Radar towers with automatic video cameras for target identification. Seismic sensors to detect and differentiate human footfalls from cattle and smugglers from border patrol agents. Satellite phones to communicate outside of cell phone contact. Laptops in border patrol vehicles linked to a satellite comm system to integrate the automated sensors with agents ready to be dispatched. And software designed to integrate it all into a seamless, virtual border fence around the Sasabe, Arizona border crossing. This is the Department of Homeland Security’s (DHS) and Boeing’s Project 28, part of the Secure Border Initiative (SBI). And in December, 2007, DHS finally took possession of the prototype for 45 days of border patrol testing – months late and way over budget.

Since I last reported on it in June, Project 28 has fallen far behind schedule and has gone far over budget. National Defense Magazine reported in October that Boeing has had a difficult time integrating all the disparate semi-custom parts together with their software, and Ted Robbins of NPR reported that Congress has threatened to reduce or cut funding for SBInet (the $8 billion Boeing project that Project 28 is a part of), that DHS Secretary Michael Chertoff has threatened to cancel Project 28, and that DHS has refused to pay Boeing $5 million because of cost overruns and months of delay. The NPR story even pointed out that private security guards from Pinkerton Government Services were having to guard the sensor towers because the built-in security systems weren’t turned on yet. Hopefully this had changed when DHS took over testing of the system….

Congress has been understandably annoyed by the whole process. Not only has Congress’ General Accounting Office (GAO) found that the DHS needs to better manage the entire SBInet project, but the DHS also informed Congress that there was going to be a significant delay in Project 28 one day after a Congressional hearing on SBInet where “representatives of DHS, SBInet prime contractor Boeing Co. and SBInet program directors testified” that there were no foreseeable delays in the project (Source). Fortunately, the oversight committments that the House and Senate appropriations committees undertook earlier in 2007 have been included in the omnibus spending bill for the DHS. Specifically, DHS only gets half of the money they requested until they develop plans to address each of the GAOs findings, and until those plans are reviewed by the GAO and the Office of Management and Budget (OMB).

According to DHS, Boeing has also been awarded an additional $64 million “to design, develop and test an upgraded Common Operating Picture software system for Border Patrol command centers and agent vehicles.” This effort was originally included in the Project 28 – was it scaled back after Boeing couldn’t make it work in a timely manner? And if so, then why was Boeing granted another $64 million when they’d shown that they couldn’t handle the job in a timely manner and in budget in the first place? Is this an example of “pay the only people who know what’s going on to fix it” or of throwing good money after bad? I don’t know, but I’d sure like to.

Creating large systems is always difficult. This is true whether the large system is designed from the ground up by a single company or whether, as with SBInet and Project 28, all the parts are purchased as commercial off-the-shelf (COTS) and then integrated. The only difference is where the difficulties lay – in designing and building all the parts so the work together seamlessly or in taking disparate parts and overlaying tools to force them to work together. In either case, though, the costs of large projects balloon due to unanticipated problems poorly managed, and this is what happened with SBInet. Whether this will be enough to convince the federal government to abandon these kinds of border controls and focus on truly effective immigration controls is an open question.