Many state departments of motor vehicles (DMVs) still rely on "mainframe computers" as part of their legacy system environments. Visions of 1950s era cabinet computers filling a room may not be far off. As DMV systems continue to age, and increased demands are placed on them, a number of states are contemplating modernizing their license, vehicle registration, and vehicle title systems. Because some states have recently modernized their mainframe environments, their experiences and lessons learned may be beneficial to those states just beginning the process of DMV modernization.
Buy or build?
With DMV modernization, many states are coming from a custom, "home-grown" environment, so the first question might well be "build or buy?" The new trend in IT shows that many state government agencies are leaning toward commercial off-the-shelf (COTS) products versus developing and supporting a custom-developed system. The prevailing wisdom is that a turnkey solution minimizes system customization, coding, and reliance on state staff who are trying to get out of the programming business. But a COTS solution may not be right in every situation. Mainframe computers and their customized code are specially designed and programmed to accommodate DMV process, such as vehicle registration application processing, which rely on batch processing. An article in GovTech reveals certain skepticism of turnkey solutions in the DMV world, citing California and Oregon's reluctance to embrace an off-the-shelf product that supports the licensing of drivers because of the number of transactions done by DMVs, stating "It's just too big." An RFI issued by New Mexico in 2011 said that an effective COTS system that can be configured for DMV driver and vehicle systems in more than one state may not exist, although the COTS market is becoming increasingly mature.
Cautionary tales from cancelled DMV projects
Once a buy-or-build decision is made, modernization cautionary tales abound, due to the complicated nature of DMV systems and the costs associated with managing such large-sized projects. California's DMV cancelled its modernization project in February for the second time in about 20 years, after spending an estimated $134 million. Vermont's DMV VT Drives project was designed to replace the state's driver license, registration, and title systems and had to be cancelled in 2011. The state was able to reach a settlement with Hewlett Packard (HP) in 2012 for approximately $8 million for the cancelled project. New Mexico cancelled a modernization effort in 2010 even after a thorough systems needs analysis was conducted.
States moving forward
Despite setbacks and lingering questions among states about best practices, systems modernization moves on. Minnesota signed a $41-million dollar contract with HP (the same company that gave an $8-million settlement to Vermont for VT Drives) for a modernized licensing and registration system. Massachusetts Registry of Motor Vehicles awarded a $76-million contract to Deloitte in November 2012 to modernize its license registration and titling systems. The question now becomes: Will these endeavors succeed, and what have these states learned, if anything, about pitfalls to avoid?
Achieving the impossible
Let's end on a positive note. The Ohio Department of Public Safety has recently published an article on a successful project. Their project wrap-up paper reflects on the enormous obstacles they overcame—with a healthy sense of humor. The report, perhaps fittingly, is called "Pigs Really Do Fly!"
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
