TITLE: B-299076, Fantastic Data, February 5, 2007
BNUMBER: B-299076
DATE: February 5, 2007
******************************************
B-299076, Fantastic Data, February 5, 2007
Decision
Matter of: Fantastic Data
File: B-299076
Date: February 5, 2007
Dale Pitman for the protester.
Peter Giella, Esq., and Vera Meza, Esq., Department of the Army, for the
agency.
Charles W. Morrow, Esq., and James A. Spangenberg, Esq., Office of the
General Counsel, GAO, participated in the preparation of the decision.
DIGEST
Protest against procuring agency's decision not to fund protester's Phase
I proposal under the Department of Defense Small Business Innovative
Research program is denied where the record shows that the agency
reasonably evaluated the proposal.
DECISION
Fantastic Data protests the decision of the Department of the Army not to
fund its phase I proposal under Department of Defense (DOD) Small Business
Innovation Research (SBIR) program solicitation No. FY06.2.
We deny the protest.
The SBIR program is conducted pursuant to the Small Business Innovation
Development Act, 15 U.S.C. sect. 638 (2000), which requires certain
federal agencies to reserve a portion of their research and development
funds for awards to small businesses. As part of its SBIR program, DOD
issues an SBIR solicitation twice a year listing the research topics for
which it will consider SBIR program admission.
Firms first apply for a 6-month phase I award to test the scientific,
technical, and commercial merit and feasibility of a certain concept. If
phase I is successful, the firm may be invited to apply for a phase II
award to further develop the concept. After the completion of phase II,
firms are expected to obtain funding from the private sector and/or
non-SBIR government sources to develop the concept into a product for sale
in private sector and/or military markets. See DOD's SBIR Website,
http://www.acq.osd.mil/sadbu/sbir/overview/index.htm.
The solicitation included Army Topic A06-039, "Versatile Sensor
Network/Data Fusion Optimization System." The objective of this topic is
to "[d]evelop a versatile wireless acoustic sensor network development
system which will allow the optimization of high performance collaborative
data fusion within scalable distributed sensor network architectures."[1]
Agency Report, Tab B, Topic A06-039, at 1.
In the area of acoustic signal processing,[2] the military customarily has
relied on low cost distributed networks using 15 to 60 randomly placed
sensor nodes.[3] The utility of the process has been enhanced by the
development of communication networks based upon MESH/Ad-hoc and ZigBee[4]
which allow scaling of the network to increase the number of nodes. The
challenge has been how to optimize "sensor algorithms and
hardware/software solutions for high performance data fusion within
scalable sensor networks" because most networks rely upon "small
collaborative networks which do not optimize what can be realized with
high performance data fusion . . . in scalable networks." Thus, the Army
sought here to acquire an "innovative development approach and wireless
system architecture to facilitate the optimization of data fusion for
scalable networks." Id.
In this connection, the topic describes a system comprised of "at least 40
wireless acoustic sensing/processing nodes capable of communicating over
distances of approximately 200 meters between nodes"; that the "[n]odes
should embrace an open architecture that will enable third-party
developers to port their software algorithms through downloadable applets
that encode, filter, condition and classify acoustic signals"; that the
"design of each node should allow common microphone interface and preamp,
a processor to house the sensor processing and data fusion algorithms,
computer interface to monitor and reprogram sensor/data fusion algorithms,
and [the] communication system"; and that the "[n]odes should be capable
of forwarding acoustic processed or unprocessed information over the
wireless network to a central processing console," including being "able
to time stamp information packets forwarded from each node with time
precision accuracy of less than 1 millisecond." Also, the nodes had to be
designed to "accommodate a wide range of potential sensor functionalities,
ranging from minimal data fusion collaboration to more complex gateway
level data fusion functionality [and] [t]he central processing console
should be capable of performing further signal processing and system level
data fusion using third-party algorithms as well as recording all acquired
data within the sensor network." Id.
With respect to phase I, the topic requires the following:
Review existing acoustic algorithms, data fusion approaches, and network
architecture designs. Examine commercially available ZigBee or other
applicable COTS [commercial-off-the-shelf] communications systems.
Develop a flexible wireless acoustics network system architecture that
will streamline the ability to develop, test, and optimize existing and
future battlefield acoustic data fusion algorithms and network
architecture designs. Demonstrate essential network functionality by
collecting and forwarding real-time acoustic signals to a central
processor using no less than 15 acoustic sensor nodes.
Id. at 1-2.
For phase I, the solicitation contemplated multiple awards of fixed-priced
contracts to those proposals offering the best-value considering three
evaluation factors listed in descending order of importance as follows:
a. The soundness, technical merit, and innovation of the proposed
approach and its incremental progress toward topic or subtopic solution.
b. The qualifications of the proposed principal/key investigators,
supporting staff, and consultants. Qualifications include not only the
ability to perform the research and development but also the ability to
commercialize the results.
c. The potential for commercial (Government or private sector)
application and the benefits expected to accrue from this
commercialization as assessed utilizing the criteria in Section 4.4.[5]
Solicitation at 11.
The Army received 27 proposals, including Fantastic Data's, for this
topic. The proposals underwent a three-tier evaluation process. In tier 1,
the proposals were evaluated by a three-member technical evaluation team,
who possessed scientific and technical knowledge in the topic area and who
rated proposals based upon a 100-point scale.[6] Based on this evaluation,
the "best" proposals were forwarded for a second level of review (Tier 2).
Here, four of the proposals, not including Fantastic Data's, were
forwarded for Tier-2 evaluation, and ultimately only one proposal
submitted by Information Systems Technologies, Inc. (ISTI), was selected
for funding (Tier 3). Fantastic Data's proposal was ranked 13^th of the
proposals received with total technical scores of 63, 46, and 56,
respectively, from each of the evaluators.
Fantastic Data challenges the agency's evaluation of its proposal. Where
an agency is conducting an SBIR procurement, it has substantial discretion
to determine which proposals it will fund. RDAS Corp., B-294848, Dec. 23,
2004, 2004 CPD para. 253 at 2. In light of this discretion, our review of
an SBIR procurement is limited to determining whether the agency violated
any applicable regulations or solicitation provisions, or acted in bad
faith. Id. Based on our review of the record, we have no basis to conclude
that the agency acted improperly in deciding not to select Fantastic
Data's proposal for funding.
The record shows that Fantastic Data's proposal was downgraded under the
first factor primarily because it relied on its own previously developed
sensor network protocol, which the evaluators regarded as innovative, but
not necessarily flexible enough to accommodate future developments related
to data fusion and algorithms outside of its proposed approach.[7] See
Agency Report, Tab D, Evaluation Reports. The evaluators believed that
Fantastic Data's proposed approach, based on its own sensor network
protocol, did not adequately recognize the need to develop an architecture
that provided more flexibility to account for future developments in
sensor networks and "data fusion approaches"; the proposal's focus did not
demonstrate maximum flexibility on how "the system will meet a wide range
of potential data fusion solutions and sensor network architectures,"
including node functionality within the network or algorithm
experimentation. See id.; Supplemental Agency Report at 6-8. While the
protester, taking various evaluators' comments out of context, asserts
that this evaluation is unreasonable, it has not shown the agency's basic
concerns about the potential lack of flexibility of Fantastic Data's
approach were unjustified.
Moreover, the record shows with regard to this factor that the proposal
did not address certain information that the agency considered critical,
such as data acquisition, data time synchronization, ease of porting
algorithms to the systems and maximum flexibility of data fusion/network
design. Although Fantastic Data's proposal focused on the sensor node,
which was an important aspect of the network, the topic contemplated that
the sensor node was to be just one aspect of the flexible wireless
acoustic network system architecture, which also included such features as
the central processing console, and gateway level data fusion
functionality (which were addressed in greater detail in the much more
highly rated awardee's proposal). Fantastic Data does not show that its
proposal contains this information or that its proposal should not have
been downgraded because it lacked this information.
Under the second factor, for which Fantastic Data's proposal received
evaluator scores of 23, 23, and 21 points, respectively, out of
30 possible points, the protester focuses on some weaknesses identified by
the evaluators which appear to be unrelated to what was supposed to be
evaluated under this factor. However, the record shows other strengths and
weaknesses identified by the evaluators, not challenged by the protester,
which are germane to this factor.
Similarly, with regard to the third factor, for which Fantastic Data's
proposal received evaluator scores of 12, 8, and 12 points, respectively,
out of 20 possible points, the protester again asserts that various
weaknesses identified under the third factor were not related to that
factor. Based on our review, we do not agree with the protester that the
referenced weaknesses are unrelated to that factor, and note that the
record contains other identified strengths and weaknesses identified by
the evaluators that are clearly germane to the factor.
Nevertheless, even if Fantastic Data's proposal received a perfect score
from each evaluator under both the second and third factors, its total
score would have been significantly less than the scores awarded to any of
the four proposals forwarded for tier 2 evaluation.[8] Thus, there was no
possibility that Fantastic Data was prejudiced, even if the agency's
evaluation of its proposal under the second and third factors was
unreasonable. See Shah & Assocs., B-257405, Sept. 30, 1994, 94-2 CPD para.
123 at 6.
Finally, Fantastic Data argues that the awardee's proposal should have
been rejected because the awardee's key investigator did not meet the
requirement in the solicitation that this individual be a full-time
employee of the SBIR company. Fantastic also argues that the evaluators
were biased in favor of the awardee, and therefore against Fantastic Data.
Fantastic Data is not an interested party eligible to protest the
evaluation of the ISTI's proposal. Under our Bid Protest Regulations, 4
C.F.R. sect. 21.0(a) (2006), only an "interested party" may protest a
federal procurement. That is, a protester must be an actual or prospective
supplier whose direct economic interest would be affected by the award of
a contract or the failure to award a contract. Determining whether a party
is interested involves consideration of a variety of factors, including
the nature of issues raised, the benefit or relief sought by the
protester, and the party's status in relation to the procurement. Four
Winds Servs., Inc., B-280714, Aug. 28, 1998, 98-2 CPD para. 57. A
protester is not an interested party where it would not be in line for
contract award were its protest to be sustained. Id. Since there has been
no challenge to the intervening proposals, including those which were
included in the tier 2 evaluation. These firms would precede the
protester's in eligibility for award under this solicitation and,
therefore, the protester lacks the direct economic interest required to
maintain a protest of the award to ISTI.
With regard to Fantastic Data's allegations of bias in the evaluation of
its proposal, we note that government officials are presumed to act in
good faith, and a protester's claim that contracting officials were
motivated by bias of bad faith must be supported by convincing proof; our
Office will not attribute unfair or prejudicial motives to procurement
officials on the basis of inference or supposition. Shinwa Elecs.,
B-290606 et al., Sept. 3, 2002, 2002 CPD para. 154 at 5 n.6. Here, we find
no evidence of bias towards Fantastic Data; to the contrary, the record
shows, as discussed above, that Fantastic Data's proposal was not funded
because it was not among the highest ranked proposals.
The protest is denied.
Gary L. Kepplinger
General Counsel
------------------------
[1] Data fusion also known as sensor fusion is the combining of sensory
data or data derived from sensory data from disparate sources such that
the resulting information is in some sense better than would be possible
when these sources were used individually. See
http://en.wikipedia.org/wiki/Sensor_data_fusion.
[2] The military utilizes acoustic signal processing to accurately track
targets, such as gunfire locations. See Agency Report, Tab B, Topic
A06-039, at 1.
[3] Sensor nodes are devices using sensors to monitor physical or
environmental conditions, such as temperature, sound, vibration,
temperature, motion or pollutants at different locations. In addition to
one or more sensors, each node in a sensor network is typically equipped
with a radio transceiver or other wireless communications device, a small
microcontroller, and an energy source, usually a battery. See
http://en.wikipedia.org/wiki/Sensor_network. The Army uses sensors and
information processing to obtain, among other things, quick and accurate
information about troop location, target location, and battlefield
conditions. See Agency Report at 2.
[4] Mesh networking is a way to route data, voice and instructions between
nodes, which allows for continuous connections and reconfiguration around
broken or blocked paths by "hopping" from node to node until the
destination is reached. See http://en.wikipedia.org/wiki/Mesh_networking.
ZigBee is the name of a specification for a suite of high level
communication protocols using low-power digital radios based on the IEEE
[Institute of Electrical and Electronics Engineers] 802.15.4 standard for
wireless personal area networks. See http://en.wikipedia.org/wiki/ZigBee.
[5] Paragraph 4.4 provided among other things that commercial potential
would be assessed based on the proposer's commercialization strategy and
the proposer's record of commercializing its prior SBIR projects.
Solicitation at 12.
[6] The first factor was worth 50 points, the second factor was worth 30
points, and the third factor was worth 20 points.
[7] Fantastic Data's proposal received evaluator scores of 28, 15, and 23
points, respectively, out of 50 possible points for the first factor.
[8] The four proposals forwarded for tier 2 evaluation received a total
average score of 90 points or higher based on the evaluation of the three
evaluators. In contrast, Fantastic Data's proposal would have scores of
82, 61, and 73, respectively, if it had received perfect scores under
these two evaluation factors.