[House Hearing, 113 Congress]
[From the U.S. Government Publishing Office]



 
   THE ARIZONA BORDER SURVEILLANCE TECHNOLOGY PLAN AND ITS IMPACT ON 
                            BORDER SECURITY

=======================================================================

                                HEARING

                               before the

                         SUBCOMMITTEE ON BORDER
                         AND MARITIME SECURITY

                                 of the

                     COMMITTEE ON HOMELAND SECURITY
                        HOUSE OF REPRESENTATIVES

                    ONE HUNDRED THIRTEENTH CONGRESS

                             SECOND SESSION

                               __________

                             MARCH 12, 2014

                               __________

                           Serial No. 113-55

                               __________

       Printed for the use of the Committee on Homeland Security
                                     

[GRAPHIC] [TIFF OMITTED] TONGRESS.#13


                                     

      Available via the World Wide Web: http://www.gpo.gov/fdsys/


                                 ______

                   U.S. GOVERNMENT PRINTING OFFICE 
88-172                     WASHINGTON : 2014
____________________________________________________________________________ 
For sale by the Superintendent of Documents, U.S. Government Printing Office, 
http://bookstore.gpo.gov. For more information, contact the GPO Customer Contact Center, U.S. Government Printing Office. Phone 202�09512�091800, or 866�09512�091800 (toll-free). E-mail, [email protected]  

                     COMMITTEE ON HOMELAND SECURITY

                   Michael T. McCaul, Texas, Chairman
Lamar Smith, Texas                   Bennie G. Thompson, Mississippi
Peter T. King, New York              Loretta Sanchez, California
Mike Rogers, Alabama                 Sheila Jackson Lee, Texas
Paul C. Broun, Georgia               Yvette D. Clarke, New York
Candice S. Miller, Michigan, Vice    Brian Higgins, New York
    Chair                            Cedric L. Richmond, Louisiana
Patrick Meehan, Pennsylvania         William R. Keating, Massachusetts
Jeff Duncan, South Carolina          Ron Barber, Arizona
Tom Marino, Pennsylvania             Dondald M. Payne, Jr., New Jersey
Jason Chaffetz, Utah                 Beto O'Rourke, Texas
Steven M. Palazzo, Mississippi       Tulsi Gabbard, Hawaii
Lou Barletta, Pennsylvania           Filemon Vela, Texas
Richard Hudson, North Carolina       Steven A. Horsford, Nevada
Steve Daines, Montana                Eric Swalwell, California
Susan W. Brooks, Indiana
Scott Perry, Pennsylvania
Mark Sanford, South Carolina
Vacancy
                        Vacancy, Staff Director
          Michael Geffroy, Deputy Staff Director/Chief Counsel
                    Michael S. Twinchek, Chief Clerk
                I. Lanier Avant, Minority Staff Director
                                 ------                                

              SUBCOMMITTEE ON BORDER AND MARITIME SECURITY

                Candice S. Miller, Michigan, Chairwoman
Jeff Duncan, South Carolina          Sheila Jackson Lee, Texas
Tom Marino, Pennsylvania             Loretta Sanchez, California
Steven M. Palazzo, Mississippi       Beto O'Rourke, Texas
Lou Barletta, Pennsylvania           Tulsi Gabbard, Hawaii
Vacancy                              Bennie G. Thompson, Mississippi 
Michael T. McCaul, Texas (Ex             (Ex Officio)
    Officio)
            Paul L. Anstine, II, Subcommittee Staff Director
                   Deborah Jordan, Subcommittee Clerk
         Alison Northrop, Minority Subcommittee Staff Director


                            C O N T E N T S

                              ----------                              
                                                                   Page

                               Statements

The Honorable Candice S. Miller, a Representative in Congress 
  From the State of Michigan, and Chairwoman, Subcommittee on 
  Border and Maritime Security...................................     1
The Honorable Sheila Jackson Lee, a Representative in Congress 
  From the State of Texas, and Ranking Member, Subcommittee on 
  Border and Maritime Security...................................     3

                               Witnesses

Mr. Mark Borkowski, Assistant Commissioner, Office of Technology 
  Innovation and Acquisition, Customs and Border Protection, U.S. 
  Department of Homeland Security:
  Oral Statement.................................................     5
  Prepared Statement.............................................     7
Ms. Rebecca Gambler, Director, Homeland Security and Justice 
  Issues, U.S. Government Accountability Office:
  Oral Statement.................................................    10
  Prepared Statement.............................................    12


   THE ARIZONA BORDER SURVEILLANCE TECHNOLOGY PLAN AND ITS IMPACT ON 
                            BORDER SECURITY

                              ----------                              


                       Wednesday, March 12, 2014

             U.S. House of Representatives,
      Subcommittee on Border and Maritime Security,
                            Committee on Homeland Security,
                                                    Washington, DC.
    The subcommittee met, pursuant to call, at 10:06 a.m., in 
Room 311, Cannon House Office Building, Hon. Candice S. Miller 
[Chairwoman of the subcommittee] presiding.
    Present: Representatives Miller, Duncan, Jackson Lee, 
O'Rourke, and Gabbard.
    Mrs. Miller. The Committee on Homeland Security, our 
Subcommittee on Border and Maritime Security will come to 
order.
    The subcommittee is meeting today to examine the CBP's 
border technology plan and its impact on securing our Nation's 
borders, and we are very pleased today to be joined by 
Assistant Commissioner, Mark Borkowski, again.
    Welcome back to the committee.
    He is from the Office of Technology Innovation and 
Acquisition at the U.S. Customs and Border Protection.
    Rebecca Gambler, we welcome you back to the committee as 
well.
    Rebecca is the director of homeland security and justice 
issues with the Governmental Accounting Office, and I will more 
formally introduce them after we do our opening statements.
    Technology has really been an integral part of the proposed 
solution to secure the vast and rugged terrain of the Southwest 
Border for a long time, and it is one part of an overall set of 
solutions that must include manpower, intelligence, and where 
appropriate, infrastructure.
    Unfortunately, a series of miscues and missteps has plagued 
Customs and Border Protection's previous efforts to produce a 
large-scale border security solution.
    The prior program known as SBInet was the subject of many 
criticisms from the GAO and the Congress before then-Secretary 
Napolitano canceled the program actually in 2011, and our goal 
in holding today's hearing is to ensure that we are finally on 
the path to success after a number of false starts.
    Hundreds of millions of dollars were spent for SBInet. We 
have a grand total of 53 miles of the border under surveillance 
to show for it. That is just 53 miles out of nearly 2,000, so I 
don't think any of us can say that that is an acceptable 
outcome.
    I am sure there is enough blame to go around for that, but 
we are not here today to assign blame. We are here today to 
make sure that moving forward the American people get the 
border security that they need, that they want, that they have 
paid for, certainly that they deserve.
    In this time of limited budgets we cannot repeat the 
mistakes of the past and this subcommittee will hold the CBP 
accountable to ensure that the largest and most expensive part 
of the Arizona Border Surveillance Technology Plan, the 
integrated fixed towers, is on track.
    Years of delay have not inspired Congressional confidence 
in the IFT project or the larger Arizona Border Surveillance 
Technology Plan. According to the GAO report said to be 
released today, some of the smaller-scale purchases do not meet 
the needs of the Border Patrol and others have been contracting 
challenges.
    More troubling, CBP did not concur with several of GAO's 
recommendations when it comes to testing the IFT and the 
integration of the smaller-scale technologies into a master 
schedule, and I certainly hope to hear more of the rationale 
for that non-concurrence.
    On a very positive note, I was certainly pleased to see 
that the Department did release the $145 million award for the 
IFT contract last year--excuse me, last week--but for many of 
us in Congress, this is a project that has certainly taken far 
too long.
    It has been 3 years since the cancellation of SBInet. We 
may not see additional border security capabilities come on-
line until later this summer. To that end, we are troubled as 
well by the current roll-out time line that will deploy eight 
towers fairly rapidly in the Nogales area, but then we are 
going to have a gap of time before other towers are deployed.
    Mr. Borkowski is shaking his head, so I know he is going to 
be addressing these concerns, and I appreciate that. I 
certainly would encourage the Department to move quickly to 
resolve any outstanding impediments to the deployment of this 
technology quickly consistent with the operational needs of the 
men and women of the Border Patrol.
    It has taken several years to get technology deployed to 
Arizona and in that time the threat has shifted actually to 
south Texas. Our procurement process has to be more agile and 
quicker otherwise by the time we deploy a solution, the threat 
may have moved elsewhere.
    I also want to work with the Office of Technology 
Innovation and Analysis to help bring effective technology to 
the border faster. We should look to the Department of Defense 
and some of the novel models that they have used to solve this 
very challenge.
    Our committee and this subcommittee in particular has a 
vested interest in securing the border. We have worked on a 
bipartisan basis to pass legislation that defines operational 
control as a 90 percent effectiveness rate and as for a 
comprehensive border security technology plan so that the 
Department is not putting technology on the border in an ad hoc 
way.
    Each piece of technology that we put on the border should 
align with the Border Patrol's operational needs and must 
support the twin goals of increasing our situational awareness 
and effectiveness in keeping those that would do us harm out of 
our country.
    The development of border security metrics will help with 
this effort because without metrics there is no way to measure 
success or failure. Congress and the American people have to 
know what increase in security we are achieving with their 
taxpayer dollars.
    GAO recommended that the Department work toward a better 
understanding of how technology contributes to border security 
and we certainly all have a very keen interest in that effort.
    We certainly look forward to hearing how the Department has 
learned the lessons from previous failures and assurances that 
the taxpayers are getting tangible, measurable border security 
from the Arizona Border Surveillance Technology Plan.
    At this time, the Chairwoman would recognize our Ranking 
Member of the subcommittee, the gentlelady from Texas, Ms. 
Jackson Lee, for any statements she may have.
    Ms. Jackson Lee. Madam Chairwoman, thank you so very much, 
and together we are concerned about this technology as we have 
been discussing technology throughout this session of Congress.
    I want to acknowledge the gentleman from Texas, Mr. 
O'Rourke, and the gentlelady from Hawaii, Ms. Gabbard, present 
for this hearing. I want to thank Chairwoman Miller for holding 
today's hearing on the Arizona Border Surveillance Technology 
Plan.
    As a senior Member of both this committee and the Judiciary 
Committee and as a Member from a border State, I have closely 
followed DHS' efforts to deploy much-needed technology and 
resources to our Southwest Border.
    Unfortunately, DHS has a poor track record with its major 
border security technology acquisitions.
    More recently, the SBInet program was canceled by Secretary 
Janet Neapolitan in 2011 after delivering only 53 miles of 
border security technology in Arizona at a cost to the American 
people of about $1 billion.
    That is a mouthful, and I hope however that that does not 
alter or dismiss the value of technology and the reality that 
technology can really work.
    Now 3 years later, CBP is moving forward with the Arizona 
Border Surveillance Technology Plan which is intended to 
provide additional border security capability in Arizona. I 
would take note of Congressman Barber and former Congresswoman 
Giffords who raised this issue continuously so I know the 
concept is important.
    Already though, there may be some cause for concern. The 
Government Accountability Office released a report today that 
CBP is not following all best practices and DHS guidance for 
acquisition management with this new program.
    Given the challenging nature of these kinds of acquisitions 
and the limited staffing resources CBP has to carry them out, 
it is imperative that the agency follows all guidelines to 
minimize risks to the plan and get a successful result.
    I hope to hear in detail from our GAO witness today about 
what the most pressing challenges are with respect to 
scheduling, cost estimates, testing, and performance metrics 
for the plan.
    I also hope our CBP witness will provide convincing answers 
about how they are addressing these challenges to prevent a 
repeat of the problems that ultimately undermine SBInet.
    We simply cannot afford to spend another $1 billion on 
border security technology that fails to deliver as promised.
    Madam Chairwoman, as I even say that amount I get a sense 
of fear for going down that route again. Certainly, the Border 
Patrol must have resources that will meet its needs to address 
the ever-changing threats it faces among our borders and as 
well those threats that impact the American people.
    With that in mind, I am particularly interested in hearing 
from CBP about how it intends to ensure the plan technology 
will meet the Border Patrol's needs especially given scheduling 
delays that have occurred and the dynamic in nature of the 
border security mission.
    I believe collectively Chairwoman Miller and myself have 
been a very supportive team of CBP. We have worked to ensure 
funding and I would hope that this hearing would not be 
perceived as accusations against CBP but only instructive 
collaborative efforts to make what we all want and that is a 
safer homeland.
    Specifically it is my understanding, the chief of the 
Border Patrol recently communicated that his agency no longer 
needs as many integrated fixed towers in Arizona and instead 
requires more mobile technology to deploy to increasingly 
problematic tech areas along the border in South Texas.
    Having been to the border, both borders, but in South Texas 
particularly at night when we first began to give mobile 
equipment to CBP I can assure you that it is a reality of how 
important that equipment is.
    I hope that the needs of the Border Patrol and the 
requirements of their mission are always being considered 
throughout this process, and I look forward to hearing from our 
CBP witness about how this recent request will be resolved.
    Again, I thank the witnesses for joining us today, and I 
look forward to a robust discussion about how we can ensure the 
Arizona border technology plan succeeds where its predecessors 
did not.
    Madam Chairwoman, just a moment as I conclude.
    All of us have had our eyes poised, if you will, on the 
Malaysian air liner tragically lost; our sympathy to those 
families who are both mourning and waiting and now in a very 
confused state.
    But I know that everyone was struck by the issue of the 
false passports that two individuals managed to get on. We have 
no conclusion. We have speculation. We do not know.
    But I know that, Madam Chairwoman, all of us on Homeland 
Security were aghast that most countries are not using the 
database check, and I believe that because we are a border 
security subcommittee that it would it be very helpful for us 
to leap into that.
    I know that there is jurisdiction, cross-jurisdiction with 
Judiciary, but I would love to request a hearing on that 
because it is brought to our attention. A tragedy has occurred. 
We cannot suggest or we do not have the facts as to what might 
have provoked that, but I believe that that is a sore that is 
waiting to spread and creating a devastating degree of 
nonsecurity around the world, so I hope that we will have a 
discussion about that and as we do many other items of 
importance.
    With that, Madam Chairwoman, I yield back.
    Mrs. Miller. I thank the gentlelady. I certainly thank the 
gentlelady for her comments about the Malaysia flight, and as 
you mentioned, we all share shock and certainly our prayers 
going out to the families of those that are missing, where no 
one knows where they are or what has happened.
    Certainly that is one of the great mysteries I think that 
any of us have ever seen, but it is interesting you mention 
about the passports because we have already been talking to 
staff about putting together sort-of looking at what is the 
entire passport issue and the other countries and how they are 
in compliance or noncompliance with looking at some of these 
things.
    Now it appears--again speculation--we are only talking 
about what we are reading in the papers I guess, that these two 
with the stolen passports were not on the watch list, but we 
will see where that goes, but I certainly appreciate those 
comments and we do, as I think this committee, subcommittee, 
and our full committee needs to take a look at that entire 
issue as well.
    In regards to the subject at hand, I would also just remind 
other Members of the committee that opening statements might be 
submitted for the record and we are certainly pleased to have 
as I mentioned two very distinguished witnesses to speak to our 
subcommittee today.
    Mr. Mark Borkowski became the assistant commissioner for 
the Office of Technology Innovation and Acquisition at the 
United States Custom and Border Protection in July 2010.
    In this role he is responsible for ensuring technology 
efforts are properly focused, on mission, and well-integrated 
across CBP. Prior to his appointment as assistant commissioner, 
Mr. Borkowski was the executive director for the Secure Border 
Initiative.
    Ms. Rebecca Gambler--I welcome back again to the 
committee--is an acting director in the U.S. Government 
Accountability's Office of Homeland Security and Justice team. 
She leads GAO's work on the border security and immigration 
issues, and their full written statements will appear in the 
record.
    At this point, the Chairwoman now recognizes Mr. Borkowski 
for his testimony.

STATEMENT OF MARK BORKOWSKI, ASSISTANT COMMISSIONER, OFFICE OF 
   TECHNOLOGY INNOVATION AND ACQUISITION, CUSTOMS AND BORDER 
        PROTECTION, U.S. DEPARTMENT OF HOMELAND SECURITY

    Mr. Borkowski. Well, thank you.
    Chairwoman Miller, Ranking Member Jackson Lee, Members of 
the committee, thank you very much for the opportunity to come 
and describe to you where we have been, where we are, where we 
think we are going.
    You laid an agenda quite a bit of material that I will do 
my best to cover at least with some introductory things, and I 
will look forward to the questions.
    I also do want to note that we certainly have appreciated 
the support of this committee. We recognize that it has 
required a tremendous amount of tolerance and forbearance on 
your part as we have gone through the last 2 to 3 years of 
trying to get ourselves to the point where we finally have 
contracts awarded, so we appreciate the support.
    We understand it has been frustrating. We understand that 
you are at least as frustrated as we have been by the delays, 
but I will try to take you through where we are, how we got 
there, and why we think maybe we are about to turn the corner.
    You both mentioned SBInet and the history of SBInet. SBInet 
taught us a lot of things as we all well know. It taught us how 
not to buy things for the Department of Homeland Security, but 
as Ms. Jackson Lee mentioned, it also taught us that technology 
does some very good things.
    Where we did deploy the two SBInet deployments, it had a 
near-immediate effect on our ability to deal with those areas 
of the border, and although I couldn't give you something more 
quantitative than this, as we fly over those areas today, a 
year or 2 years after they have been deployed, the activity in 
those areas is down tremendously.
    So technology not only supports the actual gaining of 
control of an area but it tends to stand as a sentry afterwards 
and maintain that reduced activity which then gives us more 
flexibility to move Border Patrol agents and other technology, 
so we have learned that technology is a significant factor when 
it is deployed properly. So we all share the interest in 
getting the technology out there.
    But some of the lessons that we learned from SBInet are 
lessons that we tried to apply into the current Arizona 
technology plan and there are a couple of key lessons that I 
want to really emphasize because they get to the point as to 
why we non-concurred with a couple of the GAO recommendations.
    Put simply, in our view, the GAO recommendations are 
driving us back to what we don't want to do, an SBInet. Put 
simply, that is the issue.
    What do I mean by that? SBInet was a system development. 
Now that is a term of art in my business and I have struggled 
with--because to me the difference between SBInet and IFT is so 
clear because I have done this--I have trouble explaining it--
but think of it this way.
    If you went to buy a car you would have a choice. You could 
say, ``I am going to pretend there are no cars to buy, and I am 
just going to go ask somebody to build me a car from scratch,'' 
or you could go to a car dealer and pick what is off the lot.
    We normally buy things in the Government unfortunately that 
first way as if there is nothing on the lot, and you can 
imagine what the costs, but it turns out in this business there 
is a lot of stuff on the lot and so we went to buy it on the 
lot. That is a big deal and we disaggregated pieces. We didn't 
tie it all together. We broke them apart.
    It turns out that that has had a significant effect because 
in almost everything we have bought we have bought it for lower 
than we anticipated cost. I attribute it to the non-
developmental nature and the fact that we are willing to be 
flexible about the technical definition of requirements and I 
would be prepared to discuss that in more detail as we go, to 
the point where we have the freed-up resources that have 
allowed us to do things like fly Aero stats over South Texas.
    When you talk about putting things in Texas we have started 
to do that with money that was freed up from savings generated 
by the strategy in the Arizona technology development plan.
    So we think we are making progress. As you mentioned, we 
awarded the IFT contract last week. I will tell you compared to 
the initial estimate, we saved 75 percent, and I know that 
sounds unbelievable and we are going to study that as to how 
did that happen.
    How could it be 75 percent savings, but as we have started 
to dig in, we think we are learning some lessons that may apply 
and provide that kind of experience in the future if we change 
our strategies.
    So in that context, as we have been buying these things, we 
have also been trying to strengthen our competency, the 
technical detail about how you buy these things and that is 
where the GAO has come in.
    While it is true that the GAO report continues to note 
areas of weakness, I would also point out that if you go back 
over the last few years I believe if you read these reports 
over the last few years you are going to see an improving 
trend.
    I believe that the issues that we are finding are becoming 
increasingly business frankly arcane, and we really are to the 
point where we have to make decisions about trading off cost 
versus perfection.
    So while we agree with the recommendations generally, where 
those recommendations tell us to tie things together that 
should be broken apart, which is what the IMS recommendation 
does, or tells us to build a new car rather than buying a car 
on the lot, which is what the test and evaluation 
recommendation would tend us to, we would object to those.
    We think it is a better approach to take this new method of 
buying--let's try that for a while and then generate lessons 
learned from that. So that is kind of where we are at this 
point. I look forward to your questions, and again, I very much 
appreciate the committee's support and forbearance to this 
point.
    [The prepared statement of Mr. Borkowski follows:]

                  Prepared Statement of Mark Borkowski
                             March 12, 2014

    Chairwoman Miller, Ranking Member Jackson Lee, and distinguished 
Members of the subcommittee, it is a pleasure to appear before you 
today to discuss the status of U.S. Customs and Border Protection's 
(CBP) border security technology programs in Arizona, and to reflect on 
the most recent Government Accountability Office (GAO) report about the 
management of those programs.
    I appreciate the partnership and support we have received from 
Congress, this subcommittee, and your staff, whose commitment to the 
security of the American people has enabled the continued deployment of 
key border security technologies, even in the face of significant 
challenges. I am confident that our collective efforts will continue to 
result in a better-managed and more secure border.
    This subcommittee is familiar with the outcome of CBP's SBInet 
program, an earlier component of the Department of Homeland Security's 
(DHS) Secure Border Initiative (SBI) that was designed as a 
comprehensive and integrated technology program to provide persistent 
surveillance across the northern and southern land borders of the 
United States, starting with the border of Mexico. The program 
experienced significant schedule delays and cost overruns because it 
did not allow necessary flexibility to adapt to differing needs in the 
various regions of the border. SBInet eventually delivered systems to 
two Areas of Responsibility in Arizona that continue to operate 
successfully. Nevertheless, DHS cancelled SBI on January 14, 2011, 
because it was too costly and the idea of one, all-encompassing program 
was unnecessarily complex for border technology.
    Since 2011, we have learned from the issues identified in from the 
SBInet approach and moved away from an all-encompassing SBInet concept. 
Instead, DHS and CBP have approached our border technology requirements 
in more manageable pieces tailored to specific regions on the border. 
Working closely with the Border Patrol to develop requirements, we 
created a menu of different, sophisticated technology systems, ranging 
from small to large, simple to complex. For Arizona, we selected 
systems from the menu and tailored those technology solutions based on 
realistic capabilities of current technologies and the operational 
needs of particular areas. We then created detailed acquisition plans 
for each of the technologies on the menu and have been in the process 
of buying and deploying them for the last few years. We refer to this 
approach as the Arizona Technology Plan (ATP).
    ATP or ``the Plan'' is not a program as traditionally defined 
within the acquisition business. Instead, it is a set of programs that, 
taken together, will provide what we believe is the optimal set of 
systems for our current operational needs. One key point is that the 
Plan is not a so-called ``system of systems.'' In fact, our acquisition 
strategy moved intentionally away from the ``system of systems'' 
concept because we had learned from our SBInet experience that this 
approach was unnecessarily complex and costly.
    Another change in CBP's ATP acquisition strategy based on lessons 
learned from SBInet, is a shift from pursuing what is known as ``system 
development'' toward a concept of leveraging ``non-developmental 
items.'' Put simply, system development involves the creation of a 
system that does not currently exist. System development is a very 
disciplined and exhaustive process that requires engineering design, 
analysis to compare the design to requirements, comprehensive testing, 
and eventually deployment and operation. System development is an 
appropriate acquisition approach when: (1) The requirements are 
understood with high confidence, (2) there is limited flexibility to 
relax the requirements, and (3) no existing system meets the 
requirements. However, system development is costly, challenging, and 
often risky--more so when the conditions that would support system 
development do not exist. In the case of SBInet, we did not have a 
highly confident understanding of the requirements, or a solid 
justification for why our requirements were inflexible. Therefore, it 
was unclear whether existing systems would be adequate for our needs. 
Based on lessons learned from SBInet, we explicitly and intentionally 
rejected system development as our approach for the programs within the 
Plan.
    For the programs under the ATP, we embarked on a non-developmental 
item (NDI) approach because after conducting extensive market research, 
we had high confidence that technology systems already existed that 
could provide most, if not all, of the capabilities we felt were 
required. CBP's Office of Technology Innovation and Acquisition (OTIA), 
which I oversee, worked collaboratively with the Border Patrol to 
develop the technical requirements. We also created the flexibility to 
trade those requirements against cost. Under this NDI strategy, we 
created an opportunity to do things like buy a system that met 90 
percent of our interests at 50 percent of the cost, as compared to a 
system that might have met 100 percent of our interests but at twice 
the cost.

               STATUS OF ARIZONA TECHNOLOGY PLAN PROGRAMS

    While acquisition of the programs within the Plan is admittedly 
behind schedule I believe our actions have been prudent and have 
actually resulted in some very positive outcomes. In short, we elected 
to trade schedule for higher likelihood of success in the ultimate 
deployments of the NDI technologies and to take advantage of 
opportunities to reduce costs.
    Using the NDI approach, most of the programs within the Plan are on 
contract and many have already been deployed, including: Agent Portable 
Surveillance Systems (APSS); Thermal Imaging Devices; Underground 
Sensors (UGS); and some Mobile Video Surveillance Systems (MSC). 
Although it is too early to declare complete success, the early 
indications of the ATP acquisition strategy are quite positive and, in 
some cases, far exceed our expectations.
    For example, the most complex and costly program within the Plan is 
the Integrated Fixed Tower (IFT) program. This program, ostensibly, 
looks something like the old SBInet program. As such, it is often 
treated as if it were SBInet renamed. However, IFT is not SBInet. It is 
an NDI program, and it is a narrowly-tailored solution to select parts 
of the border.
    Early external assessments of the program questioned whether NDI 
systems for IFT existed and whether CBP's program cost estimates were 
too low. While the specific numbers are still sensitive, I can report 
that we received far more proposals from industry for the IFT contract 
than we anticipated and, for that matter, more than I have ever seen 
for this type of procurement during my roughly 30 years in this 
business. The proposals were quite credible, and the sheer number 
rebuts any doubts about NDI availability. Also, almost every program in 
the Plan has been contracted at less than our initial estimates--often 
much less. The IFT contract, for example, came in at a savings 
approaching 75 percent of our initial estimate. Although we will likely 
have routine changes in the contract over time that will add slightly 
to the final cost, a 75 percent cost savings leaves a lot of room for 
those routine changes. It is also important to note that, because these 
are NDI systems, we have been able to use firm fixed-price contracting, 
which reduces the risk to the Government of substantial and 
uncontrolled cost growth, compared to cost reimbursable contracts for 
system developments like SBInet.
    We attribute these positive indications to our acquisition 
strategy, our thorough market research, our staff's hard work, our 
willingness to trade schedule for risk reduction, and our on-going 
dialogue with industry. DHS and CBP acknowledged that we needed to do 
things differently if we wanted a better result from past acquisition 
failures. In a sense, our approach to the Plan was an experiment. While 
not without risk, we believe the plan represents the most viable option 
for a successful acquisition process, one that might prove to be a 
useful model going forward. As I indicated, we are quite encouraged by 
what we have seen so far.
    The cost savings alone have already had a major impact for us. We 
have harvested those savings to do many of the things that this 
subcommittee has advocated. For instance, we have worked closely with 
the Department of Defense (DoD) to receive or borrow their 
technologies. We currently have three DoD aerostats flying over the 
Border Patrol's Rio Grande Valley Sector as part of an extended Field 
Deployment Evaluation. While undergoing evaluation, the systems 
concurrently support real-world operations and boost technological 
capabilities in a high-priority area of the border. We are able to fund 
this exercise, as well as a number of other notable efforts, because of 
the cost savings incurred as a result of our Arizona Technology Plan 
strategy.

                          GAO RECOMMENDATIONS

    CBP's border security efforts are critically important, and we 
appreciate GAO's engagement with CBP's technology acquisition 
activities from the SBInet days through the present. GAO has been 
consistently objective and has always been very open to our thoughts 
and opinions. It is important to consider the latest GAO report in the 
context of our history to date. While the recent March 2014 report, 
``Arizona Border Surveillance Technology Plan: Additional Actions 
Needed to Strengthen Management and Assess Effectiveness,'' continues 
to identify some areas of potential weakness and risk, I believe it 
also demonstrates a continuing improvement trend. Piece by piece, we 
are building the program management infrastructure that did not exist 
in the early days of SBInet. The GAO has helped us prioritize our 
efforts over the years and deserves great credit for helping to point 
the way to better performance.
    In the latest report, we concur with many of the GAO 
recommendations because they represent well-established best practices 
for any acquisition program--including the non-developmental programs 
that comprise the Plan. In most of these cases, we are aware of the 
shortcomings highlighted by the GAO. However, we also recognize that, 
we had to prioritize the activities that offered the least risk to our 
success by conducting a cost-benefit analysis. For example, although we 
did not complete formal independent cost estimates for our programs, we 
had substantial data and market research to give us high confidence in 
the conservatism of our life-cycle cost estimates. Similarly, while it 
is true that not all required acquisition documentation was formally 
approved at set times, the documents were virtually final, well-
understood, and complete enough to enable key decisions with little 
risk. Going forward, we will strive to perform better in these areas.
    We have non-concurred with two of the GAO recommendations, mainly 
because they contradict the foundation of the acquisition strategy we 
implemented for the Plan. Each program in the Plan has an Integrated 
Master Schedule (IMS), as required by our policy and practice. However, 
the GAO recommends CBP create an IMS for the Plan, as if the Plan 
itself is a program or ``system of systems.'' As discussed above, CBP 
intentionally designed the Plan not to be a system of systems. It has 
been the separation of the old SBInet program into nearly independent 
and dis-aggregated elements that has, in my view, enabled the positive 
trends we have seen to date. We maintain an appropriate level of 
integration and schedule connection among the programs in the Plan; 
however, the GAO recommendation runs counter to the lessons learned 
from SBInet and risks returning us to an acquisition strategy we 
already know to be high-risk.
    Similarly, the GAO calls for formal Operational Test and Evaluation 
(OT&E), as if the Plan were a system development. As noted above, CBP 
structured the Plan with NDI programs as a result of lessons learned 
from SBInet. Since we are familiar with the technologies, we are 
willing to trade requirements and performance for cost and other 
benefits. We have committed to purchasing, at firm-fixed price, a 
system that will perform to the specifications asserted by the 
contractor. Formal OT&E would create unnecessary bureaucracy, threaten 
the NDI nature of the program by creating a set of requirements that 
may demand system development activities, and compromise the nature of 
the Plan that has already suggested very positive results.
    For example, we will manage IFT as we have done for several of the 
other programs in the Plan. We have worked with the Border Patrol to 
define the kind of operational experience and analysis Border Patrol 
agents believe they need to understand and assess the system 
performance. We have documented this agreement in the Test and 
Evaluation Master Plan. This meets much of the intent of formal OT&E, 
does it without unnecessary bureaucracy, and provides the Border Patrol 
with oversight, control, and data to influence decisions about future 
deployments and potential system upgrades.

                               CONCLUSION

    In short, we concur with the GAO where the recommendations 
represent best practices and risk reduction for acquisitions like the 
Arizona Technology Plan. We do not concur where those recommendations 
are inconsistent with the intentional design of the programs in the 
Plan and where implementation of those recommendations would compromise 
the foundation of the Plan.
    Some have characterized our acquisition approach to the Plan as 
innovative--especially with regard to how it leverages NDI 
opportunities and offers an opportunity to trade-off requirements. 
Innovation in acquisition means we will apply lessons learned, 
experiment with new things, and break new ground. We have a solid 
understanding of where we need to break new ground, and we look forward 
to working with the GAO as we continue our efforts to develop what 
could become a new set of best practices.
    Chairwoman Miller, Ranking Member Jackson Lee, thank you for the 
opportunity to testify today. I look forward to your questions.

    Mrs. Miller. I thank the gentleman very much. I think we do 
have a lot of questions after that.
    At this time, the Chairwoman now recognizes Ms. Gambler.

 STATEMENT OF REBECCA GAMBLER, DIRECTOR, HOMELAND SECURITY AND 
     JUSTICE ISSUES, U.S. GOVERNMENT ACCOUNTABILITY OFFICE

    Ms. Gambler. Good morning, Chairwoman Miller, Ranking 
Member Jackson Lee, and Members of the subcommittee.
    I appreciate the opportunity to testify in today's hearing 
to discuss GAO's work reviewing the status of DHS' Arizona 
Border Surveillance Technology Plan.
    This plan followed DHS' announcement in January 2011 that 
it was canceling further procurements of Secure Border 
Initiative Network Systems referred to as SBInet. DHS announced 
the launch of the secure border initiative in 2005 as a multi-
year, multi-billion-dollar program aimed at securing U.S. 
borders.
    SBInet was intended to include various technologies such as 
fixed sensor towers and tactical infrastructure to create a 
virtual fence along the Southwest Border.
    After a cost of about $1 billion, SBInet systems are now 
operating along 53 miles of Arizona's border.
    The Arizona Border Surveillance Technology Plan includes a 
mix of radars, sensors, and cameras to provide security for the 
remainder of the Arizona border.
    My remarks today will focus on three areas related to CBP's 
management of the plan and its efforts to assess the 
contributions of planned and deployed technologies to border 
security.
    First, we reviewed CBP's schedules and life-cycle cost 
estimates for the plan and its highest-cost programs and we 
compared them against best practices. Those best practices, if 
followed consistently, are designed to help agencies better 
ensure the reliability of their schedules and cost estimates.
    Overall, the schedules and estimates for the plan's 
programs reflected as some but not all best practices. For 
example, we found that the schedules for some of the programs 
were not fully credible because CBP had not identified all 
risks that would be most likely to delay the programs.
    CBP also has not developed an integrated master schedule 
for the plan. Such a schedule could help provide CBP with a 
comprehensive view of the plan and more reliably commit to when 
the plan will be fully implemented.
    We also found that CBP has not independently verify its 
life-cycle cost estimates for two of the plan's programs, the 
integrated fixed towers and remote video surveillance system.
    Second, we reviewed the extent to which CBP followed key 
aspects of DHS' acquisition guidance in managing the plan. CBP 
followed this guidance in some areas, but for the plan's three 
highest cost programs we found that DHS and CBP did not 
consistently approve key acquisition documents when called for 
in DHS' guidance.
    For example, for the integrated fixed towers DHS approved 
four of the six required documents at the time they were to be 
approved but two other documents, the life-cycle cost estimate 
and the test plan were not approved on time based on 
documentation we received from CBP.
    With regard to the test plan this plan calls for CBP to 
conduct limited user testing of the integrated fixed tower to 
determine mission contributions for 30 days at one site along 
the Arizona border, however such an approach is not consistent 
with DHS' acquisition guidance which calls for operational 
tests and evaluation of systems to occur in the environmental 
conditions in which a system will be used.
    Conducting unlimited user tests at one location for 30 days 
could limit the information available to CBP on how the towers 
will perform in other locations and under different 
environmental conditions along the border.
    More robust testing is particularly important in light of 
the previous challenges we identified in testing of SBInet 
systems.
    For example, we previously reported that some SBInet test 
plans were not defined in accordance with the guidance. We 
concluded that not doing effective testing can unnecessarily 
increase the risks of problems going undetected until late in a 
system's life cycle.
    Thus, conducting more robust testing on the integrated 
fixed towers could help CBP better ensure that the towers meet 
Border Patrol's operational needs.
    Finally, DHS had the database through which Border Patrol 
can record whether or not an asset such as a camera assisted in 
an apprehension or seizure. This indicator is referred to by 
the Border Patrol as an asset assist.
    Data on asset assist, if used with other performance 
metrics or indicators, could help CBP assess the contributions 
of surveillance technologies to apprehensions and seizures. 
However Border Patrol does not require agents to record data on 
asset assist, and thus the agency does not have complete data 
to help assess technologies' contributions to border security 
efforts.
    In closing, we have made recommendations to DHS in all of 
these areas to help the Department in its efforts to manage and 
implement the plan. DHS has agreed with some but not all of 
these recommendations and has actions planned or under way to 
address some of them. We will continue to monitor DHS' efforts 
in response to our recommendations.
    This concludes my oral statement. I would be pleased to 
answer any questions that Members may have.
    [The prepared statement of Ms. Gambler follows:]

                 Prepared Statement of Rebecca Gambler
                             March 12, 2014

    Chairwoman Miller, Ranking Member Jackson Lee, and Members of the 
subcommittee: I am pleased to be here today to discuss the findings 
from our March 2014 report, being released today, in which we assessed 
the Department of Homeland Security's (DHS) U.S. Customs and Border 
Protection's (CBP) efforts to develop and implement the Arizona Border 
Surveillance Technology Plan (the Plan).\1\ In recent years, nearly 
half of all annual apprehensions of illegal entrants along the 
Southwest Border with Mexico have occurred along the Arizona border, 
according to DHS data. A top priority for CBP is preventing, detecting, 
and apprehending illegal entrants. In November 2005, DHS announced the 
launch of the Secure Border Initiative (SBI), a multi-year, multi-
billion-dollar program aimed at securing U.S. borders and reducing 
illegal immigration. CBP intended for the SBI Network (SBInet) to 
include technologies such as fixed-sensor towers, a common operating 
picture, and tactical infrastructure to create a ``virtual fence'' 
along the Southwest Border to enhance CBP's capability to detect, 
identify, classify, track, and respond to illegal breaches at and 
between land ports of entry.\2\ At a cost of about $1 billion, in 2010, 
CBP deployed SBInet systems, referred to as Block 1 systems, along the 
53 miles of Arizona's 387-mile border with Mexico that represent one of 
the highest-risk areas for illegal entry attempts. However, in January 
2011, in response to internal and external assessments that dentified 
concerns regarding the performance, cost, and schedule for implementing 
the systems, the Secretary of Homeland Security announced the 
cancellation of further procurements of SBInet systems.\3\
---------------------------------------------------------------------------
    \1\ GAO, Arizona Border Surveillance Technology Plan: Additional 
Actions Needed to Strengthen Management and Assess Effectiveness, GAO-
14-368 (Washington, DC: Mar. 3, 2014).
    \2\ The SBInet fixed-sensor towers were intended to transmit radar 
and camera information into a common operating picture at work stations 
manned at all times by U.S. Border Patrol agents. The SBInet Common 
Operating Picture was intended to provide uniform data through a 
command center environment to Border Patrol agents in the field and all 
DHS agencies, and to be interoperable with the equipment of DHS 
external stakeholders, such as local law enforcement. Tactical 
infrastructure includes pedestrian and vehicle fences, roads, and 
lighting. Ports of entry are officially designated places that provide 
for the arrival at, or departure from, the United States.
    \3\ See, for example, GAO, Secure Border Initiative: DHS Needs to 
Reconsider Its Proposed Investment in Key Technology Program, GAO-10-
340 (Washington, DC: May 5, 2010), and Secure Border Initiative: DHS 
Needs to Address Significant Risks in Delivering Key Technology 
Investment, GAO-08-1086 (Washington, DC: Sept. 22, 2008).
---------------------------------------------------------------------------
    After the cancellation of SBInet in January 2011, CBP developed the 
Plan, which includes a mix of radars, sensors, and cameras to help 
provide security for the remainder of the Arizona border. Under the 
Plan, CBP identified seven programs to be implemented ranging in 
estimated costs from $3 million to about $961 million. The three 
highest-cost programs under the Plan are the Integrated Fixed Tower 
(IFT), Remote Video Surveillance System (RVSS), and Mobile Surveillance 
Capability (MSC), accounting for 97 percent of the Plan's estimated 
cost.\4\ In November 2011, we reported on CBP's development of, and 
estimated life-cycle costs for, implementing the Plan.\5\ Specifically, 
we reported that CBP needed more information for the Plan and its costs 
before proceeding with implementation, and we recommended that CBP, 
among other things, determine the mission benefits to be derived from 
the implementation of the Plan and develop and apply key attributes for 
metrics to assess program implementation, conduct a post implementation 
review and operational assessment of SBInet, and update the cost 
estimate for the Plan using best practices.\6\ DHS concurred with these 
recommendations and has actions under way to address some of them.
---------------------------------------------------------------------------
    \4\ The IFT consists of towers with, among other things, ground 
surveillance radars and surveillance cameras mounted on fixed (that is, 
stationary) towers. The RVSS includes multiple color and infrared 
cameras mounted on monopoles, lattice towers, and buildings and differs 
from the IFT, among other things, in that the RVSS does not include 
radars. The MSC is a stand-alone, truck-mounted suite of radar and 
cameras that provides a display within the cab of the truck.
    \5\ GAO, Arizona Border Surveillance Technology: More Information 
on Plans and Costs Is Needed before Proceeding, GAO-12-22 (Washington, 
DC: Nov. 4, 2011). A Life-cycle Cost Estimate provides an exhaustive 
and structured accounting of all resources and associated cost elements 
required to develop, produce, deploy, and sustain a particular program.
    \6\ Measures and key attributes are generally defined as part of 
the business case in order to explain how they contribute to the 
mission's benefits. See Office of Management and Budget, OMB Circular 
No. A-11, Part 7, Section 300, Planning, Budgeting, Acquisition, and 
Management of Capital Assets (Washington, DC: Executive Office of the 
President, July 2010).
---------------------------------------------------------------------------
    My testimony today is based on and summarizes the key findings of 
our report on the status of the Plan, which was publicly released 
today. \7\ Like the report, my statement will address CBP's efforts to: 
(1) Develop schedules and Life-cycle Cost Estimates for the Plan in 
accordance with best practices, (2) follow key aspects of DHS's 
acquisition management framework in managing the Plan's three highest-
cost programs, and (3) assess the performance of technologies deployed 
under SBInet and identify mission benefits and develop performance 
metrics for surveillance technologies to be deployed under the Plan. To 
conduct work for the March 2014 report, we analyzed DHS and CBP program 
schedules and Life-cycle Cost Estimates and interviewed DHS and CBP 
officials responsible for developing and overseeing schedules and cost 
estimates, including officials from CBP's Office of Technology 
Innovation and Acquisition (OTIA), which manages implementation of the 
Plan. We also analyzed DHS and CBP documents, including DHS Acquisition 
Management Directive 102-01 and its associated DHS Instruction Manual 
102-01-001, program briefing slides, budget documents, Acquisition 
Decision Memorandums, and program risk sheets.\8\ Finally, we analyzed 
performance assessment documentation and metrics used by CBP to 
determine the effectiveness of technologies deployed under SBInet and 
interviewed CBP officials responsible for performance measurement 
activities, and analyzed CBP data on apprehensions, seizures, and asset 
assists from fiscal year 2010 through June 2013 to determine the extent 
to which the data could be used to measure the contributions of SBInet 
technologies in enhancing border security.\9\ We conducted this work in 
accordance with generally accepted Government auditing standards. More 
detailed information on the scope and methodology of our published 
report can be found therein.
---------------------------------------------------------------------------
    \7\ GAO-14-368.
    \8\ DHS Acquisition Management Directive 102-01, Jan. 20, 2010, and 
DHS Instruction Manual 102-01-001, Acquisition Management/Instruction 
Guidebook, Oct. 1, 2011.
    \9\ An asset assist is what happens when a technological asset, 
such as a SBInet surveillance tower, or a non-technological asset, such 
as a canine team, contributes to apprehensions or seizures. In our 
March 2014 report, apprehensions data included individuals arrested and 
identified as deportable aliens, consistent with Border Patrol's 
definition.
---------------------------------------------------------------------------
CBP'S PROGRAM SCHEDULES AND LIFE-CYCLE COST ESTIMATES REFLECT SOME BUT 
                         NOT ALL BEST PRACTICES

    In our March 2014 report, we assessed OTIA's schedules as of March 
2013 for the IFT, RVSS, and MSC programs and found that these program 
schedules addressed some, but not all, best practices for scheduling. 
The Schedule Assessment Guide identifies 10 best practices associated 
with effective scheduling, which are summarized into four 
characteristics of a reliable schedule--comprehensive, well-
constructed, credible, and controlled.\10\ According to our overall 
analysis, OTIA at least partially met the four characteristics of 
reliable schedules for the IFT and RVSS schedules (i.e., satisfied 
about half of the criterion), and partially or minimally met the four 
characteristics for the MSC schedule, as shown in Table 1. For example, 
we reported that the schedule for the IFT program partially met the 
characteristic of being credible in that CBP had performed a schedule 
risk analysis for the program, but the risk analysis was not based on 
any connection between risks and specific activities.
---------------------------------------------------------------------------
    \10\ GAO, GAO Schedule Assessment Guide: Best Practices for Program 
Schedules, GAO-12-120G (exposure draft) (Washington, DC: May 2012). We 
developed this guide through a compilation of best practices that 
Federal cost-estimating organizations and industry use. According to 
this guide, for a schedule to be comprehensive, among other things, the 
schedule should: (1) Capture all activities, as defined in the work 
breakdown structure, (2) reflect what resources are needed to do the 
work, and (3) establish the duration of all activities and have 
specific start and end dates. To be well-constructed, among other 
things, all schedule activities are sequenced in the order that they 
are to be implemented with the most straightforward logic possible. To 
be credible, the schedule should reflect the order of events necessary 
to achieve aggregated products or outcomes, and activities in varying 
levels of the schedule map to one another. Moreover, a schedule risk 
analysis should be conducted to predict a level of confidence in 
meeting the program's completion date. For a schedule to be controlled, 
the schedule should be updated periodically using actual progress and 
logic to realistically forecast dates for program activities, and a 
baseline schedule should be maintained to measure, monitor, and report 
the program's progress.

   TABLE 1.--SUMMARY OF OUR SCHEDULE ASSESSMENTS FOR THE THREE HIGHEST-COST PROGRAMS UNDER THE ARIZONA BORDER
                                          SURVEILLANCE TECHNOLOGY PLAN
----------------------------------------------------------------------------------------------------------------
                                                                               Remote Video          Mobile
               Schedule Characteristic                  Integrated Fixed       Surveillance       Surveillance
                                                             Towers              Systems           Capability
----------------------------------------------------------------------------------------------------------------
Comprehensive.......................................         Partially met      Partially met      Partially met
Well constructed....................................     Substantially met      Partially met      Partially met
Credible............................................         Partially met      Partially met      Minimally met
Controlled..........................................         Partially met      Partially met      Minimally met
----------------------------------------------------------------------------------------------------------------
Source.--GAO analysis of Customs and Border Protection data.
Note.--Not met--CBP provided no evidence that satisfies any of the criterion. Minimally met--CBP provided
  evidence that satisfies a small portion of the criterion. Partially met--CBP provided evidence that satisfies
  about half of the criterion. Substantially met--CBP provided evidence that satisfies a large portion of the
  criterion. Met--CBP provided complete evidence that satisfies the entire criterion.

    We recommended that CBP ensure that scheduling best practices are 
applied to the IFT, RVSS, and MSC schedules. DHS concurred with the 
recommendation and stated that OTIA plans to ensure that scheduling 
best practices are applied as far as practical when updating the three 
programs' schedules.
    Further, in March 2014 we reported that CBP has not developed an 
Integrated Master Schedule for the Plan in accordance with best 
practices. Rather, OTIA has used the separate schedules for each 
individual program (or ``project'') to manage implementation of the 
Plan. OTIA officials stated that an Integrated Master Schedule for the 
overarching Plan is not needed because the Plan contains individual 
acquisition programs as opposed to a plan consisting of seven 
integrated programs. However, collectively these programs are intended 
to provide CBP with a combination of surveillance capabilities to be 
used along the Arizona border with Mexico. Moreover, while the programs 
themselves may be independent of one another, the Plan's resources are 
being shared among the programs.
    OTIA officials stated that when schedules were developed for the 
Plan's programs, they assumed that personnel would be dedicated to work 
on individual programs and not be shared between programs. However, as 
OTIA has initiated and continued work on the Plan's programs, it has 
shared resources such as personnel among the programs, contributing, in 
part, to delays experienced by the programs. According to schedule best 
practices, an Integrated Master Schedule that allows managers to 
monitor all work activities, how long the activities will take, and how 
the activities are related to one another is a critical management tool 
for complex systems that involve the incorporation of a number of 
different projects, such as the Plan.\11\ Thus, we recommended that CBP 
develop an Integrated Master Schedule for the Plan.
---------------------------------------------------------------------------
    \11\ GAO-12-120G.
---------------------------------------------------------------------------
    DHS did not concur with this recommendation. In particular, DHS 
stated that maintaining an Integrated Master Schedule for the Plan 
undermines the DHS-approved implementation strategy for the individual 
programs making up the Plan and that a key element of the Plan has been 
the disaggregation of technology procurements. However, we continue to 
believe that developing an Integrated Master Schedule for the Plan is 
needed. As we reported in March 2014, this recommendation is not 
intended to imply that DHS needs to re-aggregate the Plan's seven 
programs into a ``system of systems'' or change its procurement 
strategy in any form. The intent of the recommendation is for DHS to 
insert the individual schedules for each of the Plan's programs into a 
single electronic Integrated Master Schedule file in order to identify 
any resource allocation issues among the programs' schedules. 
Developing and maintaining an Integrated Master Schedule for the Plan 
could allow OTIA insight into current or programmed allocation of 
resources for all programs as opposed to attempting to resolve any 
resource constraints for each program individually.
    In addition in March 2014, we reported that OTIA's rough order of 
magnitude estimate for the Plan and individual Life-cycle Cost 
Estimates for the IFT and RVSS programs met some but not all best 
practices for such estimates. Cost-estimating best practices are 
summarized into four characteristics--well documented, comprehensive, 
accurate, and credible.\12\ Our analysis of CBP's estimate for the Plan 
and estimates completed at the time of our review for the IFT and RVSS 
programs showed that these estimates at least partially met three of 
these characteristics--well-documented, comprehensive, and accurate. In 
terms of being credible, these estimates had not been verified with 
independent cost estimates in accordance with best practices. We 
recommended that CBP verify the Life-cycle Cost Estimates for the IFT 
and RVSS programs with independent cost estimates and reconcile any 
differences.
---------------------------------------------------------------------------
    \12\ GAO, GAO Cost Estimating and Assessment Guide: Best Practices 
for Developing and Managing Capital Program Costs, GAO-09-3SP 
(Washington, DC: March 2009). The methodology outlined in the Cost 
Estimating and Assessment Guide is a compilation of best practices that 
Federal cost-estimating organizations and industry use to develop and 
maintain reliable cost estimates throughout the life of an acquisition 
program.
---------------------------------------------------------------------------
    DHS said it concurred with this recommendation, although we 
reported that DHS's planned actions will not fully address the intent 
of the recommendation unless assumptions underlying the cost estimates 
change. In particular, DHS stated that at this point it does not 
believe that there would be a benefit in expending funds to obtain 
independent cost estimates and that if the costs realized to date 
continue to hold, there may be no requirement or value added in 
conducting full-blown updates with independent cost estimates. DHS 
noted, though, that if this assumption changes, OTIA will complete 
updates and consider preparing independent cost estimates, as 
appropriate. We recognize the need to balance the cost and time to 
verify the Life-cycle Cost Estimates with the benefits to be gained 
from verification with independent cost estimates. However, we continue 
to believe that independently verifying the Life-cycle Cost Estimates 
for the IFT and RVSS programs and reconciling any differences, 
consistent with best practices, could help CBP better ensure the 
reliability of the estimates.

    CBP DID NOT FULLY COMPLETE DOCUMENTS FOR ACQUISITION DECISIONS 
                      CONSISTENT WITH THE GUIDANCE

    In March 2014, we reported for the Plan's three highest-cost 
programs--IFT, RVSS, and MSC--DHS and CBP did not consistently approve 
key acquisition documents before or at the Acquisition Decision Events, 
in accordance with DHS's acquisition guidance. An important aspect of 
an Acquisition Decision Event is the review and approval of key 
acquisition documents critical to establishing the need for a program, 
its operational requirements, an acquisition baseline, and test and 
support plans, according to DHS guidance. On the basis of our analysis 
for IFT, RVSS, and MSC programs under the Plan, we reported that the 
DHS Acquisition Decision Authority approved the IFT program and the CBP 
Acquisition Decision Authority approved the RVSS and MSC programs to 
proceed to subsequent phases in the Acquisition Life-cycle Framework 
without approving all six required acquisition documents for each 
program. Furthermore, we reported that one document for the IFT 
program, five documents for the RVSS program, and two documents for the 
MSC program were subsequently approved after the programs received 
authority to proceed to the next phase. DHS plans to complete and 
approve those documents for the IFT, RVSS, and MSC programs that have 
not yet been completed and approved.
    With regard to one of the required documents--the Test and 
Evaluation Master Plan--we reported in March 2014 that this document 
for the IFT program, which was approved by DHS in November 2013, does 
not describe testing to evaluate the operational effectiveness and 
suitability of the system. Rather, the Test and Evaluation Master Plan 
describes CBP's plans to conduct a limited user test of the IFT. 
According to the Test and Evaluation Master Plan, the limited user test 
will be designed to determine the IFT's mission contribution. According 
to OTIA and the Test and Evaluation Master Plan, this testing is 
planned to occur during 30 days in environmental conditions present at 
one site--the Nogales station. CBP plans to conduct limited user 
testing for the IFT under the same process that is typically performed 
in any operational test and evaluation, according to the Test and 
Evaluation Master Plan. The November 2013 IFT Test and Evaluation 
Master Plan notes that, because the IFT acquisition strategy is to 
acquire non-developmental IFT systems from the marketplace (sometimes 
referred to as a commercial off-the-shelf system), a limited user test 
will provide Border Patrol with the information it needs to determine 
the mission contributions from the IFTs, and thus CBP does not plan to 
conduct more robust testing. However, this approach is not consistent 
with DHS's acquisition guidance, which states that even for commercial 
off-the-shelf systems, operational test and evaluation should occur in 
the environmental conditions in which a system will be used before a 
full production decision for the system is made and the system is 
subsequently deployed.
    As we reported, we recognize the need to balance the cost and time 
to conduct testing to determine the IFT's operational effectiveness and 
suitability with the benefits to be gained from such testing. Although 
the limited user test should help provide CBP with information on the 
IFTs' mission contribution and how Border Patrol can use the system in 
its operations, the limited user test does not position CBP to obtain 
information on how the IFTs may perform under the various environmental 
conditions the system could face once deployed. Conducting limited user 
testing in one area in Arizona--the Nogales station--for 30 days could 
limit the information available to CBP on how the IFT may perform in 
other conditions and locations along the Arizona border with Mexico. As 
of November 2013, CBP intended to deploy IFTs to 50 locations in 
southern Arizona, which can include differences in terrain and climate 
throughout the year.
    We recommended that CBP revise the IFT Test and Evaluation Master 
Plan to more fully test the IFT program, before beginning full 
production, in the various environmental conditions in which IFTs will 
be used to determine operational effectiveness and suitability. DHS did 
not concur with this recommendation and stated that the Test and 
Evaluation Master Plan includes tailored testing and user assessments 
that will provide much, if not all, of the insight contemplated by the 
intent of the recommendation. However, as we reported in March 2014, we 
continue to believe that revising the Test and Evaluation Master Plan 
to include more robust testing to determine operational effectiveness 
and suitability could better position CBP to evaluate IFT capabilities 
before moving to full production for the system, help provide CBP with 
information on the extent to which the towers satisfy Border Patrol's 
user requirements, and help reduce potential program risks.

CBP HAS IDENTIFIED MISSION BENEFITS, BUT DOES NOT CAPTURE COMPLETE DATA 
         ON THE CONTRIBUTIONS OF ITS SURVEILLANCE TECHNOLOGIES

    We reported in March 2014 that CBP has identified the mission 
benefits of its surveillance technologies, but does not capture 
complete data on the contributions of these technologies, which in 
combination with other relevant performance metrics or indicators, 
could be used to better determine the contributions of CBP's 
surveillance technologies and inform resource allocation decisions. CBP 
has identified mission benefits of surveillance technologies to be 
deployed under the Plan, such as improved situational awareness and 
agent safety.
    While CBP has defined these mission benefits, the agency has not 
developed key attributes for performance metrics for all surveillance 
technologies to be deployed as part of the Plan, as we recommended in 
November 2011.\13\ In our April 2013 update on the progress made by the 
agencies to address our findings on duplication and cost savings across 
the Federal Government, CBP officials stated that operations of its two 
SBInet surveillance systems identified examples of key attributes for 
metrics that can be useful in assessing the Plan's implementation for 
technologies.\14\ For example, according to CBP officials, to help 
measure whether illegal activity has decreased, examples of key 
attributes include decreases in the amount of arrests, complaints by 
ranchers and other citizens, and destruction of public and private 
lands and property. While the development of key attributes for metrics 
for the two SBInet surveillance systems is a positive step, CBP has not 
identified attributes for metrics for all technologies to be acquired 
and deployed as part of the Plan. Thus, to fully address the intent of 
our recommendation, CBP would need to develop and apply key attributes 
for performance metrics for each of the technologies to be deployed 
under the Plan to assess its progress in implementing the Plan and 
determine when mission benefits have been fully realized.
---------------------------------------------------------------------------
    \13\ GAO-12-22.
    \14\ GAO, 2013 Annual Report: Actions Needed to Reduce 
Fragmentation, Overlap, and Duplication and Achieve Other Financial 
Benefits, GAO-13-279SP, (Washington, DC: Apr. 9, 2013).
---------------------------------------------------------------------------
    Furthermore, we reported in March 2014 that CBP is not capturing 
complete asset assist data on the contributions of its surveillance 
technologies to apprehensions and seizures, and these data are not 
being consistently recorded by Border Patrol agents and across 
locations. Although CBP has a field within its Enforcement Integrated 
Database (EID) for maintaining data on whether technological assets, 
such as SBInet surveillance towers, and non-technological assets, such 
as canine teams, assisted or contributed to the apprehension of illegal 
entrants, and seizure of drugs and other contraband, according to CBP 
officials, Border Patrol agents are not required to record these 
data.\15\ This limits CBP's ability to collect, track, and analyze 
available data on asset assists to help monitor the contribution of 
surveillance technologies, including its SBInet system, to Border 
Patrol apprehensions and seizures and inform resource allocation 
decisions.
---------------------------------------------------------------------------
    \15\ In addition to maintaining data on asset assists, the Border 
Patrol collects and maintains data on apprehensions and seizures in 
DHS's EID.
---------------------------------------------------------------------------
    We reported that according to our analysis of EID asset assist data 
for apprehensions and seizures in the Tucson and Yuma sectors from 
fiscal year 2010 through June 2013, information on asset assists was 
generally not recorded for all apprehension and seizure events.\16\ For 
instance, for the 166,976 apprehension events reported by the Border 
Patrol across the Tucson sector during fiscal year 2010 through June 
2013, an asset assist was not recorded for 115,517 (or about 69 
percent) of these apprehension events. In the Yuma sector, of the 8,237 
apprehension events reported by Border Patrol agents during the 
specified time period, an asset assist was not recorded for 7,150 (or 
about 87 percent) of these apprehension events. Since data on asset 
assists are not required to be reported, it is unclear whether the data 
were not reported because an asset was not a contributing factor in the 
apprehension or seizure or whether an asset was a contributing factor 
but was not recorded by agents.
---------------------------------------------------------------------------
    \16\ In our March 2014 report, we defined an ``apprehension or 
seizure event'' as the occasion on which Border Patrol agents make an 
apprehension of an illegal entrant or a seizure of drugs or other 
contraband. The event is recorded in the EID and a date and unique 
identifying number are assigned. An event can involve the apprehension 
of one or multiple illegal entrants or types of items, and each 
individual illegal entrant apprehended or type of item seized in the 
event is associated with the assigned unique identifying number. Our 
analysis of apprehension events included instances in which an event 
had at least one deportable individual.
---------------------------------------------------------------------------
    As a result, CBP is not positioned to determine the contribution of 
surveillance technologies in the apprehension of illegal entrants and 
seizure of drugs and other contraband during the specified time frame. 
We reported that an associate chief at Border Patrol told us that while 
data on asset assists are not systematically recorded and tracked, 
Border Patrol recognizes the benefits of assessments of asset assists 
data, including those from surveillance technologies, such as the 
SBInet system. The associate chief further noted that these data in 
combination with other data, such as numbers of apprehensions and 
seizures, are used on a limited basis to help the agency make 
adjustments to its acquisition plans prior to deploying resources, 
thereby enabling the agency to make more informed deployment decisions.
    We recommended that CBP require data on asset assists to be 
recorded and tracked within EID and that once these data are required 
to recorded and tracked, analyze available data on apprehensions and 
technological assists, in combination with other relevant performance 
metrics or indicators, as appropriate, to determine the contribution of 
surveillance technologies to CBP's border security efforts. CBP 
concurred with our recommendations and stated that Border Patrol is 
changing its data collection process to allow for improved reporting on 
asset assists for apprehensions and seizures and intends to make it 
mandatory to record whether an asset assisted in an apprehension or 
seizure. DHS plans to change its process by December 31, 2014.
    Chairwoman Miller, Ranking Member Jackson Lee, and Members of the 
subcommittee, this concludes my prepared statement. I would be pleased 
to answer any questions that you may have.

    Mrs. Miller. Thank you very much.
    I think, Mr. Borkowski, I will pick up right where Ms. 
Gambler on her last point there about this asset assist.
    I am not quite sure the mechanics of how that would work, 
but I suppose that is something out in the field where as they 
would--as they are going through their checklist of how an 
apprehension happened or whatever the incident was.
    What is your thought about why you are not accumulating 
that kind of information? Is it just too much of a burden for 
the folks out in the field or you don't think it would be worth 
its time or what?
    Mr. Borkowski. I think the issue at this point is just 
having the systems and the capacity to do that. We generally 
have the information. The question is how do we get that 
recorded and documented, but the Border Patrol is working on 
that and it is committed to doing that because we believe it 
does tell us something.
    I would caution about how much it tells us, right, because 
you get in situations where various components came together to 
produce an apprehension and it is tempting to say this 
apprehension was caused by the agent or this apprehension was 
caused by the technology.
    Ultimately, we like to get that sensitivity--how much can 
we attribute to technology, but the reality is these things 
merge together. So I think we want to collect the data, but I 
would caution until we collect it and study it, it is still an 
open question as to how much utility that will provide us 
downstream in terms of a reasonable performance metric.
    Mrs. Miller. You know I asked that question because as you 
are aware, this subcommittee unanimously and the full committee 
unanimously passed our border bill which is going to be an 
accountability matrix at the border in utilizing various kinds 
of systems and whatever we can to really be able to effectively 
have accountability and measure our success or failure for that 
matter, so that won't be the end of that line of questioning I 
would guess.
    I also wanted to pick up on something you mentioned about a 
75 percent savings. As you said, to paraphrase, it is something 
that doesn't normally happen in Government to have a 75 percent 
savings, and I guess I would ask both of you this question of 
whether or not Ms. Gambler was surprised to see the 75 percent 
savings, and what was the construct of that budget request to 
have a 75 percent savings.
    I know that the Department of Homeland Security would have 
had to review your budget request. Did they have all the pieces 
available? It certainly calls in to--makes us think a bit about 
some of the budget requests that we are seeing.
    Mr. Borkowski. I think what is important to understand is 
first of all where those numbers come from. Life-cycle cost 
estimate, right. That was a term you heard. In life-cycle cost 
estimates there are a bunch of methodologies of generating 
them, but they are all based on experience.
    When it comes right down to it, you get a life-cycle cost 
estimate based on your past experiences and you try to find the 
closest examples of systems like the things you are buying and 
what did they cost.
    But when your experience is all based on that, ``I am not 
going to the lot to buy a car. Instead I am going to go find 
somebody; I am going to ask them to build me a car from 
scratch.'' That is where most of our cost experience is from, 
so that is what comes into a life-cycle cost estimate.
    The GAO looks at this. We look at this. Did they follow 
best practices? The answer is largely I would argue it did 
based on the way we do life-cycle cost estimates.
    But we tried to buy these differently. We tried to buy them 
by going to the lot. You can imagine the difference it would 
cost me to get a car if I went to Sam's manufacturing plant and 
said build me a car from scratch or if I go to the local Ford 
or Chrysler or GM dealer, whoever I go to.
    I think in your own mind you can imagine the difference 
between those, so that is factor No. 1. By the way, I was 
surprised by 75 percent. When we built the life-cycle cost 
estimate, and I think if we go back to testimony of this 
committee and similar discussions, there was some concern; is 
that number too low?
    We were arguing we put contingency in it because we are 
doing something that we haven't done before, but I was 
surprised by 75 percent to be honest. I was hopeful we would 
have cost savings. I was actually surprised by 75 percent.
    Here is the other thing that happens, and I have been 
trying to find a good way to make this case because these are 
subtle things that are hugely important as it turns out.
    A lot of times when we want performance in something--let's 
say I want 90 percent probability detection. When I am 
designing that from scratch because of the way we test it, I 
actually have to design it to say 93, 94, 95 percent even 
though what I want is 90 percent.
    The reason is the statistics that go into testing to be 
sure that I really get 90. So I am trying to think of an 
analogy here, and I am in your area, not mine, but this is what 
it looks like to me.
    You are running a campaign for election and you have got 
two candidates. You need 50 percent of the vote and you run a 
poll. The poll says I have 51 percent.
    Now you may feel comfortable about that but in the fine 
print it says with a margin of certainty plus or minus 3 
percent, which means the poll really said you have somewhere 
between 48 and 54, so now you have a decision I would think. Do 
you go after that last 3 percent to make sure you don't hit the 
48?
    That costs a lot more than the first 3 percent cost. That 
is exactly what happens to us when we buy things. Going from 90 
to 95 percent to guarantee I got 90 percent--that costs a lot 
more than the first 5 percent did.
    When I go and buy the things off the lot and I say look 
tell me what you have got and we will accept your definition of 
it because I don't really care about 1 or 2 percent either way, 
huge cost savings. So it looks like that is what happened here.
    I think that this is something new, and that is another 
point I think when we do something new we have to tailor the 
past practices to the thing that is new.
    I believe we need to study this some more. I believe the 
jury is still out, but it looks to me like a good deal of this 
is real.
    Mrs. Miller. I am going to ask Ms. Gambler to respond as 
well, but in regards to your analogies being from the motor 
city, I would stay away from the campaign analogies. Stick with 
the car analogies. They are better.
    [Laughter.]
    Mrs. Miller. Ms. Gambler.
    Ms. Gambler. I would just add that I think that what will 
be important going forward and in looking at the contract and 
the deployment of these IFT's is how CBP, how the Government 
will hold the contractor accountable and provide that oversight 
for the systems that are being deployed.
    From our perspective what isn't clear from the test plans 
that we have seen from CBP thus far is how they will be able to 
ensure that the IFTs once deployed are meeting the Border 
Patrol's operational needs and will work and the locations and 
under the conditions that those technologies planned to be 
deployed.
    So I think what is critical here going forward is how CBP 
will be positioned to have the information it needs to provide 
oversight for the contractor in the systems that are being 
deployed.
    Mrs. Miller. Just one other question and for both of you as 
well and this would be in regards to the testing that has been 
done which the GAO mentioned is one of the recommendations, and 
you had some concerns about all of that.
    We just want to make sure it is working, right? When we 
think about the lessons learned and the involvement or not 
enough involvement perhaps of those that are--the customers 
really--the end-users of this is the folks in the field.
    Could you expand a little bit on the testing that was done 
and why you feel that this is going to work and why you are--
why GAO is raising this as a concern about the amount of 
testing that was done and how it all unfolded?
    Mr. Borkowski. First of all, what was done was a demo. 
Remember, we wanted something that was on the lot so we wanted 
bidders to prove to us that it was really on the lot and it 
wasn't just a brochure.
    So the bidders actually had to take their system out to the 
field, take it out of the box, turn it on, and show that it did 
what they said it would do, which is unique. We don't normally 
do that.
    By the way, most of industry asked us to do that because 
they said, you know, Fred is going to lie to you but I am 
telling the truth, but you won't know it unless you make both 
of us prove it.
    So it was a demo. It was not a full-blown test. However, 
these are systems that have been around for a while and we were 
able to get an initial indication of that.
    The other thing that I want to emphasize is, it is not like 
there will be no testing. There will. There is a term of art 
that is being tossed around here called operational test and 
evaluation, but there is a whole bunch of different kinds of 
test and evaluation, so we will have extensive testing that 
these systems produce the results that were committed to in the 
contract.
    That will start with something called system acceptance 
test, which technically is developmental test and evaluation, 
so there is a whole rubric here that comes into play.
    Then limited user testing is a test that is operational 
test and evaluation designed to meet the Border Patrol's desire 
to get answers to its questions.
    So what we have actually done, rather than formal 
operational test and evaluation is sat down with the Border 
Patrol and said, what do you want to check out? What do you 
want to know about the system?
    That is all documented. It is all committed. All of that 
testing will be done to advise the Border Patrol on how much 
more of this it wants to do.
    So we are doing the intent of that, but when I go to formal 
operational test and evaluation I create aspirational goals 
that get me to that 3 percent more problem that I was 
describing and that is what we want to avoid.
    So we have tailored the approach to this, but there is 
extensive testing intended in this plan.
    Mrs. Miller. Ms. Gambler.
    Ms. Gambler. I think what is key here is that testing helps 
an agency manage risks and it doesn't necessarily mean that 
there is going to be fewer problems. It just means you are 
getting information on potential risks or potential risks of 
problems to a system earlier on in the process than you might 
otherwise have.
    That information is key to providing program managers with 
what they need to know and what they need to do to address any 
issues that comes up, and then positions them to be able to 
address those issues earlier in the process.
    So from our perspective there is an opportunity here for 
CBP to do more robust testing than what they currently plan to 
do to help ensure that the technology that is delivered by the 
vendor meets Border Patrol's needs and will operate in the 
different environments where the towers are intended to be 
placed.
    We think this is important given--or in light of some of 
the testing challenges that SBInet encountered including, for 
example, the Army Test and Evaluation Command identifying that 
there were some issues with how terrain affected the radars of 
SBInet. The ATEC officials referred to it as a technology 
terrain mismatch.
    So we think in light of those challenges there is an 
opportunity here for CBP to do some additional testing that 
would give them the information they need to effectively manage 
the program.
    Mrs. Miller. Thank you very much.
    The Chairwoman recognizes the gentlelady from Texas.
    Ms. Jackson Lee. Ms. Gambler, if I might pose a question to 
you. Based on your work on the Arizona Border Surveillance 
Technology Plan and GAO's prior work on SBInet are there 
similar challenges of warning signs you are seeing with the 
plan that GAO believed contributed to some of SBInet's 
failures? What must be done to address these warning signs?
    Ms. Gambler. Based on our work, DHS and CBP certainly have 
followed best practices and DHS' acquisition guidance in 
certain areas, but we did report that they faced similar types 
of challenges in managing the plan as they encountered under 
SBInet.
    For example, for both SBInet and the new Arizona technology 
plan we identified that CBP has schedules and life-cycle cost 
estimates that don't substantially meet best practices in all 
areas.
    As I mentioned, for both SBInet and the Arizona technology 
plan, we identified some challenges with test plans for those 
two systems and for both SBInet and the Arizona technology plan 
we identified in both cases that CBP hadn't identified 
performance metrics for assessing basically what we are getting 
for the technology. So in those areas there are similar types 
of challenges between the two systems.
    What we have done is made recommendations to DHS and CBP 
both on SBInet and the new Arizona technology plan to get at 
some of those challenges to make sure that they more fully 
adopt and use best practices for scheduling and life-cycle cost 
estimating and also that they establish metrics for assessing 
the contributions of technologies to border security and that 
they collect data to be able to assess those metrics.
    Ms. Jackson Lee. Do you have an update? Have they begun to 
do that specifically with life cycles and the test plan? Do you 
have a report that says where they are now?
    Ms. Gambler. So the recommendations we made with regard to 
SBInet have largely been closed because SBInet has been 
canceled. In terms of the recommendations that we have made as 
it relates to the Arizona technology plan, we made those----
    Ms. Jackson Lee. I know that SBI has been--I am speaking 
about going forward. The new proposal.
    Ms. Gambler. Sure. We made those recommendations in the 
report being released today and CBP does plan to take some 
actions in response to that.
    For example, the Border Patrol is starting to work to make 
changes to how it collects data on asset assists so that it can 
set some metrics for how to assess contributions of border 
surveillance technologies and CBP also has plans to look at 
their schedules and life-cycle cost estimates as well.
    As they are updating those, they have plans to make sure 
that they do more fully comply with best practices.
    Ms. Jackson Lee. How do you know that?
    Ms. Gambler. That is what they reported to us in terms of 
what they plan to do, Ranking Member Jackson Lee, and we will 
continue to monitor their progress in response to those 
recommendations as they start to implement those actions.
    Ms. Jackson Lee. How long would they need for compliance? 
How long would they need to engage in the process?
    Ms. Gambler. For the different recommendations they have 
set different time frames for completion. Some of them are 6 
months to a year out and we will, as I said, continue to 
monitor their progress and would be happy to update you on 
their progress as they go forward.
    Ms. Jackson Lee. So as we stand now from the 
recommendations made by GAO, CBP is not in compliance? Going 
forward, not on SBInet, which I know has been canceled.
    Ms. Gambler. Correct. CBP right now has not yet implemented 
the recommendations we have made.
    Ms. Jackson Lee. What is the give-and-take to--again I am 
going back--what is the give-and-take to expect compliance? 
What is the engagement and the report back that you get?
    Ms. Gambler. For recommendations that we make on any 
report, we regularly follow up with the agency to determine the 
status of actions they are taking in response to the 
recommendations.
    That could be exchange of documents or meetings with agency 
officials and then we make an assessment of the extent to which 
the actions that the agency has taken are responsive to the 
intent of our recommendation.
    Ms. Jackson Lee. When do you expect then to get back with 
CBP on these present recommendations?
    Ms. Gambler. Some of the recommendations they indicated 
about 6 months or so, that we might be able to start getting 
some information from them, so that would probably be the time 
frame that we will start to follow up.
    Ms. Jackson Lee. Mr. Borkowski, you acknowledge SBInet and 
in my statement I acknowledged that it was $1 billion in cost. 
You have now engaged with Elbit Systems of America on the 
deployment of integrated fixed towers and that valued contract 
was $145 million although the initial projected cost was $600 
million.
    To what do you attribute the significant cost discrepancy? 
Do we have a quality product? You just heard GAO, Ms. Gambler, 
mention that you are not yet in compliance. You have indicated 
at least 6 months on certain aspects of the testing aspect.
    So would you comment on again the cost discrepancy? Is this 
a lesser technology than originally planned? Do you think you 
will see more inaccuracies? Are you confident that we have got 
the best contractor, but more importantly, is this a true 
answer to SBInet?
    Mr. Borkowski. With the last question, we are not trying 
to--in terms of the correction of the errors that led to 
SBInet, I think it is a good chunk of the way, 80, 90 percent 
of the way there. I do think that.
    Is this a good contractor? The system--we had many bidders, 
many bidders, and of those many bidders we saw several systems 
that appear to work right out of the box.
    The one that we awarded is actually used for border 
security in Israel as it turns out, but there were other 
bidders who had very effective systems of that worked right out 
of the box, so that is something different.
    Our sense is that the quality is high, and I think the cost 
difference is due to what I tried to describe earlier. When we 
buy things and go to a specialty shop to build them from 
scratch we pay more than when we buy something off the lot.
    Now if there is nothing off the lot to buy then we are 
stuck with going to a specialty vendor and that is the way we 
bought SBInet. The problem was there was stuff on the lot, so 
when we changed to that you have a significant impact.
    Then when I go to those folks and say look, I am not asking 
you. I am going to tell you what I am interested in, but I am 
not asking you to go to the last 3 percent. Tell me what you 
can do, give me some reason to believe that, and if that is 
good enough for me I will buy it.
    It is that last 3 percent cost that cost you an arm and a 
leg, so I think those are all things that led to this 
significant cost reduction. By the way, that $145 million 
covers 50 towers in Arizona and 7 years of operations and 
maintenance, so it is quite a bit of functionality that we get 
out of it.
    What we saw in the demos was very impressive. I don't know 
if the Border Patrol would say this, but when they described 
what they saw in that system in the demo, they almost were 
raving about it to be frank. Hopefully that will get us to 
where we want----
    Ms. Jackson Lee. Thank you.
    Let me ask if I might, will you wrap this into the question 
that I am going to give you? I just want to make sure you 
answer, Ms. Gambler, that you will be making a definitive 
effort to follow through on the life cycle and test plan 
testing.
    But my question that I want to ask is in my statement I 
mentioned the Border Patrol Chief Michael Fisher informed the 
office that the Border Patrol needs fewer integrated fixed 
towers in Arizona than originally planned and instead wants 
more mobile surveillance technology to be used in South Texas--
more mobile surveillance technology to be used which has 
increasingly become a hot spot for illicit cross-border 
activity in recent years meaning more of that in South Texas.
    Given that it has been 3 years since the cancellation of 
SBInet, keeping in mind the dynamic situation along the border, 
how confident are you that the number and type of technologies 
planned for the Arizona border is appropriate? How do you plan 
to address Chief Fisher's inquiries, if you would?
    Thank you, Madam Chairwoman.
    Mr. Borkowski. Certainly. Chief Fisher has maintained the 
requirement for all of the IFTs but he has said they are lower 
priority, so there is a subtle distinction there, but he says 
that is still a requirement.
    However, he would like to divert resources and do things in 
South Texas first. So what we have done there is, as you know, 
we have for example put up and these are demonstrations and 
evaluations, they are kind of temporary activities, but we have 
flown some aero stats there.
    The one part of the Arizona technology plan is a system 
called the mobile video surveillance system also known as a 
scope truck. It has got an infrared camera that is long-range 
that sees. The Border Patrol asked us to divert those to South 
Texas. That contract will be awarded around June.
    In addition, the contract that awarded it mobile 
surveillance capabilities, those are going to Arizona and will 
free up resources that will then be moved to South Texas; 
mobile resources that will be moved to South Texas.
    The other element of this is that again, as I have said we 
have saved some cost and the Border Patrol has asked us to 
prioritize those costs savings not in filling in the IFT in 
Arizona just yet, but in applying those resources to continue 
the kinds of things we have been doing in moving mobile video 
surveillance systems using DOD technology in South Texas. So to 
this point that is what we are able to do relatively quickly.
    Ms. Jackson Lee. Thank you. I look forward to this 
continuing dialogue. I think it is going to be crucial as this 
system is put in place. I hope there is some synergism between 
GAO and DHS in getting this right. Thank you.
    I yield back, Madam Chairwoman.
    Mrs. Miller. The Chairwoman now recognizes the gentleman 
from South Carolina, Mr. Duncan.
    Mr. Duncan. Thank you, Madam Chairwoman.
    Just to remind the committee, we are well over $17 trillion 
in debt. That is how big our financial hole is as a Nation. 
Hearing after hearing before the Oversight Management and 
Efficiency Subcommittee which I chair, we have identified DHS 
programs that have overspent and failed to fully deliver.
    Having been down there to the border in Arizona I know 
first-hand how rugged the terrain is. I was surprised to see 
that CBP plans only to test the towers in one location for 1 
month, and I question how much sense that makes.
    Mr. Borkowski, I get the need to get these towers deployed. 
Listen, I appreciate technology being used, and I like the idea 
of integrated towers. I like the idea of aero stats. I like the 
idea of more different surveillance that can aid the CBP 
officers on the ground.
    We have got to have a multi-pronged approach. That includes 
fencing, that includes personnel, and that includes 
surveillance, so I get that. I get the need to deploy these 
quickly, but we have been down this road before of inadequate 
testing with SBInet. So what makes you think that you are in 
the best position for success with only one test?
    Mr. Borkowski. Again, I think it is a misnomer to say there 
will be only one test. There will be extensive tests and the 
plan as it is written says we will test for as long as the 
Border Patrol wants to test.
    In fact, the appropriations act that was passed has 
language that we supported and already intended to do which 
says we will not do subsequent deployments until the chief of 
the Border Patrol says he is comfortable doing subsequent 
deployments.
    Mr. Duncan. Let me ask you this. This is pretty expensive 
technology. This contract is a fairly substantial reward. How 
much testing, real testing have been done by the contractor at 
no cost to the Government because we have got to be good 
stewards and if they want the contract they have got to--in the 
private sector, if somebody wants a contract, they get out 
there, they do all the testing, and they prove to the purchaser 
of their equipment that they have done the testing, so how much 
of that has gone on to save the taxpayers?
    Mr. Borkowski. Well, we did run, as I said, demonstrations. 
We did make them take their system and put it up out along the 
Southwest Border, turn it on, and show us how it worked for a 
couple of days, but remember----
    Mr. Duncan. For a couple of days.
    Mr. Borkowski. Well, but they also, remember, they have 
deployed these in Israel. They have been using them for a 
significant amount of time.
    There is another element to testing that I think is getting 
lost in the discussion here, and it goes to this idea of a 
system development versus an off-the-shelf system.
    A lot of this discussion about testing incrementally comes 
when you say okay, I am going to build--let's go to the car 
example. If I were going and building a car from scratch and 
designing it from scratch and went to Joe's manufacturing 
company, when Joe built the carburetor, I would test the 
carburetor.
    When Joe built the engine, I would test the engine. I would 
test all of the sub pieces, and then I start to put 
subcomponents together and test, and then I put the whole thing 
together and test. That is the kind of testing you do as you 
go.
    But when you have a whole car already built and you bought 
it off the lot, you don't tear it apart to test the carburetor 
and the engine. You test the whole car, and you test it once 
you have bought it.
    That is exactly what we are doing. It is a change in 
paradigm because we are not doing system development, but we 
are doing testing.
    So I don't think we are all getting the accurate picture 
here. That testing as I say we have designed with the Border 
Patrol to say we will test everything you want to test to 
answer your questions, and we will not do another thing until 
you tell us you are comfortable but I still have to buy the 
car. I went and bought it off the lot. I do have to buy the car 
so I can test it and that is what we are doing.
    Mr. Duncan. I appreciate that. Let me shift gears here for 
just a minute. The GAO's report raises questions as to whether 
acquisition management lessons learned from SBInet are being 
applied to the new technology plan.
    I understand that there is some disagreement and that CBP 
has not concurred with all of the recommendations of the most 
recent report, so why did the Department not concur on two of 
GAO's suggested recommendations?
    Mr. Borkowski. Basically, because in our view, those two 
recommendations drive us back to system development, back to 
that approach that says I am not going to buy the car off the 
lot, I am going to go to Joe's manufacturing company and build 
one from scratch.
    Plus, the other thing that they do is they take us back to 
the days of SBInet where we had everything all tied together as 
one program and couldn't pull the pieces apart.
    So the one recommendation we non-concurred on was for 
something called an IMS, an integrated master schedule, which 
again is a sort of a term of art. It is a schedule but it is a 
schedule on steroids.
    It has got all kinds of interconnections and networking at 
very low levels of detail and each program has an integrated 
master schedule, and we did that on purpose. We want separate 
programs. I want to be able to buy IFT whether or not I buy 
RVSS and so forth and so on.
    The GAO says that we should take all of those separate 
programs and create an IMS for them as one program. That is not 
consistent with DHS practice and what it has the effect of 
doing is turning us back into SBInet. We think that is a really 
bad idea.
    The second recommendation that we non-concurred with had to 
do with operational test and evaluation. Not in the spirit, 
because as I have already explained, we are going to do all of 
those things with the Border Patrol, but when you do formal 
operational test and evaluation you set targets.
    As I was explaining to the Chairwoman, if I set a target of 
90 percent I actually have to pay to get to 95 percent to prove 
I can do 90 percent. We do not want to do that. We do not want 
to pay that premium. We do not need to pay that premium, so in 
our view, those two recommendations are driving us exactly back 
to the things we tried not to do after we got out from under 
SBInet.
    Mr. Duncan. I am following acquisition very closely, so I 
am interested to see how this actually transpires.
    I don't want to sit here, Madam Chairwoman, 2 years from 
now and try to justify to the American taxpayer why we had 
another SBInet and wasted their money. I hope this works.
    I am interested in seeing it, and I am going to wish for 
the success because security at the Southwest Border is very 
important to folks in South Carolina and across this Nation.
    I have got some other comments and questions. I will just 
wait.
    Ms. Gambler, I want you to chime in so I am going to wait 
for the second round hopefully and with that, I will just yield 
back.
    Ms. Jackson Lee. Could I just ask you to yield just for one 
moment?
    I just want to thank you for the work you have done on 
contracts or procurement. I have said it on some of the 
hearings. I just want this one sentence. We all want this to 
work. We are in a new world and again I just want to repeat 
that our critique today is to make it better for America.
    I yield back.
    Mr. Duncan. Thanks to the lady and the critique will make 
it better. I think that is what the Chairwoman is wanting on 
this to make sure that, (A) acquisition is happening. We are 
not spending taxpayer dollars recklessly and that we are making 
good decisions for the Nation. We are applying best management 
practices and we will secure the border and stop the flow of 
illegals into the country.
    With that, I yield back.
    Mrs. Miller. Absolutely.
    The Chairwoman now recognizes the gentleman from Texas, Mr. 
O'Rourke.
    Mr. O'Rourke. Thank you, Madam Chairwoman.
    For Mr. Borkowski, how much will we spend on the Arizona 
Border Surveillance Technology Plan?
    Mr. Borkowski. I am trying to add up the pieces in my head 
here. Our original estimates back when we first advertised the 
planned----
    Mr. O'Rourke. I just want to know what we are going to 
spend.
    Mr. Borkowski. I am thinking it will be $500 million to 
$700 million for deployment plus up to 10 years of operation 
and maintenance. I am thinking it will be in that ballpark when 
it is all said and done.
    Mr. O'Rourke. Five hundred million dollars to $700 million 
for deployment and what is O&M cost over 10 years?
    Mr. Borkowski. It is included in that number, so I would 
expect it would be roughly 50-50 between the deployment cost 
and the O&M cost for 10 years. That is a ballpark figure.
    Mr. O'Rourke. Okay. Going back to SBInet which I understand 
you were the executive director of before moving to this 
current position, we spent $1 billion and have a 53 miles 
monitored, so roughly just under $19 million a mile.
    In learning from SBInet and looking forward to what we are 
doing here with the Arizona Border Surveillance Technology 
Plan, one of the pieces that was missing before I think were 
performance metrics that we would be able to look at and know 
whether or not we had a success.
    What are the performance metrics for this?
    Mr. Borkowski. The performance metrics for this are 
basically to--and we tailored them--to ask technology to look 
in areas where technology can see.
    So you heard for example that we couldn't see through rough 
terrain----
    Mr. O'Rourke. I want to know to answer my colleague from 
South Carolina's question, in 2 years, in 5 years, in 10 years, 
when we are looking back at the $600 million plus that we have 
spent, how do we know that it was a success?
    What are the objective numbers that I am going to be able 
to look at to make an objective judgment on whether or not that 
was money well-spent?
    Mr. Borkowski. When you combine this technology with 
effectiveness ratio, I think you will get the numbers you are 
looking for. So what we will be able to tell you is, is or is 
not something going on, how much activity is in that area, and 
then the Border Patrol will be able to tell you of that 
activity how much did they interdict. I believe that is what 
you are looking for.
    Mr. O'Rourke. Okay. I don't know that I totally understand 
that. I would hope that when we are going to spend this kind of 
money on the heels of I think one of the worst, most missed 
managed Federal projects in SBInet, a billion-dollar boondoggle 
that didn't work and netted us 53 miles, I am not sure how we 
are doing in those 53 miles.
    When the GAO assesses that, makes recommendations, and you 
choose not to implement those recommendations when you don't 
have clear performance metrics, when you don't have an 
integrated master schedule, when you don't have the true life-
cycle cost estimates, when you say something like tell me what 
you can do and I will buy it to a vendor when I think we had 
vendors really controlling the situation in SBInet and kind of 
designing it as they went along.
    When you say that you are flexible on the technical details 
but you don't have a coherent way to measure what we are going 
to do, I am worried that we are going to have something akin to 
SBInet again.
    To again answer the concerns from my colleague from South 
Carolina I watched a couple of years ago as a private citizen 
in an El Paso border community the god-awful waste that was 
SBInet and I swore that I would never allow something like that 
to happen again.
    So I don't want to hope that we are not going to have this 
problem again. I want to stop it in its tracks now before we 
get there. If we had not had the problem with SBInet, I think 
some of what we are talking about today might be forgivable or 
understandable, but following that why not follow the 
recommendations of the GAO?
    I think it is--frankly it is a very difficult to hear you 
say that the GAO is trying to point you in the direction of 
replicating the errors of SBInet basically blaming the GAO for 
sending you back into SBInet when I think it was the GAO that 
uncovered the awful problems within SBInet.
    I really--I have just got to say for the record, Madam 
Chairwoman, I think we need to stop this program now until you 
can adequately describe what the performance metrics are, what 
the value is that we are going to get for this money that is 
adding to the National debt and a better answer as to why you 
are not going to comply with the recommendations from the GAO.
    I find today's testimony just very troubling for all of the 
reasons we have talked about from border security to the value 
of taxpayer dollars to official oversight and accountability 
from CBP.
    I think this money could be much better used in much better 
ways.
    With that, Madam Chairwoman, I will conclude.
    Mrs. Miller. I thank the gentleman for his statement and 
his concerns.
    Neither myself or the Ranking Member have any additional 
questions.
    Does the gentleman from South Carolina wish to ask--take 
some additional time?
    Mr. Duncan. Thank you.
    I would just like for Ms. Gambler to chime in on the last 
question I asked about the Department not concurring with GAO's 
suggested recommendations and the differences between what GAO 
has said and what maybe Mr. Borkowski testified to.
    Ms. Gambler. Thank you, Congressman Duncan.
    From our perspective our findings and our recommendations 
are not intending for DHS or CBP to create a system of systems 
or re-aggregate programs under the Arizona technology plan.
    Rather, they are intended to help ensure that CBP 
consistently follows best practices and acquisition guidance 
and those best practices and acquisition guidance exist and are 
designed to help ensure that acquisitions are managed 
effectively and efficiently and that acquisition programs 
deliver to meet operational needs.
    So for example, with regard to the integrated master 
schedule recommendation that we made--again, our intent isn't 
for CBP to re-aggregate programs under that schedule.
    The intent of our recommendation is that DHS or CBP would 
take the schedules that they have for the individual programs, 
insert them into one master schedule file, and by doing that 
CBP could look at resource constraints or resource issues 
across the different programs and resolve those issues as they 
come up rather than resolving them on a program-by-program 
individual basis.
    Mr. Duncan. I apologize for interrupting you. They don't 
have a master schedule now for the complete program to my 
understanding. Is that correct?
    Ms. Gambler. That is what our finding is, yes.
    We would also added that an integrated master schedule is 
viewed as being a good practice because even where there aren't 
direct linkages between programs, in cases where programs are 
being reported to a single customer or single client, in this 
case the Border Patrol, it is helpful to have the schedules 
inserted into one file so that for reporting purposes the 
customer or client can have a comprehensive or a concise view 
of the programs and understand when the programs will be 
completed and when the overall plan will be completed.
    Mr. Duncan. Thank you.
    I really don't have anything further, Madam Chairwoman. I 
just wanted to give Ms. Gambler a chance to chime in and I ran 
out of time.
    Thank you so much for going back.
    Mrs. Miller. You are welcome.
    Ms. Jackson Lee. Madam Chairwoman, I don't have a question 
either. I just want to say this as I keep hearing integrated 
systems let me be very clear. Coming from Texas and being 
engaged in the border for all of these years, I want to see the 
strategic approach that is in a bill that has not yet passed 
the floor of the House, but a very good bill.
    I want to hear about California, Texas, New Mexico, 
Arizona, and as we are discussing this, this is the title of 
this hearing--dealing with the Arizona fixed issues, I want to 
make sure that we are consistently saying to DHS, we want a 
consistent, responsible, respected border approach to allow for 
the free flow of those who come to do us good and who want to 
come and do many things that have added to our economy but also 
to be able to be in a block for those who want to do us harm.
    But we want to hear about Texas and Arizona, New Mexico, 
and California collectively together as it relates to the 
Southern Border and of course I always make mention of the 
Chairwoman's very important area, the Northern Border.
    I yield back, Madam Chairwoman.
    Mrs. Miller. I thank all of the--certainly thank the 
witnesses today, and I thank my colleagues for that which I 
think is a very informative hearing. As it was just mentioned 
by the Ranking Member, we actually have, as you know Mr. 
Borkowski, SBInet that was deployed on the Northern Border and 
a section actually in my district and we have had pretty good 
success with it.
    It may be the one area of SBInet that people point to as it 
has been quite helpful in that particular area, but also as has 
been mentioned here we have this border bill that we hope to 
see some movement on in the House and one of the biggest 
components of that and I think has been a lot of debate across 
the country is how the Senate bill in our mind treats border 
security as just throwing huge pots of money at the border 
again in an ad hoc fashion without really having these 
accountability matrixes, et cetera and something that we all 
discussed during the construct of that bill was SBInet and the 
amount of money that we spent there and the failures that we 
have had there, et cetera.
    So I think as you can see clearly from this subcommittee 
and I think we can say as well the full committee we are really 
going to be watching this thing. We are really going to be 
looking at it, so there remains a number of questions.
    At this time I would say to the Members of the committee if 
they have any additional questions for the witnesses, we would 
ask you witnesses then to respond in writing.
    Pursuant to Committee Rule 7(e), the hearing record will be 
held open for 10 days, and without objection, the committee 
stands adjourned. Thank you.
    [Whereupon, at 11:07 a.m., the subcommittee was adjourned.]