6. CONCLUSIONS AND RECOMMENDATIONS
This section provides the overall conclusions and specific recommendations as they relate to lessons learned regarding institutional and technical challenges. A benefits summary also is included to aid other States and agencies in determining the value of integrating a CAD-TMC system.
6.1 CONCLUSIONS
Utah is fortunate that it had a well developed and established program in place prior to this field test. A joint process for handling incidents had been developed and refined over several years and included access to 9-1-1/CAD information for all types of incidents. Many of the benefits of an integrated TMC-CAD system were realized well before the field test got underway.
The FOT has proven worthwhile for the agencies to continue their quest to develop a true real- time data exchange system. As improvements are completed, operators from both agencies will recognize the benefits.
The real value of this FOT can be applied in Region 4 within Utah, and in other states that do not have the interoperability and strong institutional relationships that are already in place in the Salt Lake Valley region. This is especially true for areas where multiple agencies from state and local government agencies may respond to incidents on freeways, such as home rule states, where interoperable CAD would be a huge benefit in trying to provide real-time traffic information. This would apply both to other regions within Utah and to other states. Delays in obtaining information in these outlying areas far exceed the delays that occur in the Salt Lake City region and sometimes significant events are not reported to UDOT at all.
6.2 RECOMMENDATIONS
The recommendations developed by the Evaluation Team are intended to serve as a general guideline that other states could consider when planning similar CAD-TMC integration projects. The intent was to help states proactively identify issues that may impact deployment cost, schedule, and technical performance, and reflect the lessons learned by Utah during the FOT.
The recommendations are presented in two sections. Section 6.2.1 presents general recommendations for consideration by all stakeholder groups involved in this type of project. Section 6.2.2 captures recommendations specific to the involvement of transit agencies in this type of integration, which were derived from UTA's experience.
6.2.1. General Recommendations
#1: Involve IT staff early-on in the project planning process. Interviewees mentioned the importance of involving agency information technology staff early in the development of the integrated system. This is important so the IT organization provides technical input to the system to assure that the computing and communication environment fit within each agency and can be effectively maintained.
#2: Understand the importance of close working relations from the start. All of those interviewed by the Evaluation Team mentioned the importance of the close working relationship among the agencies involved in this FOT. The close working relationship was strengthened by the work these agencies did in preparation for and during the 2002 Winter Olympic Games. Although not every region can strengthen relationships among agencies by hosting the Olympic Games, agencies should consider how to build these relationships in advance implementing an integrated system.
#3: Provide dedicated staff working on integration, or staff with emphasis on integration. Interviewees mentioned that it was often difficult to spend enough time on the integrated system. Decisions and work items sometimes took longer than those involved would have preferred. Even though every agency supported the integrated system, staff had normal responsibilities with integration duties added on. It would be ideal if staff involved had a priority on the integrated system tasks.
#4: Build in short development cycles to reduce staff turnover issues. Interviewees mentioned that some agencies had critical staff turnover during the implementation of the integrated system. Staff turnover can be disruptive to implementation schedules and budgets as new people have to come up to speed on the system. If the system is planned to have incremental implementations (see section 5.4, Technical Challenges), then the development cycles for each incremental implementation can be short to minimize the likelihood that staff will turnover during a given development cycle. Staff turnover between cycles is not as disruptive as turnover during a development cycle.
#5: Understand the importance of considering role of business practices in the integrated system. As discussed earlier in this document, it is important that the integrated system not require a change in the operator's or dispatcher's work process. However, if other aspects of an agency's business practice would improve the integrated system, it should be considered. For example, VECC agencies were concerned about providing certain information to the integrated system. UDOT is planning to develop an MOU with the VECC agencies that will specify how the information will be used. This may allow a change in those agencies' business practices that will lead to more information shared in the integrated system.
#6: Understand the importance of coordination meetings. Interviewees mentioned the importance of ongoing, periodic coordination meetings with the partner agencies. These meetings kept communication open and emphasis on the integrated project.
#7: Coordinate deployment schedule with vendor schedule for system modifications and upgrades. As mentioned in section 5, CAD systems are generally off-the-shelf products. Vendors have a fixed release schedule. It is important to coordinate project schedules with the vendors' release schedules.
#8: Define what data is exchanged and when. In the Utah system, the IEEE 1512 standard was selected for incident management messages and codes. However, not all vendors supported those codes. It is important for agencies to prepare for differences in codes and determine how to handle these differences.
#9: Decide what incidents will be shared among agencies and what information will be exchanged when an incident is shared. The experience in Utah is leading the participating agencies to automatically send incidents of interest and allow the receiving systems to filter those incidents to display the ones that are likely to be of most interest to the operators.
#10: Understand the importance of incremental implementation. In the Utah system, agencies learned a lot in the initial implementation of the integrated system. The agencies are using that knowledge to plan improvements to the integrated system. For agencies planning an integrated system, it is recommended that they plan an initial implementation and at least one subsequent, incremental improvement. Any group of agencies is almost certain to learn how they would prefer to have the system operate. The project and related contracts should be arranged to allow the agencies to implement what they learn in the initial implementation.
#11: Understand the importance of redundant communication path. As discussed in section 5, a back-up communication pathway is important. Agencies should plan to include redundant communications in an integrated system.
#12: Minimize or avoid duplicate entry. Because not all needed information is transferred from VECC to the integrated system, UDOT operators have to enter data in their system that was already entered by VECC dispatchers in their system. Ideally, any given piece of information would only be input once by any operator in the integrated system. This is an important concept to plan for in any integrated system.
6.2.2. Transit-Specific Recommendations
UTA cited the following general technical and institutional lessons learned, with recommendations provided as appropriate:
Transit #1: Expect a great deal of complexity in interfacing with the various network protocols and security infrastructures for multiple public sector agencies, in particular, given the sensitive nature of much of the subject matter for the messages. Not everything UTA thought it understood at the outset turned out to be correct, both technically and institutionally. There is no effective way to learn these things other than by working through them with the other agencies, and it is useful to understand that extra time and effort will be needed.
Transit #2: A technical example was the need to make various unexpected changes in UTA's messaging interface to accommodate the specific configurations of the messaging system interfaces developed later by other agencies. UTA did not anticipate the amount of time that would be needed for such adjustments to the configuration of its software.
Transit #3: An institutional example was the unexpected difficulty for dispatchers in being able to quickly interpret public safety agency incident messages, due to the various codes and jargon used.
Transit #4: For agencies that need to work with a vendor for the necessary enhancements to their respective CAD systems, it will be useful to establish strong working relationships and effective contractual mechanisms for ongoing technical support. It would be difficult to anticipate the specifics of all required vendor support for incorporation into system specifications. This leads to vendor support being needed for requirements that were not necessarily incorporated into UTA's original specifications. Since UTA developed and enhanced its software using in-house resources, it did not need to work with a vendor and did not experience this directly. However, several of the other agencies did need to work with their respective CAD software vendors to implement the changes, and this was UTA's observation on the effect.
Transit #5: Incident information generated by public safety agencies needed to be filtered and processed before being presented, for effective use by transit dispatchers. In their raw form, it was found that only some of these incident messages would affect traffic. In addition, the message description contained a range of information not needed by transit operations and in a format that was difficult to decipher. The filtering and processing could be performed either by a designated staff person, or by another agency such as UDOT. The purpose of this filtering and pre-processing for UTA would ideally be to (1) limit messages to those that could affect traffic in main corridors of the UTA service area; (2) provide a plain language description of the potential traffic impact location; and (3) distinguish between messages about new incidents and updates on existing incidents.
6.3 BENEFITS SUMMARY
The Benefits Summary presented in the Evaluation Plan identified the key metrics to be assessed during the course of the evaluation. The following benefits were identified:
#1. Enhanced field operations was associated with locating and responding to incidents. To a significant extent, this benefit was previously realized by Utah. UDOT and UHP had previously co-located staff at TMCs, and CAD terminals were placed in TMCs to enable data sharing. The most significant benefit realized by the project was the ability to engage in direct data exchange between legacy systems rather than having an operator observe two or more terminals. This real-time exchange of data adds to the benefits previously obtained through inter-agency cooperation and represents an additional enhancement of field operations and fills what had been a gap in the existing incident management and response program already in place in Utah.
#2. Geo-location for placing incidents and marginal improvement in scene clearance. Observed benefits included the use of Geo-location in providing a mechanism to place incidents without operator intervention, and from interviews, a qualitative assessment that scene clearance time seemed to improve marginally. Better traveler information allows the public the opportunity to bypass the incident which leads to less congestion and better response sooner (response units getting to the scene via a clear route). This logic seems sound; however, data was not available to support these conclusions.
#3. Enhanced communications among responders; enhanced on-scene activities. The evaluation was not able to completely assess this benefit. The system is newly deployed and while operational is still undergoing refinement. This benefit would be more accurately assessed when the system has matured and has been in use for a period of several years instead of several months.
#4. Enhanced efficiency in documenting the incidents. In the first 2 months of operation, UDOT increased documented incidents of 800 percent, as noted in section 4.2.1. The number of incidents for which the TMC maintained data increased significantly after the CAD-TMC integration. The main difference observed between the before and after data discussed above was that UDOT seemed to maintain much more complete incident records after the deployment, both in terms of the number of incidents recorded and the details recorded about each incident. It is believed that this increase is due in large part to the fact that CAD data was more readily available to TMC operators after the CAD-TMC deployment. This is supported, in part, by the large number of incidents in the after data for which Dispatch Services/9-1-1 were listed as the reporting agency.
#5. Improved data quality. The electronic data collection, particularly, recording the incident start and stop times, has significantly improved overall data quality. An additional example of this is reflected in a decrease in the error rate for the coding of incidents by type.
#6. Improved interagency working relationships. Utah had already achieved substantial progress in this area, and the project represented a continuation of this benefit. Utah's success in this area is represented by the inter-agency discussions on the amount and type of data that should be exchanged between the systems?the inter-agency cooperation that enabled this data exchange established the venue for addressing this type of system refinement based on initial deployment experience.
#7. Enhanced communication with the traveling public and media. This benefit would be more properly addressed at system maturity. While anecdotal evidence obtained during after project interviews indicates that enhanced communication is occurring, assessing this metric based on several years of implementation experience will provide a more accurate measure the benefit of enhanced communication to the traveling public and the media. From observations, efficiency in documenting incident management improved. This was presented in section 4.1, and was also an objective of the system impact study. Input for some fields was automated so UDOT operators did not have to enter this data.