D), all subject to a deduct weighting value (a). Ex. 1038, 34.
Uzarski discloses distress types T1 through T8, with T1 indicating a “Single
Defective Tie” and T8 indicating an “Improperly Positioned Tie.” Id. at 56–
57. Both of these factors, by their plain language, concern the condition of
an individual railroad tie. Id.; see also Ex. 1049, 57:11–17 (stating that
“skew angle of a tie,” i.e. improper positioning of a tie, can be a factor when
determining the condition of a tie). Moreover, T1 can be further broken
down into severity levels low (L) and medium (M), and T8 can be broken
down into severity levels low (L), medium (M), and high (H), thereby
providing additional metrics relating to individual crossties. Ex. 1038, 57–
58. Thus, it appears from our review of Uzarski that “determining a series
of condition metrics of the at least one railroad tie” is disclosed, whether the
claim language is interpreted as determining a series of condition metrics for
one railroad tie or a series of conditions metrics for multiple railroad ties.
In response to Patent Owner’s second argument, Petitioner replies that
“attacking Uzarski individually and ignoring its combination with Villar and
Haas” misapplies the legal standard for obviousness when based on a
combination of references. Reply 8. We agree that Patent Owner’s
assertions regarding Uzarski alone, namely, that Uzarski does not employ a
IPR2019-00620
Patent 9,441,956 B2

36

“processor” as recited in element 1[d] to implement its system of grading
ties, are not tenable when Petitioner’s obviousness argument does not rely
on Uzarski for the processor limitation. Petitioner relies on a combination of
three references, and therefore, attacking Uzarski individually for not having
a processor is further untenable. See In re Keller, 642 F.2d 413, 426 (CCPA
1981); In re Merck & Co., 800 F.2d 1091, 1097 (Fed. Cir. 1986) (non-
obviousness cannot be established by attacking references individually
where the obviousness argument is based a combination of references).
In response to Patent Owner’s third argument, Petitioner argues that
Uzarski discloses claim 1’s “applying a weight factor to each condition
metric,” because the record and the claim’s plain language do not support
Patent Owner’s construction that “applying a weight factor to each condition
metric” requires applying a “different” weight factor to each “individual”
condition metric. Id. at 20. Although Patent Owner relies on one example
from the ’956 patent Specification in which “a unique weight factor” is
“applied to each variable” (PO Resp. 40–41; Ex. 1001, 21:10–17), this
example in the Specification is just “one embodiment” (id. at 21:10) and
“other mathematical scoring methods and formulas can also be used” (id.
at 21:19–20). See also Tr. 34:14–35:12 (“there’s nothing in the claim
language expressly prohibiting” the weight factor from being the same and
applied to each condition metric, and “nothing explicit like that in the
specification”). A non-limiting, exemplary embodiment in a specification
does not mean that the claim language should be read as importing and
requiring that one embodiment. Liebel-Flarsheim Co. v. Medrad, Inc., 358
F.3d 898, 906 (Fed. Cir. 2004) (only when “the patentee has demonstrated a
clear intention to limit the claim scope using ‘words or expressions of
IPR2019-00620
Patent 9,441,956 B2

37

manifest exclusion or restriction’” is the scope of a claim properly restricted
to that of a disclosed embodiment). Because we need not read the limitation
“applying a weight factor to each condition metric” as requiring application
of a different weight factor to each individual condition metric, we accept
Petitioner’s argument that Uzarski’s application of weight factor “a”
discloses this limitation of claim 1. Pet. 35 (Uzarski “includes applying a
weight factor (i.e., ‘a( )’) to each of the condition metrics—i.e., the distress
type (T), severity (S), and density or amount (D)”).
e. Additional Arguments
In response to Patent Owner’s final arguments, Petitioner argues that
Villar in combination with Haas and Uzarski render obvious claim
limitations 1[d] and 1[f], because the “Petition relies on Villar, not Haas, for
claim 1[d]’s processor, and Uzarski, not Haas,” for the condition metric.
Reply 21. We agree that Petitioner’s combination relies on Villar for the
processor limitation and Uzarski for the condition metric limitations.
Because Petitioner relies on Villar, rather than Uzarski or Haas, to disclose
the processor required in element 1[d] (Pet. 25), Patent Owner’s attempts to
discredit Petitioner’s argument with respect to element 1[d] by arguing that
Haas and Uzarski do not disclose a processor are misdirected. Regarding
element 1[f], we are persuaded by Petitioner’s mapping of the respective
elements of Villar, Uzarski, and Haas to the challenged claim limitations.
Id. at 29–38. Petitioner supports this mapping with extensive discussion of
the references themselves, as supported by the testimony of Dr.
Papanikolopoulos. Id. (citing Ex. 1004 ¶¶ 58–70). As we have found above,
Petitioner adequately demonstrates that Villar discloses the processor
limitation and Uzarski discloses the condition metric limitation, and
IPR2019-00620
Patent 9,441,956 B2

38

Petitioner also provides well-supported and persuasive reasoning for
combining the three relied-upon references in the manner proposed.
Accordingly, Patent Owner’s arguments regarding Haas’s alleged lack of
processor and regarding the alleged failure of a Haas-Uzarski combination to
disclose or suggest elements [d] and [f] are inapposite in view of the actual
arguments presented in the Petition.
On this record, we are persuaded by Petitioner’s mapping of the
disputed elements of claim 1 to the elements of Villar, Haas, and Uzarski,
and by Petitioner’s arguments supporting reasons to combine the three
references with a reasonable expectation of success.
iii. Claims 8 and 17
Petitioner argues that the combination of Villar, Haas, and Uzarski
renders obvious claim 8. Pet. 42–45. Specifically, regarding element [d] of
claim 8, which requires “computing a condition grade for the at least one
railroad tie based on a weighted summation of the condition metrics,”
Petitioner argues that, as “discussed in claim 1[f], Villar, Haas, and Uzarski
render obvious these limitations of claim 8, and thus also claim 8.” Id. at 45
(citing Ex. 1004 ¶ 90).
Petitioner further argues that the combination of Villar, Haas, and
Uzarski renders obvious claim 17. Pet. 50–54. Petitioner characterizes
element [c] of claim 17 as requiring “analyzing the 3-D profile via a
computer processor to determine a set of condition metrics representing each
railroad tie in the set of railroad ties,” and “computing a condition grade for
each railroad tie based on a weighted summation of the condition metrics.”
Id. at 53. Petitioner argues that “as discussed in claim 1[f], Villar in
combination with Haas and Uzarski does” disclose element 17[c], and
IPR2019-00620
Patent 9,441,956 B2

39

“Villar, Haas, and Uzarski thus render obvious these limitations of claim
17.” Id. (citing Ex. 1004 ¶ 106).
Patent Owner argues that the Petition is deficient regarding
element [d] of claim 8, because “the Petition and Dr. Papanikolopoulos rely
on its reasoning why Villar, Haas, and Uzarski purportedly disclose
element [f] of claim 1.” PO Resp. 49 (citing Ex. 2001 ¶ 102). Patent Owner
further argues that “the language of those two elements is not identical”
because “claim 8 recites ‘computing a condition grade for the at least one
railroad tie based on a weighted summation of the condition metrics’” and
claim 1 “does not recite ‘a weighted summation.’” Id. at 50.
Patent Owner similarly argues, regarding element [c] of claim 17, that
the Petition is deficient because it “relies entirely on its analysis of
element [f] of claim 1,” but “the language of those two elements is not
identical.” PO Resp. 55. Rather, Petitioner argues, claim 1 “does not recite
‘a weighted summation’” as in element 17[c], but instead recites “applying a
weight factor to each condition metric.” Id. at 55–56.
Petitioner replies that there is no improper incorporation by reference,
as argued by Patent Owner, because “Petitioner merely references
paragraphs from the same document, not external sources.” Reply 23.
Petitioner also argues that Patent Owner “provides no substantive
explanation or evidence supporting why the arguments presented with
respect to claims 8 and 17 framed from the analysis set for[th] with respect
to claim 1 are inapplicable.” Id.
We do not view Petitioner’s reference to its arguments regarding
claim 1[f] as improper incorporation by reference, because Petitioner is not
incorporating arguments or evidence from another document. See 37 C.F.R.
IPR2019-00620
Patent 9,441,956 B2

40

§ 42.6(a)(3) (“Arguments must not be incorporated by reference from one
document into another document.”). Furthermore, in reviewing Petitioner’s
arguments regarding claim 1[f], Petitioner appears to address the very
problem Patent Owner raises, namely, the issue of weighted summation.
Petitioner argues:
Uzarski’s TCI assessment method includes applying a weight
factor (i.e., “a( )”) to each of the condition metrics—i.e., the
distress type (T), severity (S), and density or amount (D)—
“summing the appropriate individual component group distress
types over their applicable severity and density levels through
the use of appropriate weighting factors.”

Pet. 35 (emphasis added). As shown in Uzarski’s Equation 2, relied upon
for Petitioner’s arguments concerning claim 1[f] as well as claim 8[d] and
claim 17[c], there appears to be at least one weighted summation (Ʃ as
applied to a( )). Ex. 1038, 34; see also Ex. 1004 ¶ 68. Although Petitioner
ideally would have included these arguments in the section of the Petition
addressing claim 8[d] and claim 17[c], Petitioner’s weighted summation
arguments are in the Petition and sufficiently apparent to support Petitioner’s
burden with respect to claim 8[d] and 17[c].
iv. Claim 12
Petitioner argues that the combination of Villar, Haas, and Uzarski
renders obvious claim 12. Pet. 46–48. Specifically, regarding the portion of
claim 12 that requires “creating a tie replacement plan that is based upon the
location of each railroad tie,” Petitioner argues that Haas discloses creating a
tie replacement plan based upon the location. Id. (citing Ex. 1004 ¶¶ 94–95;
Ex. 1037 ¶¶ 25, 27–29, 36, 38, 40–41, 43, Figs. 1–7).
IPR2019-00620
Patent 9,441,956 B2

41

Patent Owner argues Petitioner “fails to show a reasonable likelihood
that claim 12 is unpatentable over Villar, Haas, and Uzarski.” PO Resp. 51.
According to Patent Owner, Petitioner admits that “Villar does not disclose
creating a tie replacement plan based upon the location,” and “erroneously
contends that Haas ‘does’ disclose creating such a plan.” Id. Patent Owner
argues that what Haas does, i.e., “sending exemption information and
positioning data to a server for ‘maintenance planning purposes,’ is not even
vaguely the same as ‘creating a tie replacement plan’ as recited by claim 12.
Id. at 51–52. Patent Owner reiterates that “sending information to a service
is not the same as, and does not mean that, the information is being used to
create a tie replacement plan that is based upon the location of each railroad
tie.” Id. at 54 (citing Ex. 2001 ¶ 107).
Petitioner replies that Patent Owner
provides no credible reason why Haas’s disclosure of
determining at least one crosstie’s condition, severity, and
location pursuant to a tiered maintenance timeline for repairing
detected rail component problems such that crossties with more
severe condition grades are addressed/repaired “immediately,”
for safety, while those with acceptable condition grades can
alternatively be addressed, e.g., “soon,” “during the next
inspection,” or “during the next scheduled maintenance window
or for use during planning,” even if created based on information
sent to the server, would not be understood as a “tie replacement
plan.”

Reply 24 (citing Ex. 1037, Abstract, ¶¶ 27–29, 36, 40, Figs. 1–7).
We agree that Petitioner’s cited disclosures of Haas appear to disclose
a tie replacement plan. Claim 12 requires “creating a tie replacement plan
that is based upon the location of each railroad tie,” but the claim places no
further requirements to the creation of the tie replacement plan apart from
IPR2019-00620
Patent 9,441,956 B2

42

basing it upon the location of each railroad tie. Petitioner shows adequately
that Haas’s inspection system 100 includes a data acquisition module 102
that records positioning information 112 (shown in Figure 1 as a box labeled
“GPS”). Pet. 46–47. Petitioner also relies on Haas’s track condition
monitoring module 104 that uses “various image analytics 114 to identify
and detect the current condition of the rail components in the image and a
defect severity analysis and temporal condition analysis module 106 to
determine a severity of the detected problems in the rail components.” Id.
(citing Ex. 1037 ¶¶ 25, 29, 36, 41, Figs. 1–7). Haas then sends this
information, including “positioning data,” to “server 132 for maintenance
planning purpose.” Id. at 48 (emphasis added) (citing Ex. 1037 ¶ 28).
Petitioner also indicates that Haas sets forth “a four-level time frame for
repairing detected rail component problems: (1) immediately, (2) soon, (3)
during the next inspection, or (4) during the next scheduled maintenance
window or for use during planning.” Id. (citing Ex. 1037 ¶ 38). All
together, these portions of Haas disclose creating a tie replacement plan,
broken down into several levels of urgency, based upon the location of each
railroad tie as well as other data collected by Haas.
v. Remaining Challenged Claims
We have also considered Petitioner’s and Patent Owner’s arguments
regarding the remainder of the independent and dependent claims challenged
under this ground, namely, claims 2, 3, 7, 11, 13, 16, and 18–20, as follows.
We note generally that Patent Owner does not present substantive arguments
directed to claims 2, 3, 7, 11, 13, 16, and 18–20 discussed below, apart from
referring back to its arguments for claims 1, 8, 12, and 17.
a. Claim 2
IPR2019-00620
Patent 9,441,956 B2

43

Claim 2 depends from claim 1 and further requires that “the processor
is further configured to: a. determine a location of each railroad tie with
respect to the railroad track; and b. correlate the tie location with GPS and/or
milepost location.” Petitioner argues that “Villar and Haas disclose
claim 2’s additional limitations, and, Villar, Haas, and Uzarski render
obvious claim 2.” Pet. 38–39 (relying on Ex. 1036, 6:5–25; 11:18–20;
Ex. 1037 ¶¶ 27, 40, 43; Ex. 1004 ¶ 75).
b. Claims 3, 11, and 19
Claim 3 depends from claim 1 and further requires that “the condition
metrics include at least one of: tie warpage, tie surface roughness, tie
condition, tie entropy, tie heuristics, tie length, tie width, tie skew angle, tie
plate cut, differential plate cut, spike height, depth of adzing, broken
condition, curvature, tonnage, speed and decay zone.” Petitioner argues that
“Villar, Haas, and Uzarski render obvious claim 3,” relying on Uzarski to
disclose determining a series of tie condition metrics relating to the
condition of the tie, “including: (1) the type of distress, (2) severity of
distress, and (3) amount or density of distress.” Pet. 40 (citing Ex. 1038, 33–
40; Ex. 1004 ¶ 76).
Claim 11 “depends from claim 8 and recites the same limitations as
claim 3,” and Petitioner argues that, as “discussed in claim 3, Uzarski
discloses claim 11’s additional limitation” so that “Villar, Haas, and Uzarski
thus render obvious claim 11.” Pet. 46 (citing Ex. 1004 ¶ 91).
Claim 19 “depends from claim 17 and recites the same limitations as
claim 11,” and Petitioner argues that, as “discussed in claim 11, Villar,
Haas, and Uzarski render obvious claim 19.” Pet. 55 (citing Ex. 1004
¶ 110).
IPR2019-00620
Patent 9,441,956 B2

44

c. Claims 7 and 16
Claim 7 depends from claim 1 and requires that “the processor is
further configured to identify at least one of un-ballasted bridge ties, a
fastener type, the start and end of a rail crossing, a switch, switch points,
switch ties, switch frogs and joint bars.” Petitioner relies on Haas to disclose
that diagnosed rail components “can include, e.g., rails, crossties, spikes,
spike holes, clips, tie plates, anchors, joint bars, joint bar bolts, nuts and
washers.” Pet. 40–41 (emphasis added) (citing Ex. 1037, Abstract, ¶¶ 2, 25,
27, 29, 36, claim 15, Fig. 1). Petitioner further explains that it “would have
been obvious to implement Villar’s processing device 60 to further identify
joint bars and/or nuts and washers when inspecting the railroad track, as
taught by Haas.” Id. at 41 (relying on Ex. 1004 ¶ 78; Ex. 1036, 1:39–2:6).
Petitioner argues that “Villar, Haas, and Uzarski thus render obvious
claim 7.” Id. at 42.
Claim 16 depends from claim 8, and Petitioner argues that it “recites
the same limitations as claim 7,” and thus, “Villar, Haas, and Uzarski render
obvious the additional limitation of claim 16.” Pet. 49 (citing Ex. 1004
¶ 98).
d. Claim 13
Claim 13 depends from claim 8 and further comprises “a. determining
a location of each railroad tie in the set of railroad ties with respect to a
railroad track bed; and b. correlating the tie location with GPS and/or
milepost location.” Petitioner argues that “Villar’s inspection system 30
includes a GPS device 64 that determines the geographical location of
inspected railroad track components, including crossties,” as does Haas.
Pet. 49; Ex. 1036, 6:3–14; Ex. 1037 ¶¶ 27, 40, 43. Petitioner also relies on
IPR2019-00620
Patent 9,441,956 B2

45

Villar’s inspection system 30 that “measures the spacing between crossties
based on the plurality of image data of multiple ties and determines a
geographic location of the image data.” Id. (citing Ex. 1036, 2:22–24, 6:20–
21, 7:55–8:3, claims 1, 9, 22, 31, 47, 55). Petitioner argues that “Villar,
Haas, and Uzarski thus render obvious claim 13.” Id.
e. Claim 18
Claim 18 depends from claim 17 and further requires “a. determining
locations of ties with respect to the railroad track; b. correlating the tie
location with GPS and/or milepost location relative to the railroad track; and
c. creating a tie replacement plan that is based upon the GPS and/or milepost
location.” Petitioner relies on Villar and Haas to disclose “determining the
location of railroad components including ties (claim 18[a.]) and correlating
the locations with GPS relative to the railroad track (claim 18[b.]).” Pet. 54.
Petitioner refers to its claim 12 discussion to argue that “it would have been
obvious to implement Villar, Haas, and Uzarski’s system with Haas’ tie
replacement plan based on the GPS location of the captured images, as
recited by claim 18[c.].” Id. (citing Ex. 1004 ¶ 108–109).
f. Claim 20
Claim 20 depends from claim 19 and further comprises “adjusting the
condition grade of the tie based upon at least one of the condition metrics.”
Petitioner refers to its claim 1[f] arguments and Uzarski’s Equation 2 to
argue that Uzarski discloses this limitation. Pet. 55. Petitioner relies on
Uzarski’s explanation that from “a rating perspective, it was found that as
additional distress types and/or severity levels occurred in the same sample
unit, the impact of any given distress on the condition rating became less,”
and to account for this in the model, “an adjustment factor must be applied
IPR2019-00620
Patent 9,441,956 B2

46

to the sum of the individual deducts.” Id. at 55–56. Petitioner further relies
on Dr. Papanikolopoulos’s testimony that “one of ordinary skill in the art
would have understood that Uzarski’s adjustment factor ‘adjusts the
condition grade of the tie.’” Id. at 56 (citing Ex. 1004 ¶ 113).
For the same reasons provided above for claims 1, 8, 12, and 17, as
well as the arguments and evidence presented by Petitioner, we are
persuaded that Petitioner has demonstrated by a preponderance of the
evidence that claims 2, 3, 7, 11, 13, 16, and 18–20 are obvious in view of
Villar, Haas, and Uzarski. Thus, based on the complete record currently
before us, we are persuaded that Petitioner has demonstrated by a
preponderance of the evidence that claims 1–3, 7, 8, 11–13, and 16–20 of
the ’956 patent are obvious over Villar, Haas, and Uzarski.
G. Asserted Obviousness Over Villar, Haas, Uzarski, Velten, and Kanade
(Ground 2)
Petitioner asserts that claims 9 and 10 of the ʼ956 patent are
unpatentable as obvious under 35 U.S.C. § 103 over Villar, Haas, Uzarski,
Velten, and Kanade. Pet. 57–61.
Claim 9 depends from claim 8, and further requires a 3-D optical
scanning system, and generating a computer-readable image. Ex. 1001,
24:40–44. Petitioner argues that one of ordinary skill in the art would
understand “that Villar’s inspection system is a ‘3-D’ optical scanning
system,” but that Villar does not disclose using processing device 60 to
analyze the computer-readable image representative of a profile of the
railroad ties. Pet. 57–58 (citing Ex. 1036, 5:44–49, 7:40–50, Figs. 11, 12;
Ex. 1004 ¶ 115). It would have been obvious, according to Petitioner, to
“identify boundaries of at least one railroad tie in Villar’s computer-readable
IPR2019-00620
Patent 9,441,956 B2

47

image, as evidenced by, e.g., Velten (Ex-1041) and Kanade (Ex-1010).” Id.
at 58 (citing Ex. 1004 ¶ 117). The motivation to make this modification
would be “to further automate tie fault detection in a computer-readable
image, as taught by Velten and Kanade.” Id. at 60 (citing Ex. 1004 ¶ 120)).
Claim 10 depends from claim 9 and further requires “a. projecting a
beam of light across a railroad track bed that includes the railroad ties;
b. receiving light reflected from the railroad ties with a camera and
producing a camera signal; and c. generating, from the camera signal, the
computer-readable image.” Ex. 1001, 24:45–53. Petitioner argues that
Villar’s inspection system 30
(1) projects a beam of light 42 across the railroad track bed that
includes railroad ties 10, (2) receives light 42 reflected from the
track bed that includes railroad ties with a camera and
producing camera image signals based on received light; and
(3) generates from the camera image signals, the computer-
readable image representative of a profile of the scanned
railroad track bed.

Pet. 61 (citing Ex. 1036, 3:7–8, 3:29–32, 3:45–49, 4:9–14, 4:37–58, 5:32–
38, 5:44–49, 7:40–50).
Patent Owner, noting that claims 9 and 10 are dependent on claim 8,
argues that because “claim 8 is patentable for at least the reasons stated
above, claims 9 and 10 are patentable.” PO Resp. 57.
Petitioner replies that, apart from the claim 8 argument, Patent
Owner’s only argument against this ground is that “Kanade is non-analogous
art,” which, according to Petitioner, is incorrect. Reply 25 (citing PO
Resp. 7 (a one-paragraph summary of Kanade concluding: “As such,
Kanade is non-analogous art”). Petitioner argues that “Kanade and the
claims overlap in the broader field of machine vision, and thus, Kanade is
IPR2019-00620
Patent 9,441,956 B2

48

reasonably pertinent to the problem of object recognition.” Id. (citing
Unwired Planet, LLC v. Google Inc., 841 F.3d 995, 1000–1001 (Fed.
Cir. 2016)).
Two separate tests define the scope of analogous prior art:
“(1) whether the art is from the same field of endeavor, regardless of the
problem addressed and, (2) if the reference is not within the field of the
inventor’s endeavor, whether the reference still is reasonably pertinent to the
particular problem with which the inventor is involved.” In re Bigio, 381
F.3d 1320, 1325 (Fed. Cir. 2004). Neither party makes well-developed
arguments regarding whether Kanade meets this two-part test. Applying this
two-part test to the evidence before us, we find no discrete arguments that
Kanade is from the same field of endeavor, which is characterized by Patent
Owner as an “automated inspection system for railroads” (PO Resp. 7), and
by the challenged patent as a “system and method for inspecting railroad
ties.” Ex. 1001, code (54). We do agree with Petitioner, however, that
Kanade, a textbook titled “Three-Dimensional Machine Vision” deals
broadly with the field of machine vision. Reply 25; Ex. 1010, 2. More
particularly, Kanade “contains an entire section on 3D recognition
algorithms” in addition to its broader treatment of machine vision. Ex. 1004
¶ 119. Kanade’s broad scope and its particular descriptions of 3D imaging
are reasonably pertinent to the field of endeavor, which seeks to “automate
tie fault detection in a computer-readable image.” Ex. 1004 ¶ 120. Thus,
Kanade satisfies the second prong of the test as being “reasonably pertinent
to the problem of object recognition.” Reply 25.
Based on the arguments, testimony and evidence before us, we are
satisfied Petitioner has demonstrated by a preponderance of the evidence
IPR2019-00620
Patent 9,441,956 B2

49

that claims 9 and 10 are unpatentable as obvious under 35 U.S.C. § 103 over
Villar, Haas, Uzarski, Velten, and Kanade.
III. CONTINGENT MOTION TO AMEND
A. Introduction
Having determined that claims 1–3, 7–13, and 16–20 of the ’956
patent are unpatentable, we address Patent Owner’s Contingent Motion to
Amend.
B. Legal Standard
In an inter partes review, amended claims are not added to a patent as
a matter of right; rather, they must be proposed as a part of a motion to
amend. 35 U.S.C. § 316(d). The Board must assess the patentability of
proposed substitute claims “without placing the burden of persuasion on the
patent owner.” Aqua Prods., Inc. v. Matal, 872 F.3d 1290, 1328 (Fed.
Cir. 2017) (en banc); see Lectrosonics, Inc. v. Zaxcom, Inc., IPR2018-01129,
Paper 15 at 3‒4 (PTAB Feb. 25, 2019) (precedential). Subsequent to the
issuance of Aqua Products, the Federal Circuit issued a decision in Bosch
Automotive Service Solutions, LLC v. Matal, 878 F.3d 1027 (Fed. Cir. 2017),
as well as a follow-up order amending that decision on rehearing. See Bosch
Auto. Serv. Sols., LLC v. Iancu, No. 2015-1928 (Fed. Cir. Mar. 15, 2018)
(Order on Petition for Panel Rehearing).
In accordance with Aqua Products, Bosch, and Lectrosonics, a patent
owner does not bear the burden of persuasion to demonstrate the
patentability of the substitute claims presented in the motion to amend.
Rather, ordinarily, “the petitioner bears the burden of proving that the
proposed amended claims are unpatentable ‘by a preponderance of the
evidence.’” Bosch, 878 F.3d at 1040 (as amended on rehearing); see
IPR2019-00620
Patent 9,441,956 B2

50

Lectrosonics, Paper 15 at 3–4. In determining whether a petitioner has
proven unpatentability of the substitute claims, the Board focuses on
“arguments and theories raised by the petitioner in its petition or opposition
to the motion to amend.” Nike, Inc. v. Adidas AG, 955 F.3d 45, 51 (Fed.
Cir. 2020). Where the finding of unpatentability with respect to the
proposed amended claims departs materially from grounds or issues
addressed by the parties, the parties must ordinarily be provided prior notice
and an opportunity to be heard. Hunting Titan, Inc. v. Dynaenergetics
Europe GmbH, IPR2018-00600, Paper 67, 22 (PTAB July 6, 2020)
(precedential) (explaining that “raising a ground of unpatentability for the
first time in a final written decision, after the parties have exhausted their
ability to submit briefing and argument to the Board, does not provide
sufficient notice of, and opportunity to respond to, such a ground”). Thus,
the Board determines whether substitute claims are unpatentable by a
preponderance of the evidence based on the entirety of the record, including
any opposition made by a petitioner.
Notwithstanding the foregoing, a patent owner’s proposed substitute
claims must meet the statutory requirements of 35 U.S.C. § 316(d) and the
procedural requirements of 37 C.F.R. § 42.121. Lectrosonics, Paper 15 at 4–
8. Accordingly, a patent owner must demonstrate: (1) the amendment
proposes a reasonable number of substitute claims; (2) the proposed claims
are supported in the original disclosure (and any earlier filed disclosure for
which the benefit of filing date is sought); (3) the amendment responds to a
ground of unpatentability involved in the trial; and (4) the amendment does
not seek to enlarge the scope of the claims of the patent or introduce new
subject matter. See 35 U.S.C. § 316(d); 37 C.F.R. § 42.121.
IPR2019-00620
Patent 9,441,956 B2

51

C. Proposed Substitute Claims
Patent Owner proposes contingently substituting claims 30–44 for
claims 1–3, 7–13, and 16–20. Mot. Amend 1. Proposed independent
substitute claims 30, 34, and 41 are reproduced below, with underlining
indicating text added to the claims that Patent Owner proposes to replace.
30. A system for inspecting railroad ties, comprising:
a. a rail vehicle, configured for moving along a railroad
track in a travel direction;
b. a light generator, attached to the rail vehicle, oriented to
project a beam of light across the railroad track while moving
thereon;
c. an optical receiver, attached to the rail vehicle, oriented
to receive at least a portion of the beam of light reflected from
the railroad track and configured to generate image data
representative of a profile of at least a portion of the railroad
track; and
d. a processor, configured to analyze the image data by
applying one or more algorithms configured to:
i. find boundaries of at least one railroad tie in the
railroad track; and
ii. determine the condition of the at least one
railroad tie by:
a. determining a series of condition metrics of
the at least one railroad tie;
b. applying a unique weight factor to each
condition metric; and
c. computing a condition grade of the at least
one railroad tie based on the weighted condition
metrics.
Mot. Amend 2–3.
IPR2019-00620
Patent 9,441,956 B2

52

34. A method for inspecting railroad ties, comprising:
a. scanning a set of railroad ties with an optical scanning
system to produce image data;
b. analyzing the image data using a processor running a
machine vision algorithm;
c. identifying, via the processor, boundaries of at least one
railroad tie in the set of railroad ties;
d. determining, via the processor, condition metrics for the
at least one railroad tie in the set of railroad ties; and
e. computing a condition grade for the at least one railroad
tie based on a weighted summation of the condition metrics,
wherein the weighted summation of the condition metrics
comprises applying a unique weight factor to each of the
condition metrics.
Id. at 4–5.
41. A method for grading railroad ties of a railroad track
in situ, comprising:
a. scanning a set of railroad ties with an optical scanning
system and generating image signals representing the railroad
ties;
b. applying, via a computer processor, a machine vision
algorithm to the image signals and generating a 3-D profile of
the railroad ties that locates tie features;
c. analyzing the 3-D profile via a computer processor to
determine a set of condition metrics representing each railroad
tie in the set of railroad ties; and
d. computing a condition grade for each railroad tie based
on a weighted summation of the condition metrics;
e. determining a location of each railroad tie in the set of
railroad ties;
f. determining a spacing between railroad ties with a good
or bad condition grade; and
IPR2019-00620
Patent 9,441,956 B2

53

g. correlating the condition grade, location, and spacing
between ties with good or bad tie condition grades to create a tie
replacement plan.
Id. at 7–8.
D. Statutory and Procedural Requirements
A motion to amend may not present substitute claims that introduce
new subject matter. 35 U.S.C. § 316(d)(3); 37 C.F.R. § 42.121(a)(2)(ii).
New matter is any addition to the claims that lacks sufficient support in the
subject patent’s original disclosure. TurboCare Div. of Demag Delaval
Turbomach. Corp. v. Gen. Elec. Co., 264 F.3d 1111, 1118 (Fed. Cir. 2001)
(“When the applicant adds a claim . . . , the new claims . . . must find support
in the original specification.”). The Board requires that a motion to amend
set forth written-description support in the originally filed disclosure of the
subject patent for each proposed substitute claim and also set forth support
in an earlier-filed disclosure for each claim for which a patent owner seeks
the benefit of the earlier-filed disclosure’s filing date. See 37 C.F.R.
§ 42.121(b)(1)–(b)(2).
In its Motion to Amend, Patent Owner identifies citations in the
original disclosure of the ’956 patent that it asserts provide support for the
proposed amendment of each claim. Mot. Amend 10 (citations omitted).
Petitioner contends that proposed substitute claims 30–40 do not
comply with the requirement that the proposed claims are supported in the
original disclosure, arguing that proposed substitute claims 30–40 are invalid
for lacking written description support. See Opp. 6–7; see 35 U.S.C.
§ 316(d); 37 C.F.R. § 42.121. Proposed substitute claim 30 recites
“applying a unique weight factor to each condition metric.” Mot. Amend 3
(proposed amendment underlined). Petitioner argues the claim fails “to
IPR2019-00620
Patent 9,441,956 B2

54

comply with the written description requirement because the ’956 original
disclosure fails to show that the inventors invented applying a ‘unique’
weight factor to each condition metric.” Opp. 6. However, Equation (1) of
the original disclosure shows a mathematical formula in which “wn
represents the unique weight factor that is applied to each variable.”
Ex. 1002, 68 (¶ 111); Ex. 1003, 33 (¶ 111). As discussed above,
Equation (1) describes a formula that multiplies a particular weight factor
uniquely with a particular tie metric. Accordingly, the ’956 original
disclosure provides adequate written description support for a “unique”
weight factor.
Petitioner does not otherwise dispute that proposed substitute
claims 30–40 satisfy the requirements that the amendment propose a
reasonable number of substitute claims; the amendment respond to a ground
of unpatentability involved in the trial; and the amendment does not seek to
enlarge the scope of the claims of the patent or introduce new subject matter.
See generally Opp.; see 35 U.S.C. § 316(d); 37 C.F.R. § 42.121. Petitioner
also does not dispute that proposed substitute claims 41–44 satisfy the
statutory requirements of 35 U.S.C. § 316(d) and the procedural
requirements of 37 C.F.R. § 42.121. See generally Opp. On this record, we
determine that Patent Owner’s Motion to Amend meets the statutory and
regulatory requirements under 35 U.S.C. § 316(d) and 37 C.F.R. § 42.121.
Because we ultimately find Petitioner has proven unpatentability, we need
not analyze these requirements further.
E. Analysis of the Substitute Claims
Petitioner challenges the patentability of claims 30–44 on the grounds
that the claims are invalid as indefinite, as lacking written description, and as
IPR2019-00620
Patent 9,441,956 B2

55

obvious over prior art. Opp. 1, 6–7. First, regarding indefiniteness,
Petitioner argues that the term “unique” recited in proposed substitute claims
30–40 is indefinite. Id. at 1. Second, Petitioner argues that the term “a good
or bad condition grade” recited in the proposed substitute claims 41–44 is
indefinite. Id. at 5. Third, regarding unpatentability over the prior art,
Petitioner challenges the patentability of proposed substitute claims 30–40 as
obvious over combinations including Villar, Haas, and Uzarski (id. at 9–10)
as well as over combinations including Villar, Haas, and Nielsen11 (id. at
11–19). Finally, Petitioner challenges the patentability of proposed
substitute claims 41–44 as obvious over the combination of Villar, Haas,
Uzarski, and Palese.12 Id. at 19–25. We address each of these challenges in
turn.
i. Indefiniteness of Proposed Substitute Claims 30–40
Proposed substitute claim 30 recites “applying a unique weight factor
to each condition metric.” Mot. Amend 3 (proposed amendment
underlined). Proposed substitute claim 34 includes a similar amendment.
Id. at 5. Patent Owner submits that the claim amendment requires that “a
different weight factor—i.e., a factor not ‘like’ or ‘equal’ to any other
factor—is applied to each variable.” PO Resp. to Opp. 2. Patent Owner
further asserts that “unique” “plainly means that the weighting factor is
either unique or it is not. In other words, the weighting factor cannot be the
same as another weighting factor.” Id. at 1. Petitioner argues that the

11 U.S. Pub. No. 2010/0010863 A1, published January 14, 2010 (Ex. 1055).
12 J. W. Palese et al., Tie Planning Tools for the Track Inspector, The
American Railway Engineering and Maintenance of Way Association, Track
& Structures Annual Conference (1999) (Ex. 1056).
IPR2019-00620
Patent 9,441,956 B2

56

“phrase ‘unique’ is a subjective term of degree” and, thus, is indefinite.
Opp. 1–2.
We are persuaded the claims fail to inform a person of ordinary skill
in the art with reasonable certainty the scope of “a unique weight factor.”
We perceive an inconsistency between Patent Owner’s definition of “a
unique weight factor” and the ’956 patent’s use of “a unique weight factor”
as described in Equation (1). Ex. 1001, 21:10–17. Patent Owner asserts that
“the ordinary meaning of the word ‘unique’” is “commonly defined as
‘being the only one’ or ‘being without like or equal.’” PO Resp. to Opp. 1.
Patent Owner further argues that “applying a unique weight factor to each
condition metric” means “the weighting factor cannot be the same as another
weighting factor” and “a different weight factor—i.e., a factor not ‘like’ or
‘equal’ to any other factor—is applied to each variable.” Id. at 1–2
(emphasis added); see Mot. Amend 14. When asked whether a weight factor
could be a common value applied to each condition metric (“What prohibits
the weight factor from being say, five in each case when applying each
condition metric?”), Patent Owner responded “the words each implies that
you’re going to have a weight factor for each individual metric. So that
gives you some clue that it’s not going to be the same one, otherwise it
would say applying a weight factor to the condition metric.” Tr. 34:14–21
(emphasis added). That is, Patent Owner’s arguments suggest that “a unique
weight factor” cannot have the value of any other weight factor. See id.; see
also PO Resp. to Opp. 1–2.
However, in contrast to Patent Owner’s description, the ’956 patent
supports an interpretation in which a weight factor is “unique” to the
particular condition metric to which it is applied, but not necessarily a value
IPR2019-00620
Patent 9,441,956 B2

57

different than or unequal to another weight factor. The ’956 patent discloses
that tie score ST can be calculated by Equation (1):

“where wn represents the unique weight factor that is applied to each
variable, and Mn represents each tie metric that is used.” Ex. 1001, 21:10–17
(emphasis added). Equation (1) does not necessarily require that its unique
weight factor variables wn, e.g., w1, w2, w3, etc., are different factors, i.e.,
values, from one another. Rather, Equation (1), which multiplies a particular
weight factor and a particular tie metric, suggests that a weight factor wn is
unique respective to its corresponding tie metric Mn. For example, w1 is
uniquely applied to M1, w2 is uniquely applied to M2, and so forth. Such an
application includes instances in which w1 and w2 may be the same value,
but are still unique to the tie metric that the weight factor is applied to, i.e.,
M1 and M2 respectively. Patent Owner’s asserted interpretation, however,
precludes w1 and w2 from having the same value (see PO Resp. to Opp. 1–4;
Tr. 34:14–35:12), and so highlights the uncertainty of the term “unique.”
That is, the claim fails to inform a person of ordinary skill in the art with
reasonable certainty whether “a unique weight factor” applied to each
condition metric is unique in that the weight factor is only applied to its
corresponding condition metric, or unique in that it is different than any
other weight factor, i.e., there appear to be two plausible interpretations.
“As the statutory language of ‘particular[ity]’ and ‘distinct[ness]’ indicates,
claims are required to be cast in clear—as opposed to ambiguous, vague,
indefinite—terms. It is the claims that notify the public of what is within the
protections of the patent, and what is not.” In re Packard, 751 F.3d 1307,
1313 (Fed. Cir. 2014); see also Nautilus v. Biosig Instruments, Inc., 572 U.S.
IPR2019-00620
Patent 9,441,956 B2

58

898, 898–899 (2014) (“A patent is invalid for indefiniteness if its claims,
read in light of the specification delineating the patent, and the prosecution
history, fail to inform, with reasonable certainty, those skilled in the art
about the scope of the invention.”).
Accordingly, we conclude that proposed substitute claims 30 and 34
are indefinite. Proposed substitute claims 31–33 and 35–40 depend from
proposed substitute claims 30 and 34 and, as such, also are indefinite.
ii. Indefiniteness of Proposed Substitute Claims 41–44
Proposed substitute claim 41 recites, in relevant part, “d. computing a
condition grade for each railroad tie . . . f. determining a spacing between
railroad ties with a good or bad condition grade.” Mot. Amend 7 (proposed
amendment underlined). Petitioner argues the Specification of the ’956
patent “fails to provide an objective basis” for one of ordinary skill in the art
“to determine whether a tie has a ‘good or bad’ condition grade.” Opp. 5.
Patent Owner’s reply does not dispute that determining “a good or bad
condition grade” for a railroad tie is subjective. PO Resp. to Opp. 2.
Instead, Patent Owner argues:
No “objective basis” is needed for determining whether a tie’s
condition is “good” or “bad.” The point is that condition criteria
are applied (regardless of how implemented), and that ties are
assigned such a condition grade. It does not matter if anyone else
may apply the grades based on different criteria.
Id.
We are persuaded that the claim fails to inform a person of ordinary
skill in the art with reasonable certainty of the scope of “a good or bad
condition grade.” The ’956 patent itself states that “[i]t is to be appreciated
that this grading scale is somewhat arbitrary” in determining whether a tie as
IPR2019-00620
Patent 9,441,956 B2

59

good or bad. Ex. 1001, 21:21–30. Moreover, Petitioner’s expert asserts that
“one of ordinary skill in the art would understand the term ‘good or bad’” is
“a qualitative and subjective term of degree.” Ex. 1050 ¶ 35; see Opp. 5. As
such, both the Specification and expert testimony support Petitioner’s
assertion that the ’956 patent “fails to provide an objective basis for a one of
ordinary skill in the art to determine whether a tie has a ‘good or bad’
condition grade.” Opp. 5–6; Ex. 1050 ¶ 35.
Additionally, as Patent Owner argues, the limitation reciting
“f. determining a spacing between railroad ties with a good or bad condition
grade” does not recite determining a “good or bad” condition grade, because
that limitation determines spacing between ties, rather than determining tie
condition. PO Resp. to Opp. 2. However, to the extent that we can locate an
antecedent basis for the proposed “good or bad” limitation, the claim also
recites “d. computing a condition grade of the at least one railroad tie based
on the weighted condition metrics.” Ex. 1001, 23:57–58 (emphasis added).
Taking limitations d. and f. together, the claim suggests that the condition
grade determination includes “a good or bad” determination. As discussed
above, the Specification does not provide any basis for the ordinary skilled
artisan to objectively determine a “good or bad” condition, and, instead,
states that such condition grades are “somewhat arbitrary.” Id. at 21:21–30.
Determining the spacing between railroad ties with “a good or bad
condition grade” (emphasis added) is indefinite for an additional reason:
due to the use of “or,” the structure of the limitation makes the conditions for
determining spacing unclear. For example, determining the spacing between
railroad ties with “a good or bad condition grade” includes any of:
1) determining the spacing between consecutive good railroad ties,
IPR2019-00620
Patent 9,441,956 B2

60

2) determining the spacing between consecutive bad railroad ties, and
3) determining the spacing between good railroad ties and bad railroad ties.
All three interpretations are supported by the claim language, but it is
unclear which interpretation is the prevailing one. Furthermore, if the claim
includes all three options, then the limitation would be rendered meaningless
because the limitation would determine the spacing between any railroad ties
in any condition, i.e., good or bad, and the spacing between ties would be
arbitrarily determined. As such, the claim language, in view of the
Specification, fails to inform a person of ordinary skill about the scope of the
invention. See In re Packard, 751 F.3d at 1313; see also Nautilus, 572 U.S.
at 898–899.
Accordingly, we conclude that proposed substitute claim 41 is
indefinite. Proposed substitute claims 42–44 depend from proposed
substitute claim 41 and, as such, are also indefinite.
iii. Obviousness of Proposed Substitute Claims 30–40 over Villar,
Haas, and Nielsen
Proposed substitute claim 30 recites “applying a unique weight factor
to each condition metric.” Mot. Amend 3 (amendment underlined).
Although we conclude that “a unique weight factor” is indefinite, as
discussed above, the indefiniteness of the term does not impede our ability to
analyze the obviousness of the claim over the combination of Villar, Haas,
and Nielsen. Whether “a unique weight factor” is unique and only applied
to a respective condition metric, or has a unique value different than another
weight factor, we conclude that substitute claim 30 would have been obvious
over the combination of Villar, Haas, and Nielsen.
IPR2019-00620
Patent 9,441,956 B2

61

Nielsen is titled “Methods and Apparatus for Quality Assessment of a
Field Service Operation Based on Multiple Scoring Categories.”
Ex. 1055, code (54). As Nielsen explains, field service operations are “any
operation in which companies dispatch technicians and/or other staff to
perform certain activities, for example, installations, services and/or
repairs,” e.g., a “locate and marking operation.” Id. ¶¶ 5–6. Nielsen is
drawn to a “quality assessment system” that “may receive information
related to a locate and marking operation,” and that will “automatically
assess the quality of the operation.” Id. ¶ 64. In one embodiment, Nielsen
describes performing an automatic quality assessment by “selecting . . . a
weighting factor for each of the plurality of quality assessment criteria, the
weighting factor for each quality assessment criterion indicating the relative
importance of the quality assessment criterion to each of the other quality
assessment criteria.” Id. ¶ 30. Nielsen’s weighting factors for each of the
plurality of quality assessment criteria are “unique”—both uniquely applied
to a respective quality assessment criteria, and unique as having a unique
value different than another weight factor, thus, satisfying both
interpretations of “unique” discussed above. For example, unique weight
factor of “x1” is applied to “Accelerometer data” and a different unique
weight factor of “x5” is applied to “Battery strength data.” Id. at Table 6.
Petitioner asserts that Nielsen teaches “applying a unique weight
factor to each condition metric.” Opp. 15–16. Patent Owner does not
dispute that Nielsen teaches “applying a unique weight factor to each
condition metric” in performing a quality assessment and, instead, argues
that the combination of Villar, Haas, and Nielsen is improper. See PO Resp.
to Opp. 5–9. In particular, Patent Owner argues Nielsen is not analogous art.
IPR2019-00620
Patent 9,441,956 B2

62

Id. at 6. Patent Owner further argues that “there is absolutely no teaching or
suggestion in Haas that there needs to be ‘weighting’ of the alleged ‘tie
condition metrics,’” and so “[n]othing in Haas indicates any need for further
analysis that would prompt a POSITA to look” to Nielsen. Id. at 8.
As discussed above, two separate tests define the scope of analogous
prior art: “(1) whether the art is from the same field of endeavor, regardless
of the problem addressed and, (2) if the reference is not within the field of
the inventor’s endeavor, whether the reference still is reasonably pertinent to
the particular problem with which the inventor is involved.” In re Bigio, 381
F.3d at 1325. “A reference is reasonably pertinent if, even though it may be
in a different field from that of the inventor’s endeavor, it is one which,
because of the matter with which it deals, logically would have commended
itself to an inventor’s attention in considering his problem.” In re Clay, 966
F.2d 656, 659 (Fed. Cir. 1992).
Nielsen generally relates to “any operation in which companies
dispatch technicians and/or other staff to perform certain activities, for
example, installations, services and/or repairs,” e.g., a “locate and marking
operation.” Ex. 1055 ¶¶ 5–6. Nielsen’s generality indicates that its field of
endeavor may overlap with the field of the inventor’s endeavor, which is
characterized by Patent Owner as an “automated inspection system for
railroads” (PO Resp. 7), and by the challenged patent as a “system and
method for inspecting railroad ties.” Ex. 1001, code (54).
Even if Nielsen is in a different field of endeavor, Nielsen is
reasonably pertinent to the particular problem with which the invention is
concerned. The ’956 patent involves a type of automated quality
assessment: “to automatically identify tie defects and grade the ties.”
IPR2019-00620
Patent 9,441,956 B2

63

Ex. 1001, 1:26–29, 14:13–14. Similarly, Nielsen provides “automated
quality assessment” solutions. Ex. 1055 ¶¶ 63–64. Petitioner’s expert
confirms that Nielsen addresses “optimiz[ing] . . . computed condition
grades.” Ex. 1050 ¶ 48. Petitioner’s expert further states that a person of
ordinary skill in the art would have understood that Nielsen describes
improvements to quality assessments, e.g., “tak[ing] into account the relative
importance of one condition metric relative to another when determining the
condition grade, as some condition metrics can have a greater predictor of
the condition score.” Id. ¶ 49 (citing Ex. 1055 ¶¶ 30, 214). Accordingly,
because the ’956 patent is concerned with how to perform automatic quality
assessments, and a person of ordinary skill in the art would understand that
Nielsen provides solutions for automatically assessing quality, the inventor
of the ’956 patent would have looked to Nielsen in seeking solutions for
providing automated quality assessments. As such, Nielsen is analogous art.
Furthermore, even considering Patent Owner’s arguments that “there
is absolutely no teaching or suggestion in Haas that there needs to be
‘weighting’ of the alleged ‘tie condition metrics’” and “[n]othing in Haas
indicates any need for further analysis that would prompt a POSITA to look”
to Nielsen (PO Resp. to Opp. 8), the rationale Petitioner provides for the
combination of Villar, Haas, and Nielsen is reasonable. Patent Owner’s
argument appears to demand that some teaching, suggestion, or motivation
for the combination be found in the references themselves, but that approach
to analyzing the obviousness inquiry has been rejected as too rigid. KSR
Int’l Co., 550 U.S. at 415. Instead, obviousness requires “some articulated
reasoning with some rational underpinning to support the legal conclusion of
obviousness.” In re Kahn, 441 F. 3d 977, 988 (Fed. Cir. 2006). In this case,
IPR2019-00620
Patent 9,441,956 B2

64

Petitioner’s expert states a person of ordinary skill in the art would have
combined Villar, Haas, and Nielsen to apply
a unique weight factor to each of Haas’ condition metrics (e.g.,
moved or broken ties), such that the weight factor applied for
each condition metric is based on and indicative of the relative
importance of that condition metric (e.g., moved ties) to each of
the other condition metrics (e.g., broken ties), and the condition
grade is computed based on the weighted condition metrics.
Ex. 1050 ¶ 49. Petitioner’s expert explains that the skilled artisan “would
have been motivated to make this modification because, as Nielsen explains,
it would take into account the relative importance of one condition metric
relative to another when determining the condition grade, as some condition
metrics can have a greater predictor of the condition score.” Id. (citing
Ex. 1055 ¶¶ 30, 214). Petitioner’s expert further explains that the skilled
artisan would have been motivated to combine Villar, Haas, and Nielsen in
order “to make informed decisions by creating more comprehensive metrics
(often with the use of weight factors).” Id. As such, improving the
calculation of a score, i.e., improving a computed condition grade, rationally
supports the combination of Villar, Haas, and Nielsen.
Additionally, proposed substitute claim 34 would have been obvious
over the combination of Villar, Haas, and Nielsen, for reasons similar to
proposed substitute claim 30. Proposed substitute claim 34 recites, “e.
computing a condition grade for the at least one railroad tie based on a
weighted summation of the condition metrics, wherein the weighted
summation of the condition metrics comprises applying a unique weight
factor to each of the condition metrics.” Mot. Amend 5 (proposed
amendment underlined). Like proposed substitute claim 30, which recites
“applying a unique weight factor to each condition metric,” proposed
IPR2019-00620
Patent 9,441,956 B2

65

substitute claim 34 similarly recites “applying a unique weight factor to each
of the condition metrics.” Id. at 3, 5. Patent Owner argues that “substitute
claim 34[d] does not recite a weight summation of the condition metrics”
and we “should find substitute claim 34[d] patentable for at least this
reason.” PO Resp. to Opp. 8–9. Proposed substitute claim 34, however,
expressly recites “a weighted summation of the condition metrics.”
Furthermore, Petitioner argues, the combination of Villar, Haas, and Nielsen
teaches the proposed amendment in claim 34 reciting “wherein the weighted
summation of the condition metrics comprises applying a unique weight
factor to each of the condition metrics,” for the same reason as the
combination teaches “applying a unique weight factor to each condition
metric,” as recited in proposed substitute claim 30. Opp. 16–17. Notably,
Petitioner points out that Nielsen teaches “combining the weighted scoring
value or grades for all of the quality assessment criteria to generate at least
one indication of the quality assessment.” Opp. 16 (citing Ex. 1055 ¶ 30).
Thus, proposed substitute claim 34 would have been obvious over the
combination of Villar, Haas, and Nielsen for essentially the same reasons as
proposed substitute claim 30.
For the foregoing reasons and for the reasons discussed in our
unpatentability analysis of claim 1, we conclude that the subject matter of
proposed substitute claims 30 and 34 would have been obvious based on the
combined teachings of Villar, Haas, and Nielsen.
The proposed substitute dependent claims recite the same subject
matter as the original dependent claims but are amended to change their
dependency from the proposed substitute independent claims. Therefore, the
proposed substitute dependent claims are unpatentable for reasons discussed
IPR2019-00620
Patent 9,441,956 B2

66

above with respect to the proposed substitute independent claims and the
original dependent claims. More particularly, proposed substitute dependent
claims 31–33 and 35–40, which correspond, respectively, to claims 2, 3, 7,
9–13, and 16, are unpatentable as having been obvious over the combination
of Villar, Haas, and Nielsen.
In addition, Petitioner further asserts claims 30–40 are unpatentable
over the combination of Villar, Haas, and Uzarski or the combination of
Villar, Haas, Uzarski, Velten, and Kanade. Opp. 9–10. Because we already
conclude that claims 30–40 are unpatentable over the combination of Villar,
Haas, and Nielsen, we need not address the patentability of the claims over
the combinations of Villar, Haas, and Uzarski or the combination of Villar,
Haas, Uzarski, Velten, and Kanade.
Having considered the full record developed during trial, we deny
Patent Owner’s Motion to Amend as to proposed substitute claims 30–40
because they are unpatentable under 35 U.S.C. § 103.
iv. Obviousness of Proposed Substitute Claims 41–44
Proposed substitute claim 41 recites, in part,
e. determining a location of each railroad tie in the set of
railroad ties;
f. determining a spacing between railroad ties with a good
or bad condition grade; and
g. correlating the condition grade, location, and spacing
between ties with good or bad tie condition grades to create a tie
replacement plan.
Mot. Amend 7–8 (proposed amendments underlined).
Petitioner contends that the combination of Villar, Haas, Uzarski, and
Palese teaches the proposed amendment. Opp. 19–25. Villar, as discussed
above, describes an automated railroad track inspection system in which
IPR2019-00620
Patent 9,441,956 B2

67

“defects in [track] crossties 10 can be determined from [captured] image
data.” Ex. 1036, 9:30–31. Villar also describes that “the spacing between
crossties can be determined from the plurality of image data.” Id. at 7:55–
56. Palese, titled “Tie Planning Tools for the Track Inspector,” describes
“tools . . . to assist [a] tie inspector and planner in making more accurate and
optimal decisions concerning crosstie replacement.” Ex. 1056, 6, 9.
Specifically, tools “for recording, storing and analyzing tie condition
information” and a tool “evaluat[ing] historic tie installation information for
forecasting future tie requirements on a macro basis.” Id. Among other
features, Palese’s tools include a “lateral track strength measurement system
[that] can be used to locate clusters of ties not suitable in the lateral
direction.” Id. at 19. Palese defines a “tie cluster” as “continuous counts
of 2 to 10 (or more) bad ties in a row.” Id. at 25.
Patent Owner contends that proposed substitute claim 41 is not
unpatentable over Villar, Haas, Uzarski, and Palese. PO Resp. to Opp. 9–12.
First, Patent Owner argues that “Petitioner fails to identify how Palese
qualifies as prior art.” Id. at 9–10. Second, Patent Owner argues Palese’s
“tools [do] not provide a link to using an ‘automated’ track inspection
system.” Id. at 11. Third, Patent Owner argues Palese does not teach “a
‘spacing between ties with a good or bad tie condition grades.’” Id. at 12.
Fourth, Patent Owner argues a person of ordinary skill in the art “would not
look to combine Palese with Villar” because “Palese discloses the manual
inspection of railroad ties” while “Villar teaches away from the manual
inspection of railroad ties.” Id. at 10.
Initially, we disagree with Patent Owner that Palese has not been
properly established as qualified prior art by Petitioner. PO Resp. to Opp. 9.
IPR2019-00620
Patent 9,441,956 B2

68

Palese is an excerpt from the proceedings of the American Railway
Engineering & Maintenance of Way Association 1999 Annual Conferences.
Ex. 1056. On its face, Palese displays a 1999 copyright date and indicia that
it was produced in connection with conferences held in Baltimore, Maryland
from August 29–September 1, 1999 and in Chicago, Illinois from September
12–15, 1999. Id. at 1, 6. On Palese’s face is a stamp reading “M.I.T.
LIBRARIES RECEIVED” on December 18, 2000. Id. at 1. A Declaration
of Michael P. Carney, a research analyst at Finnegan, Henderson, Farabow,
Garrett & Dunner, LLP, was submitted with Palese, in which Mr. Carney
explains the significance of the stamp on Palese’s face. Ex. 1057 ¶ 1.
Mr. Carney opines that the “December 18, 2000, date stamp” on the face of
Palese “indicates that the MIT Libraries received a published copy of Palese
by that date and that it would have been shelved for the public within a few
days or weeks after that date.” Ex. 1057 ¶ 14. Further, Mr. Carney asserts
that Palese includes a Machine-Readable Cataloging (“MARC”) system
record, which includes a date field listing “the first six characters ‘010111’
in ‘YYMMDD’ format, indicating that Palese was first catalogued by the
UC Berkley Library on January 11, 2001” and, thus, was “publicly available
at least as early as January of 2001.” Id. ¶¶ 17–18.
On this complete record, we are persuaded that Petitioner has
adequately shown Palese is prior art. Citing the Palese reference and Mr.
Carney’s declaration, Petitioner explains that substitute claims 41–44 would
have been obvious over Villar, Haas, Uzarski, and Palese because “the
creation of a tie replacement plan accounting for the spacing between ties
with good or bad condition grades . . . was well-known in the art at the
relevant time, as evidenced by, e.g., Palese.” Opp. 19 (citing Ex. 1056;
IPR2019-00620
Patent 9,441,956 B2

69

Ex. 1057) (emphasis added). As such, Petitioner reasonably relies on
Mr. Carney’s declaration to establish that the teachings of Palese qualify as
prior art under 35 U.S.C. § 103. Patent Owner does not otherwise contest
that Palese was publicly available and is prior art before the critical date.
See Opp. 9–10. On this record, we credit the indicia of public accessibility
available on the face of the Palese reference, and the testimony of
Mr. Carney supporting Palese’s public accessibility as of the critical date.
Further, we disagree with Patent Owner that Palese does not teach an
“automated” track inspection system. PO Resp. to Opp. 11. Palese’s “tools
[are] available to railway inspection and maintenance planning personnel to
aid in a better understanding of tie degradation and future crosstie
requirements.” Ex. 1056, 19; see id. at 9. For example, one “class of tools
that aid the tie inspector is the lateral track strength measurement system”
that “can be used to locate clusters of ties not suitable in the lateral
direction.” Id. at 19. Palese describes that lateral track strength measuring
tools are automated: for example, an exemplary system “applies vertical and
lateral loads,” performs “extrapolation of the measured data under load,” and
“provides planning software in the form of exception reports, mile-by-mile
tie requirements, cluster breaking tie requirements, and a host of other
statistics.” Id. at 21–22, Fig. 10; see id. at 19–21, Fig. 8.
Moreover, Patent Owner’s attack on Palese individually does not
address the combination of Villar, Haas, Uzarski, and Palese. As discussed
above, Villar teaches the automated railroad track inspection system relied
upon by Petitioner to make its obviousness arguments. See Ex. 1036, 7:22–
34, 9:62–10:22. As such, even assuming that Palese does not teach an
“automated” railroad track inspection system, the combination of Villar,
IPR2019-00620
Patent 9,441,956 B2

70

Haas, Uzarski, and Palese teaches an automated railroad track inspection
system.
We also disagree with Patent Owner’s argument that “Petitioner never
contends that Palese determines a ‘spacing between ties with a good or bad
tie condition grades.’” PO Resp. to Opp. 12. Petitioner points out that
Palese’s system locates “clusters of bad ties” and “ascertains tie condition
distribution.” Opp. 22 (citing Ex. 1056, 19–22, 25). Indeed, Palese analyzes
a “detailed distribution of ties . . . to provide the user with number of tie
clusters.” Ex. 1056, 25. Palese defines “tie clusters” as “continuous counts
of 2 to 10 (or more) bad ties in a row.” Id. As such, the identification of tie
clusters by Palese, as pointed out by Petitioner (Opp. 22; Paper 25, 24:14–
20, 25:5–7) determines a spacing between ties with bad conditions—namely,
that there is no spacing between bad ties because the bad ties are continuous.
We also disagree with Patent Owner’s argument that Villar and Palese
would not have been combined by one of ordinary skill in the art because
“Villar teaches away from the manual inspection of railroad ties,” while
Palese states that “walking tie inspections are still very important.”
PO Resp. to Opp. 10. Like our discussion above regarding the combination
of Villar and Uzarski, we are not directed to any portion of Villar that
discredits or discourages incorporation of a method for determining tie
spacing and tie conditions such as the one disclosed in Palese. Polaris
Indus., Inc., 882 F.3d at 1069. Furthermore, although Palese states “walking
tie inspections,” i.e., manual inspections, “are still very important,” in
context, Palese makes that statement because its system is “limited to testing
groups of ties in the lateral plane only.” Ex. 1056, 22. That is, Palese
describes that its automated track inspection system is unable to detect
IPR2019-00620
Patent 9,441,956 B2

71

problematic ties in a non-lateral plane, and, so, the reason manual
inspections are important is to identify problematic ties in the non-lateral
plane. As such, Palese fails to teach or suggest that its manual inspections
should not be replaced by automatic tie inspections. Moreover, Patent
Owner’s argument is based on the improper bodily incorporation of manual,
non-lateral inspections. PO Resp. to Opp. 10. The proffered combination
does not require the incorporation of manual inspections. Accordingly,
neither Villar nor Palese teaches away from their combination. Instead, we
agree with Petitioner’s expert that a person of ordinary skill in the art would
have “been motivated to” combine the references in order “to create an
optimized crosstie inspection and maintenance plan to help direct an
operator to where and how many ties should be installed.” Ex. 1050 ¶ 62.
We also agree with Petitioner’s expert that a person of ordinary skill in the
art would have “been motivated to” combine the references in order to
improve “inspection of rail components” because Palese teaches
improvements for rail inspections system. Id. (citing Ex. 1056, 8–9, 11, 18,
25–28, 31). As discussed above, we agree that implementing the automated
track inspection and tie grading system of Villar and Haas and Uzarski to
help direct an operator to where and how many ties should be installed
pursuant to an optimized tie replacement plan, as taught by Palese, would
simply be a predictable extension of Villar’s teachings. Opp. 23. For the
foregoing reasons and for the reasons discussed in our unpatentability
analysis of claim 17, we conclude that the subject matter of proposed
substitute claim 41 would have been obvious based on the combined
teachings of Villar, Haas, Uzarski, and Palese.
IPR2019-00620
Patent 9,441,956 B2

72

The proposed substitute dependent claims recite the same subject
matter as the original dependent claims but are amended to change their
dependency from the proposed substitute independent claims. Therefore, the
proposed substitute dependent claims are unpatentable for reasons discussed
above with respect to the proposed substitute independent claims and the
original dependent claims. More particularly, proposed substitute dependent
claims 42–44, which correspond, respectively, to claims 18–20, are
unpatentable as having been obvious over Villar, Haas, Uzarski, and Palese.
Having considered the full record developed during trial, we deny
Patent Owner’s Motion to Amend as to proposed substitute claims 30–44.
IV. CONCLUSION13
For the reasons discussed above, we determine Petitioner has proven,
by a preponderance of the evidence, that the challenged claims are
unpatentable, as summarized in the following table:
Claims 35 U.S.C. § References/ Basis
Claims
Shown
Unpatentable
Claims Not
Shown
Unpatentable
1–3, 7, 8,
11–13, 16–
20
103(a) Villar,
Haas,
Uzarski
1–3, 7, 8, 11–
13, 16–20

13 Should Patent Owner wish to pursue amendment of the challenged claims
in a reissue or reexamination proceeding subsequent to the issuance of this
decision, we draw Patent Owner’s attention to the April 2019 Notice
Regarding Options for Amendments by Patent Owner Through Reissue or
Reexamination During a Pending AIA Trial Proceeding. See 84 Fed.
Reg. 16654 (Apr. 22, 2019). If Patent Owner chooses to file a reissue
application or a request for reexamination of the challenged patent, we
remind Patent Owner of its continuing obligation to notify the Board of any
such related matters in updated mandatory notices. See 37 C.F.R.
§ 42.8(a)(3), (b)(2).
IPR2019-00620
Patent 9,441,956 B2

73

Claims 35 U.S.C. § References/ Basis
Claims
Shown
Unpatentable
Claims Not
Shown
Unpatentable
9, 10 103(a) Villar,
Haas,
Uzarski,
Velten,
Kanade
9, 10
Overall
Outcome
1–3, 7–13,
16–20

We also determine that proposed substitute claims 30–44 are
unpatentable and deny the motion to amend as to these claims, as
summarized in the following table:

Motion to Amend Outcome

Claims
Original Claims Cancelled by Amendment
Substitute Claims Proposed in the Amendment 30–44
Substitute Claims: Motion to Amend Granted
Substitute Claims: Motion to Amend Denied 30–44
Substitute Claims: Not Reached

V. ORDER
In consideration of the foregoing, it is hereby:
ORDERED that Petitioner establishes, by a preponderance of the
evidence, that claims 1–3, 7–13, and 16–20 of U.S. Patent No. 9,441,956 B2
are unpatentable; and
IPR2019-00620
Patent 9,441,956 B2

74

FURTHER ORDERED that Patent Owner’s Contingent Motion to
Amend is denied; and
FURTHER ORDERED that this is a Final Written Decision;
therefore, parties to the proceeding seeking judicial review of the decision
must comply with the notice and service requirements of 37 C.F.R. § 90.2.

IPR2019-00620
Patent 9,441,956 B2

75

FOR PETITIONER:

Aaron Parker
David Reese
Nicholas Cerulli
FINNEGAN, HENDERSON, FARABOW,
GARRETT & DUNNER, LLP
aaron.parker@finnegan.com
david.reese@finnegan.com
nicholas.cerulli@finnegan.com

FOR PATENT OWNER:

Rexford Johnson
Dana Herberholz
C. Kevin Speirs
Jordan Stott
PARSONS, BEHLE & LATIMER
rjohnson@parsonbehle.com
dherberholz@parsonsbehle.com
kspeirs@parsonsbehle.com
jstott@parsonbehle.com