Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
15 The objective of this synthesis project is to summarize the state-of-the-practice and state-of-the-art utilization of tech- nologies for improving safety data. One critical source of information for the synthesis came from a survey of state agencies. The surveys provided information not only on successful uses of technology, but also helped to identify critical gaps in technologies for acquiring, processing, and maintaining safety data. The survey design was significantly complicated by the number and breadth of data collection and maintenance activities undertaken by each of the surveyed agencies. Owing to the breadth of the survey content, three different surveys were developed and distributed to contacts in each state for each of the three core safety data areas (crash, road- way inventory, and traffic operations). The survey design included a set of 12 general database questions followed by a varying number of specific data and technology-related items specific to the type of database in question. The three surveys can be found in Appendixes C, D, and E. The road- way inventory survey (see Appendix D) contained elements related to both roadway inventory and pavement manage- ment, whereas the traffic operations survey (see Appendix E) covered Advanced Traffic Management Systems and traffic operations. Even though specific surveys were not distrib- uted to obtain information for citations and convictions, emergency medical services, and medical, driver licensing, and vehicle registration databases, information regarding such databases was included with the responses from states using electronic crash records systems. In total, approxi- mately 60 surveys from 34 states were received out of the total of 150 that were distributed. CRASH RECORD DATABASE SURVEY RESPONSES Crash report data are compiled for most crashes that involve fatalities, injuries, and property damage above specific state thresholds. Items of general interest include location, date, and time of crash; driver information including age, gender, and license; vehicle information including year, make, model, and registration; and environmental elements such as road design features, traffic controls, and weather. Responses to the crash survey were received from 24 states: Arizona, Arkansas, California, Connecticut, Delaware, Georgia, Illinois, Iowa, Maryland, Michigan, Missouri, Nevada, New Hampshire, North Dakota, Ohio, Oklahoma, Oregon, Pennsylvania, Texas, Utah, Washington, West Virginia, Wisconsin, and Wyoming. Figure 2 shows the geographic distribution of responding state agencies. Respondents from Departments of Public Safety or other Patrol Agency were nearly half (11 of 24) and the remainder (13 of 24) were from DOTs. A sum- mary of survey responses follows. It is not surprising that the majority of state crash records custodians are active in statewide safety initiatives. Of 24 responding states, 21 have a Traffic Records Coor- dinating Committee and 11 are also involved in a Safety Management System. Three states did not indicate involvement in either of these initiatives. Two states indi- cated that they were also Crash Outcome Data Evaluation System (CODES) states. Acquisition All 24 states use paper-based forms to report crashes. Of these, 17 also use some type of electronic form in conjunc- tion with a portable computer, and 3 of the 17 include elec- tronic data obtained from event data recorders (automobile blackbox). Six states, Arkansas, Delaware, Iowa, Mary- land, North Dakota, and Wisconsin, use Traffic and Crimi- nal Software (TraCS) as the electronic software for crash record capture, and the remaining 11 states use other elec- tronic crash report software. Of the 17 states that indicated collecting electronic crash records, 10 responded to the question regarding the percentage of total crash records captured electronically. One-half (5 of 10) reported that they capture less than 10% electronically, one stated approximately 30%, and the remaining four capture more than 60%. The two states capturing the most electronic reports are Nevada and Delaware with 75% and 85% cap- tured, respectively. A minority of states reported the use of various technolo- gies to support crash data collection. Two states use the mag- netic strip reader to record information from a driverâs license/registration, and seven states use a barcode reader to obtain data from a driverâs license/registration or vehicle identification number (VIN). Laser-based measurements of crash sites are used by three states, and video recording is used by six. Digital photography is in use in nine states, whereas aerial photography is only used in one. Wisconsin CHAPTER THREE STATE OF THE PRACTICE
16 locate crashes post hoc using a linear referencing system. The distribution of location data formats is provided in Figure 4. For severe crashes and special studies, several states (including California, Delaware, Iowa, Maryland, Missouri, New Hampshire, and Oregon) employ a variety of tools in addition to the technologies used for regular crash investiga- tions. For example, Delawareâs Fatal Accident Investigation and Reconstruction officers use laser management tech- niques and digital photography to record crash site measure- ments and crash details. Specialized crash reconstruction teams in Maryland and Oregon measure crash sites using survey-grade technologies. Most states indicated that data obtained for special studies are not included in general crash databases; rather, these are often filed in hard-copy format with the fatal crash reports. One last question regarding data collection referred to the types of training provided to police officers who collect crash record data. In most cases, officers receive 4â8 h of crash record training at the police academy. On-the-job or in-service training is utilized by some states. The state DOT or the state police department can organize this type of train- ing. Most states provide an accident-reporting guide, which can be used by officers in the absence of or supplemental to organized training courses. Safety conferences and the Insti- tute of Police Training are also sources for crash reporting education. uses all of these technologies except the magnetic strip reader. The distribution of technology use for crash reporting can be found in Figure 3. Most states use multiple methods of locating crashes. The most popular method is the text description, including street name and number and offset to closest cross street. The text description is used by 19 of 23 responding states. Alternative methods supplement the text descriptions in all but three states, where they are used as the primary meth- ods. The alternative methods include grid coordinates taken from map systems (electronic or paper), latitude and longi- tude coordinates taken from GPS receivers (handheld or in- vehicle), and linear references to route/milepost/reference marker systems. In the absence of data from one of the coordinate/reference systems, text descriptions are used to 1 1 2 3 6 7 9 0 1 2 3 4 5 6 7 8 9 10 Electronic mapping Aerial photographs Magnetic strip reader Laser measurement Digital video Barcode reader Digital photography FIGURE 2 Geographic distribution of states responding to crash survey. FIGURE 3 Reported technology use for crash data collection.
17 Processing All of the responding states maintain crash record data at a centralized state repository. Nine states also indicated that data are maintained at the jurisdictional or local level. Only one-third of the states that indicated maintaining crash record data at lower levels reported data sharing with the central repository. Currently, 10 of 24 states indicated acceptance of electronic crash record transmission to the central repository, and 3 other states are in the process of implementing electronic transmis- sion and receipt systems. All responding states indicated receipt of crash records in hard-copy format. Original hard-copy crash report forms are maintained in four states; one of the four scans the original to microfilm and another scans to a digital image file. In total, 15 states scan crash reports to digital image files, whereas 4 states still use microfilm technology. Eighteen of the 24 responding states have crash records databases that contain specific data elements that allow link- age to other databases. The distribution of linkages between crash records and other data sources is shown in Figure 5. Additional linkages were reported between crash records and hospital, bridge, and pavement records. Challenges or obsta- cles to data linkage include a lack of accuracy and exactness of data, amount of data entry time, a lack of necessary tech- nologies, out-of-date storage devices (i.e., mainframes and tape systems), inadequate information technology staff sup- port time, inadequate technical support and knowledge, and connectivity issues between database platforms. Storage and Maintenance In terms of data storage, the majority of the 21 responding states use Oracle (7, 33%), SQL (7, 33%), or DB2 (9, 43%) databases. Most states in transition indicated moving toward Oracle and SQL solutions. Some of these states also maintain mainframe systemsâpresumably for maintaining legacy data. Several states responding to the survey, however, still use mainframe systems as their primary data storage system. Access to crash records data is provided to users within the custodial agency in each of the 23 responding states. Fifteen states provide access to other state agencies, and 12 provide access to local agencies. Access is provided in a number of formats. The most common format for intra- agency use is a secure intranet site. External access is provided through secure Internet sites, unsecured Internet sites, FTP transfers, CD-ROM/DVD, e-mail, and hard copy. Crash Record Database Survey Summary All states maintain crash records in a centralized location to fulfill reporting requirements mandated at the federal level. However, a number of states indicated that redundant data repositories were maintained by local jurisdictions. A few 19 11 9 7 0 2 4 6 8 10 12 14 16 18 20 MilepostGridGPSText FIGURE 4 Distribution of crash location data format.
states noted that this occurrence was the result of strict data protections required by the state agency. Limited data sharing is occurring from a centralized crash database to other state agencies and local agencies. The survey of states indicated that the use of technologies for collecting data was limited, and so too were efforts to research new methods for data collection and maintenance. Although a number of states reported the capture of elec- tronic crash reports, few had widespread use throughout the state. A number of technologies exist within these electronic reporting systems; however, these too are not employed extensively. When technologies are used for special studies (i.e., laser or survey grade measurement and digital photog- raphy), much of the data is filed away in hard copy, limiting its long-term usefulness. Many states still use route and milepost linear referencing systems. States that have been moving toward linked crash records systems have discovered the usefulness of moving toward link node reference systems. In link node systems, links can be retired yet maintained with new ones without los- ing the ability to link the historical road characteristics records with historical crash data. Of the three main data components, crash records and road inventory are reported as being linked far more often then crash and traffic flow databases. A few states indicated that they were still using mainframe legacy systems for crash records storage and maintenance, 18 which severely limits their ability to perform analysis and link to other data sources. Several states indicated their frustration with the lack of funding available for obtaining technical sup- port, new technologies for crash data collection, and updating data storage devices. ROAD INVENTORY DATABASE SURVEY RESPONSES Roadway inventory data include the physical features within a roadâs right-of-way. These data include geometric data, cross-sectional elements, traffic control devices, and pavement-related data. All state highway agencies are responsible for collecting, processing, managing, and analyz- ing roadway inventory data. The primary uses of these data are for federal reporting, planning, and design. The survey of states indicated that roughly half of those that responded to the survey use roadway inventory data for input to safety eval- uation. Responses to the survey were received from 20 state DOTs: Arkansas, Colorado, Florida, Georgia, Hawaii, Idaho, Illinois, Iowa, Massachusetts, Minnesota, Mississippi, Mis- souri, Oklahoma, Oregon, South Carolina, South Dakota, Ten- nessee, Texas, Washington, and Wisconsin. The geographical distribution of responding states is shown in Figure 6. A summary of the survey responses follows. Unlike the responses from the crash records survey, fewer than half (9 of 20) of the respondents are involved with a Traffic Records Coordinating Committee; a similar number FIGURE 5 Reported linkages from crash records database survey (24 respondents). CMV = commercial motor vehicle; EMS = emergency medical services. 4 5 7 7 7 9 10 11 13 14 0 2 4 6 8 10 12 14 16 Traffic flow Citation and conviction CMV registration Medical records EMS run records GIS database Vehicle registration Driver history Driver's license Road inventory
19 (8 of 20) are involved with Safety Management Systems. However, several other safety initiatives were provided by respondents including 402 program initiatives, Local Com- munity Traffic Safety Teams, Toward Zero Deaths Commit- tee, Highway Safety Committee, and Hazard Elimination Safety Program. Acquisition The survey of state DOTs included approximately a dozen questions related to data acquisition. The survey indicated that most roadway inventory data is collected in the districts or by headquarters staff dedicated to this function. Sixteen respondents indicated that the data are maintained at a cen- tral state repository, whereas four states reported that the data are maintained in district or local repositories. Respondents stated that data maintained at district and local levels are shared with a central database. The vast majority of the states indicated that they do not participate in research efforts related to improving data col- lection technology. Fifteen states reported that they plan to deploy new technologies over the next 5 to 10 years, includ- ing video-logging, high-speed data collection, more extensive GPS usage, web applications, and road profiling. The timeliness of data varied greatly depending on what was being collected. All of the states conduct bridge inspec- tions on a two-year cycle. This is not surprising because the inspections are required every two years by federal mandate. Also, the procedures for conducting bridge inspections are standardized. Many states collect road inventory data on an annual basis for some characteristics, whereas a complete reinventory usually approaches 3 to 5 years. Several states had no set schedule, and one state indicated that it never does a reinventory. Table 1 provides information on collection samples for several categories of road characteristics data for limited access roadways for 18 of the 20 responding states. The table indicates how many states collect data comprehensively, versus sampling or collecting data on an as-needed basis. Similar data collection samples are used for arterial road- ways, whereas data for collector and local roads showed less comprehensive coverage with more sampling and as-needed data collection activities. For geometry data, several states (7 of 20) extract these data from as-built or construction plans. Seven states also indicated that they use sensor technology to collect certain attributes such as roadway grades or cross slope. Some data, such as horizontal and vertical curve information, is collected by post-processing raw spatial data, which is discussed in the next section. Five states use video-log, two use pen and paper, and two others do not collect geometry data. Road pavement data are collected annually on average. The survey indicated that many states had a great deal of automation in collecting pavement data for the stateâs pave- ment management function. Over half of the states reported that they used specialized vans to collect pavement data [six Automotive Road Analyzer (ARAN), one Mandli, and four custom-developed]. The survey responses noted that data collection is somewhat subjective for pavement distress data. Other measurements such as rutting depth, skid resistance, and structural strength are usually made with precise mea- suring devices. Some states collect these data continuously, whereas others collect the data as a sample (e.g., the first 500 ft of every mile). The survey indicated that pavement data were collected by both in-house staff and by contractors. The survey responses for data collection of cross- sectional elements such as barriers and clear zone mainte- nance were sporadic. Four states indicated that they have an extensive video-logging program for this. Others do field surveys using pen and paper and four stated that they do not collect these data. Much of this information would not be available digitally for most of the states with the exception of shoulder width and scanned as-built plans. The responses for sign and pavement inventories were similar to those for other cross-sectional elements. Most states relied on video-logs or as-built plans. Colorado and Washington noted that they use GPS technology for their sign inventory and add attributes. Only four states mentioned that they employ inventory sign retroreflectivity. The same four states indicated that they also collect pavement-marking retroreflectivity. Retroreflectivity values do not appear in a central database. Processing State DOTs use data processing for all of the areas listed previously. Processing of the data is required to calculate different performance indicators such as distress, skid, or rut- ting indices for pavements. Specialized software such as AASHTOWare (e.g., Trns*port, RoadWare, and Bridge- Ware) allows various functions including construction and maintenance, contract, bridge, and pavement management. All of the states that were surveyed have some level of GIS use. Florida indicated that it is able to estimate horizontal FIGURE 6 Geographical distribution of states responding to the road inventory survey.
curvature with their GIS application by processing centerline data. The GIS can establish the location of the point of cur- vature and point of tangency, as well as calculate a variety of curve parameters including the length of curve and the radius. Massachusetts indicated a similar capability, and also men- tioned that it estimates vertical alignment data using their GIS to drape centerline information over a digital terrain model. Neither state provided information regarding the accuracy of the data obtained from these procedures. Storage and Management The survey of state DOTs incorporated several questions with regard to the storage and management of roadway inventory data. Of the states surveyed, 80% have state repositories that are accessible relational databases. Many states (70%) indicated that they have electronic transmission capability to the central repository, but an even greater pro- portion (75%) noted that they still rely heavily on hard-copy 20 submission of at least some road data inventory items. Most of the original hard-copy data are archived. All states indi- cated that data are checked using either manual or auto- mated checks. States make data readily available; however, the method variesâ70% of the states make data available through the intranet, but some rely on FTP transfer, e-mail, and CD-ROM/DVD. Roughly half of the states with intranet data availability make their data available externally through secured or nonsecured Internet. Most externally available data are aggregated and available only at interval releasesâusually annually. The timeliness of the data varies; many states keep their road inventory databases current (or almost current) for state roads. Local road databases, if available, can range from current to more than 20 years old. Many states are on a 5-year cycle to ensure that all roads are inventoried and the data are accurate. Historical data for most of the states replying were available only on archives and most archive annually. Road Characteristics Data Collect Comprehensively Sample As-Needed Number of lanes 17 Lane width 17 1 1 1 1 1 8 6 6 6 7 Lane type 15 Shoulder width 17 Shoulder type 17 Edge treatments (SafetyEdge) 2 Median width 17 Median type 17 ROW width 7 Cross slope (normal crown) 4 Barriers (type, length) 9 Bridges 18 Railroad crossings 16 Multi-use paths/bike paths 7 Pedestrian facilities 2 2 Tunnels Grade 8 Vertical curvature 7 Horizontal curvature 9 Superelevation 3 1 Sight distance 2 2 Speed limit 15 1 2 4 Sign inventory 4 Truck/weight restrictions 6 Number of lanes/approach 1 5 Signal timing 1 2 2 Traffic control 5 1 1 1 1 Pavement material 15 Pavement distress data 10 1 Skid resistance 6 Ride quality 12 2 Pavement markings 1 Note: ROW = right-of-way. Pavement Elements Cross-Section Elements Roadway Structure Elements Geometric Elements Intersection Elements TABLE 1 ROAD CHARACTERISTICS DATA AND COLLECTION SAMPLE
21 Linkage to crash, traffic, and GIS was possible in all but a few of the states. Figure 7 provides a distribution of the reported linkages to multiple sources. Florida stated that database linkage was possible to a great number of state agency databases outside of the DOT. GIS usage with com- plete integration of roadway inventory and traffic data was reported by 9 states and limited integration for specific appli- cations was reported in 10 others. Other linkages with bridge, work zone, maintenance, project management, financial, and pavement monitoring systems were also reported by a small number of states. Management of road inventory data is accomplished through varying types of linear reference systems. The two primary linear reference systems used by state DOTs are route-milepost (75%) and node-to-node (25%). All states noted that they can link to GIS; however, few had complete GIS integration. Most of the GISs used by the state can link to either linear reference system method. Route-milepost data can be accessed using dynamic segmentation. Road Inventory Database Survey Summary As noted in the crash database survey summary, route-mile- post is the most common linear referencing system in use by state DOTs. Unfortunately, the route-milepost system does not easily allow for multiple years of historical road inven- tory data to be readily available for analysis. This makes linkage to historical crash records difficult. Roadway inven- tory databases are dynamic and constantly changing. Using the route-milepost system, there is no good way to manage change dates for specific pieces of information (e.g., the date that raised pavement markers were added to a section of roadway or the date when a traffic signal was added to an intersection). With hundreds of data fields for each record, managing change dates is a challenge. Most states archive the road inventory database each year, but not all roads are updated annually. Therefore, it is difficult to establish a link between road characteristic changes and safety improve- ments without chasing a significant paper trail of contract documents. Federal mandated data programs such as those for bridge structures, railroad crossings, and HPMS have a strong effect on statesâ comprehensive collection of data. For bridge and pavement data elements, all states reported collecting these data elements comprehensively, whereas for elements such as road geometric elements (horizontal and vertical curvature), less than half of the responding states mentioned collecting the data elements comprehensivelyâmost collect samples or on an as-needed basis if at all. Overall, a number of data elements required by safety programs (i.e., cross-slope, barriers, and intersection elements) are also underreported. According to the survey, technologies were used spar- ingly except for pavement management/video-log. Where federal programs require more comprehensive data, states 1 1 1 1 1 1 2 15 16 19 0 2 4 6 8 10 12 14 16 18 20 Citation and conviction CMV registration Medical records Vehicle registration Driver history Driver's license EMS run records Crash Traffic flow GIS database FIGURE 7 Reported linkages from road inventory database survey (20 respondents).
have adopted technologies to increase their efficiency with these data collection activities. However, many of the data elements are still collected manually with pen and paper or pulled from plan and profile sheets. Expansion of this type of data collection system would be cost prohibitive. Addi- tionally, much of the data collected in hard-copy format must be entered or scanned, requiring additional labor costs. Availability of AASHTOWare products for data manage- ment for federal requirements is seen as a benefit to the states. These software packages are typically designed to be flexible for use in most states, while allowing nationwide analysis of exported data. Further development of the antic- ipated TSIMS, while daunting, is expected to be an impor- tant venture. TRAFFIC OPERATIONS DATABASE SURVEY RESPONSES Technologies used for traffic data acquisition, processing, storage, and maintenance requirements vary by their intended applications. Although the survey was intended to query states regarding specific traffic operations databases, most states reported on the use of technologies for traffic operations data collection in general or with respect to the pavement monitoring system or advanced traffic management systems. Many of the problems in reporting arise because most states do not have one centralized database for collection of traffic operations data. Given these disparities in reporting, the following survey summary provides examples of acquisition technologies as related to specific applications. General trends are provided when applicable. Survey responses were received from 16 state DOTs: Colorado, Connecticut, Florida, Georgia, Hawaii, Illinois, Iowa, Massachusetts, Minnesota, Missouri, Oregon, South Dakota, Tennessee, Texas, Washington, and Wisconsin. Figure 8 shows the geographical distribution of states responding to the traffic operations survey. Acquisition Traditionally for planning, design, and evaluation different data acquisition tools have been applied for various data 22 types. Only a few survey respondents identified any specific types of data collection technologies. Missouri reported sat- isfaction with nonintrusive sensors for collecting traffic data. Examples of technology-oriented data acquisition tools, compiled from reviews of literature and survey responses, are provided here. Electronic count boards (with computer interfacing capa- bility) for counting intersection turning volumes have been used extensively. Additionally, the following automated detection systems can provide both count and presence data: inductive loops, magnetic detectors (two types: two-axis fluxgate and inductive coil), microwave radar, active and passive infrared sensors, ultrasonic sensors, acoustic sensors, and video image processors. In the existing systems, mag- netic detectors, passive infrared, and ultrasonic sensors are not able to generate classification data. When asked which of these technologies are currently being used by the states, only a few states indicated using technologies beyond count boards, road tubes, loop detectors, and video detectors. Tables 2 and 3 provide detailed usage summaries for each technology as related to the collection of speed, volume, and occupancy traffic data elements in both permanent and mobile applications. A few states indicated participating in a pooled-fund study on Portable Non-Intrusive Traffic Detec- tion Systems. FIGURE 8 Geographical distribution of states responding to the traffic operations database survey. TABLE 2 PERMANENT TRAFFIC MONITORING TECHNOLOGY USE: TECHNOLOGY DATA Technology Speed Volume Occupancy Manual Count Boards 8 4 Road Tubes, Pneumatic Counter 8 12 7 Video Detectors Radar Detectors 1 2 1 Laser Lidar Detectors Acoustic Detectors Magnetic Detectors 1 1 1 Microwave Detectors 1 1 Ultrasonic Detectors Active/Passive Infrared Sensors Automatic Vehicle Identification Vehicle Probes TABLE 3 MOBILE TRAFFIC MONITORING TECHNOLOGY USE Technology Speed Volume Occupancy Manual Count Boards 1 Road Tubes, Pneumatic Counter 1 1 1 Loop Detectors 12 12 7 Video Detectors 2 3 Radar Detectors 4 5 Laser Lidar Detectors Acoustic Detectors 1 1 Magnetic Detectors 1 1 Microwave Detectors 1 1 Ultrasonic Detectors Active/Passive Infrared Sensors Automatic Vehicle Identification Vehicle Probes
23 Handheld radar meters have been used extensively to con- duct speed studies. Automated detection technologies include inductive loop detectors, magnetic detectors (two types: two- axis fluxgate and inductive coil), microwave radar, active and passive infrared sensors, ultrasonic sensors, acoustic sensors, and video image processors. Of the traffic operations data elements surveyed, the most comprehensively collected element is traffic volume. This makes sense because it is required for HPMS report- ing. The next most frequently collected element was vehi- cle classification, followed by speed and vehicle weight. Table 4 identifies several traffic operations data elements and summarizes statesâ responses to whether the data are collected comprehensively, sampled, or collected only on an as-needed basis. For travel time and delay studies, automated means of data collection includes test cars equipped with distance measuring instruments that are connected to a laptop or hand- held computer. Automated detector systems used for speed and presence data acquisition can also be applied for collect- ing travel time data. Processing Software is the primary means of processing the raw data collected in the field. Two types of processing may take place: processing of sensor data to convert into traffic flow measure (e.g., the use of presence data at two sensors of known distance to compute speed) and processing of traf- fic flow measures to identify parameters important for con- trol or other management functions (e.g., processing of speed and volume data to identify whether an incident has taken place). Vendors that are supplying the automated detectors usu- ally supply the software required for processing the raw data to convert into required traffic flow measures. Processing of the traffic flow measures to generate decision support input is done through an algorithm developed by an agency or through off-the-shelf software purchased by an agency. In addition, a major area of transportation research has been in developing these decision support algorithms using deter- ministic-, stochastic-, and artificial intelligence-based mod- els. Several simulation models, including DYNASMART, DYNAMIT, INTEGRATION, VISSIM, and PARAMICS, have been developed to support real-time decision making by processing the traffic flow measures acquired by the sensors. According to the survey responses, several states use software, either off-the-shelf or developed in-house, to process data collected by the sensors. Iowa uses software developed in-house to process manually collected turning movements. Missouri and Wisconsin use off-the-shelf soft- ware, TRADAS, for automated processing of raw data, per- forming quality checks, summarizing the data, preparing reports, and managing the data. Florida uses polling soft- ware, developed by a consultant, to continuously manage traffic monitoring sites. Florida reported that they have provided this software to the South Carolina DOT. Florida made enhancement to a survey processing software, origi- nally developed in-house, to process short-term (seasonal, portable, noncontinuous coverage) traffic counts and vehi- cle classification surveys. Their software can read from a number of traffic counters (Peek, Diamond, PAT, Mitron, Numetrics, and ITS) and transform data to a common format for processing. The number one complaint identified in the state surveys was related to proprietary processing software. Storage and Maintenance As mentioned earlier, Missouri and Wisconsin use TRADAS software to manage their traffic data. The efforts are on-going in the real-time traffic operations, such as those included in intelligent transportation systems (ITS) applications for archiv- ing the data collected to support real-time decision making and reporting. Traffic Operations Database Survey Summary Given the difficulties that the states had in responding to this survey, it is clear that additional efforts to define a more com- prehensive system for traffic operations data collection and maintenance are needed. Several states indicated that multiple databases were maintained to house traffic operations dataâ some for ITS data archives, some for traffic signal data, and others to support volume requirements for HPMS. One of the overarching requirements of both IHSDM and SafetyAnalyst is the need for volume data. More integrated and effective data systems will be needed to meet these demands. Although several states mentioned plans for deploy- ment of portable nonintrusive traffic detection systems in the near future (5â10 years), few states have been early adopters. The use of portable nonintrusive technologies will be required to expand the traffic data needs of future safety analysis programs. Traffic Data Collect Comprehensively Sample As-Needed Speed 2 6 5 Volume 10 4 2 Vehicle Classification 6 8 2 Vehicle Weight 4 6 2 Axle Load 1 6 Axle Spacing 1 7 1 Occupancy or Density 1 1 Origin and Destination 1 1 2 Air Quality 0 1 0 Pedestrian Counts Traffic Control 2 TABLE 4 TRAFFIC CHARACTERISTICS DATA AND COLLECTION SAMPLE
