Pavement Management Applications Using Geographic Information Systems (2004)

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.

7 CHAPTER TWO PAVEMENT MANAGEMENT SYSTEMS AND SPATIAL ANALYSIS TECHNOLOGIES This chapter includes a brief introduction to PMS, GIS, and spatial analysis, and discusses how the technologies have been combined to enhance the highway management process. PAVEMENT MANAGEMENT SYSTEMS There are more than 6 million km (approximately 4 million mi) of public roads in the United States, of which approxi- mately 64% are paved. Pavement management is a business process that allows DOT personnel to make cost-effective decisions regarding the pavements under their jurisdiction. Two AASHTO documents provide a complete treatment of pavement management and PMS, including objectives, components, and benefits. The Guidelines for Pavement Management Systems, published in 1990, provides the ba- sic information needed to develop a framework for PMS (1). The 2001 Pavement Management Guide discusses in detail the technologies and processes used for the selection, collection, reporting, management, and analysis of data used in pavement management at the state level (4). Exten- sive information about the development, implementation, and use of PMS by towns, cities, and counties can be found in National Highway Institute (NHI) course 13426, Road Surface Management for Local Governments (5). Although in its broadest definition pavement management covers all phases of pavement planning, programming, analysis, de- sign, construction, and research (6), most implemented PMS are restricted to addressing pavement maintenance and rehabilitation (M&R) needs (4). PMS assist in provid- ing answers to the following questions (1): • What general M&R strategies would be the most cost-effective? • Where (what pavement sections) are M&R treatments needed? • When would be the best time (condition) to program a treatment? Because of increasing system and budget demands, more public accountability, and limited personnel re- sources, and in particular the GASB (Governmental Ac- counting Standards Board) 34 accounting procedures, state DOTs are changing their way of doing business and em- bracing an asset management business approach (7). Asset management is the term commonly used by business to de- scribe the systematic process of maintaining, upgrading, and operating physical assets cost-effectively, efficiently, and comprehensively (8). Under the leadership of the FHWA, state DOTs have realized the benefits of this ap- proach and are starting to reengineer their business proc- esses accordingly. Many agencies have focused attention on asset inventory and condition data integration, in many cases using a GIS for data integration, and are working on integrating management decisions of existing “stovepipe” management systems—such as PMS and bridge manage- ment systems—for executive-level decisions (7). In addi- tion, there is a trend toward supplementing subjective pol- icy-based decision making with objective, performance- oriented tools. PMS are one of the key components of asset management, not only because they provided the frame- work for their development, but also because they are the main business process and account for up to 60% of the to- tal assets in a typical DOT. A PMS has been defined as a “set of tools or methods that can assist decision makers in finding cost-effective strategies for providing, evaluating, and maintaining pave- ment in a serviceable condition” (1). A PMS provides a systematic process for collecting, managing, analyzing, and summarizing pavement information to support the se- lection and implementation of cost-effective pavement construction, rehabilitation, and maintenance programs (2). To effectively support these types of decisions, a PMS must include reliable and sufficient data; calibrated analy- sis models and procedures; and effective, easy-to-use tools that help visualize and quantify the impact of the possible solutions considered. Although the approaches used by agencies differ, the foundation of all PMS is a database that includes the fol- lowing four general types of data: 1. Inventory (including pavement structure, geometrics, and environment, among others); 2. Road usage [traffic volume and loading, usually mea- sured in equivalent single-axle loads (ESALs)]; 3. Pavement condition (ride quality, surface distresses, friction, and/or structural capacity); and 4. Pavement construction, maintenance, and rehabilita- tion history. Figure 1 shows the percentage of the responding states and provinces that collected or used each of these specific data elements. It was surprising that not all the agencies reported that they are collecting inventory data, because

8 FIGURE 1 Types of pavement management data collected. these data are necessary for supporting the other data col- lection activities. However, this may be because the re- sponsibility for collecting inventory data often does not re- side with the PMS office. PMS analysis capabilities include network-level and project-level tools. “Network-level” analysis tools support planning and programming decisions for the entire network or system. A PMS usually includes tools to • Evaluate the condition of the pavement network and predict pavement performance over time; • Identify appropriate M&R projects; • Evaluate the different alternatives to determine the network needs; • Prioritize or optimize the allocation of resources to generate plans, programs, and budgets; and • Assess the impact of the funding decisions. “Project-level” analysis tools are then used to select the final alternatives and to design the projects included in the work program. The pavement management cycle then con- tinues with the execution of the specified work. Changes in the pavement as a result of the work conducted, as well as normal deterioration, are periodically monitored and fed back into the system. From an asset management perspec- tive, the network-level goals and available budgets are de- fined by higher-level strategic decisions that set the overall goals for system performance and agency policies. PMS produce reports and graphic displays that are tailored to different organizational levels of management and execu- tive levels, as well to the public (9). Enhanced spatial capabilities for data storage, manage- ment, and analysis augment many of the aforementioned functions and tools. For example, GIS and other spatial tools can facilitate the following PMS functions: • Data collection and processing—GIS and global posi- tioning systems (GPS) could allow collecting data us- ing a coordinate-based method and relate the infor- mation to the base highway network. The display of inventory and condition data on color-coded maps may also facilitate data cleaning and gap detection. These maps can highlight contradictory or redundant information as well as sections with missing data. • Data integration—The use of database management tools that can handle spatial data can facilitate the in- tegration of the data used for supporting PMS deci- sions—inventory, pavement condition, traffic, and maintenance history—and is collected or stored in different DOT units. • Incorporation of spatial data into the PMS analysis— Spatial GIS tools allow users to efficiently overlay point and area data, which is not route specific, with the linear road network for PMS modeling. Examples include the use of weather or regional information in the development of pavement performance models, the computation of average treatment cost by district or region, or the use of land use and regional devel- opment models for enhancing traffic predictions. Spatial analysis tools can also facilitate grouping pro- jects based on geographic proximity or other criteria to obtain economies of scale or reductions in traffic disturbances. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% P er ce nt o f R es po nd en ts Road Inventory Pavement Condition Traffic Volume ESALs M&R History

9 • Output presentation—The user can easily generate color-coded maps and graphic displays depicting road conditions, coverage of evaluation campaigns, and maintenance and rehabilitation schedules, among many other applications. GIS can also facilitate the computation of statistics by areas or regions; for ex- ample, the average condition of the roads by county. These maps are an integral part of condition reports and work programs usually generated by the DOTs. It is for these reasons that many agencies have used, or are actively pursuing the use of, GIS and other spatial technologies for developing PMS applications. According to the survey of practice conducted for the preparation of this synthesis, 31 agencies (60%) reported that they are currently using spatial applications for PMS and 14 agen- cies (27%) indicated that they are not. An additional seven agencies (13%) provided conflicting information; although the PMS respondent indicated that spatial tools were not used or was unsure if they were used, the GIS representa- tive indicated that the PMS did used spatial tools. The dis- crepancies were resolved through follow-up telephone calls that revealed that although a GIS is not used to support PMS decisions, it is used to prepare maps and displays. Furthermore, seven of the agencies (50%) that are not cur- rently using spatial applications for PMS indicated that they have plans for their use. In addition to indicating if they were using spatial tools to support PMS activities, the survey asked each respon- dent to indicate the primary current and planned uses of GIS and other spatial applications. The responses to these questions are summarized in Figure 2. Almost all DOTs currently using these technologies (28) use them to prepare maps, and approximately half (15) use spatial database management tools to help them with data integration. A very limited number of respondents (5) indicated that they are using some of the spatial analysis capabilities. How- ever, the planned activities show a trend toward the use of the more advanced capabilities, such as data integration and spatial analysis. SPATIAL ANALYSIS TECHNOLOGIES Spatial analysis is broadly defined as a “set of methods useful when the data are spatial” (10). It consists of a series of transformations, manipulations, and other techniques and methods that can be applied to spatial data to add value to them, support decisions, and reveal patterns and anoma- lies that may or may not be immediately obvious. Spatial data consist of “geographically referenced features that are described by geographic positions and attributes in an ana- log and/or computer-readable (digital) form” (11). Spatial analysis allows users to create, query, map, and analyze cell-based raster data; to perform integrated raster/vector analysis; to derive new information from existing data; to query information across multiple data layers; and to fully integrate cell-based raster data with traditional vector data sources. NHI course 151039, Applying Spatial Data Tech- nologies to Transportation Planning (12), provides detailed coverage of the subject. The field of spatial analysis has grown significantly in recent years, thanks to the introduction of relatively inex- pensive and relatively easy-to-use GIS. More recently, FIGURE 2 Current and planned PMS applications of GIS and other spatial technologies. 0 5 10 15 20 25 30 Map Generation Data Integration Spatial Analysis N um be r of R es po nd en ts Current Planned

10 other spatially enabled databases and software components have been developed specifically for highway manage- ment. These software components, or middleware, sit be- tween the database that resides on a server computer and the end-user applications, and they provide many of the functions and procedures that an end-user application re- quires. Therefore, such middleware may provide savings in coding and the total cost and effort of building end-user applications in DOTs with respect to the traditional “from the ground up” approach used in the 1960s and 1970s. GIS A GIS can be defined as a system of computer hardware, software, personnel, organizations, and business processes designed to support the capture, management, manipula- tion, analysis, modeling, and display of spatially referenced data for solving complex planning and management prob- lems (13,14). Because any definition of a GIS represents a simplistic view of a complex system, the preceding defini- tion is provided only to illustrate the capabilities of the sys- tem. Additional definitions, more detailed information, and training materials on GIS can be found in the FHWA Dem- onstration Project No. 85: GIS/Video Imagery Applications (14) and NHI course 151029, Application of Geographic Information Systems for Transportation (15). For the pur- pose of this synthesis, the concept of a GIS is discussed as a “process” for integrating spatial data into the decision- making process, rather than as specific GIS technologies or software packages. A comprehensive GIS includes procedures for conduct- ing the following activities: 1. Data input, either from maps, aerial photographs, sat- ellite images, surveys, or other sources; 2. Data storage, retrieval, and querying; 3. Data transformation, analysis, and modeling, and 4. Output generation, including maps, reports, and plans. GIS link geographic (or spatial) information displayed on maps, such as roadway alignment, with attribute (or tabular) information, such as pavement structure, condi- tion, and age (Figure 3). Although many of the current ap- plications are limited to map generation, a major strength of a GIS is its ability to use topology (i.e., spatial relation- ships among features) to support decision making for spe- cific projects and/or limited geographic areas. A branch of geometrical mathematics, topology deals with spatial rela- tionships between spatial entities and is concerned with the connectedness, enclosure, adjacency, nestedness, and cer- tain other properties of objects that may not change when the geometry of objects change (15). It is in large part what makes GIS different from other spatial technologies; and it is vital to many GIS analysis operations (such as proximity, buffer, overlay, etc.). A comprehensive GIS includes three important charac- teristics. First, it includes a database management system (DBMS), which uses georeferences as the primary means of indexing information. This could be the DOT agency DBMS or a GIS vendor-supplied DBMS. Second, a com- prehensive GIS integrates spatial analysis functions that in- corporate statistical and conceptual models. This feature differentiates a GIS from traditional computer-assisted de- sign/computer-assisted mapping (CAD/CAM) tools. How- ever, recent versions of several CAD programs have incor- porated many GIS features. Spatial analysis methods allow FIGURE 3 GIS functional scheme. A B C Route: I - xx Section: A - B Pav . type: Flexible PSI: 3.9 PCI: 87 ADT: 5000

11 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1 2 2 4 2 2 1 1 1 1 2 2 2 2 2 1 1 1 1 2 2 3 2 2 1 1 1 1 1 1 3 2 2 1 1 1 1 1 1 3 2 2 1 1 1 1 1 2 3 2 2 2 2 2 2 2 2 3 (a) (b) FIGURE 4 Main GIS data structures: (a) Vector data structure; (b) Raster data structure (15). users to perform computations on data groups or layers and to view relationships that would otherwise not be obvious. Third, with its vast array of functions, GIS should be viewed as a process rather than as merely software or hard- ware. The way in which data are entered, stored, and analyzed using GIS and other spatial tools must mirror the way information will be used for a specific research or de- cision-making process, such as pavement management. The manipulation of attribute data is performed by means of a DBMS, which comprises a set of programs that manipulate and maintain the database attributes and geo- metric objects. Current GIS data structures include raster and vector data structures (Figure 4). Raster data structures are defined by dense arrays of values that represent fea- tures requiring large storage capacities and a lower nominal spatial resolution or byte (e.g., digital pictures and satellite images). Processing raster data involves massive element- wise calculations. In contrast, vector data structures are represented by nonuniform, sparse sets of vertices that re- quire less storage by delineating features. Vector data have a higher nominal spatial resolution or byte, but require a com- plex two-level, arc-node data structure to manage gap-and- overlay problems. Processing involves more complex data manipulations, including numerical integration. Most GIS incorporate both raster and vector functions (15). In vector data, there are three basic geometric or foun- dational elements that are currently used: points, lines, and polygons. Points are defined by single vertices and are used to represent features such as cities and intersections. Lines are defined by nonclosed sets of vertices and are used to represent linear objects, such as roads and power lines. Polygonal areas or regions (e.g., counties or DOT districts) are defined by closed sets of bounding vertices. For transportation applications, lines or arcs are usually also combined in routes and networks. The choice of data structure depends on what is being analyzed, the applica- tion requirements, and the spatial resolution required; most highway management and PMS applications use vector data structures. Most GIS manipulations of spatial ele- ments involve predefined package theory—overlay, split, buffer, and point-in-polygon—that are basically union, in- tersection, and membership operations that take advantage of the topological relationships between objects. The at- tributes of a feature describe or characterize the feature. A GIS can assist in the analysis of many planning and operational problems, such as pavement management, which varies by scale, time, and format, while allowing the enhancement of measurement, mapping, monitoring, and modeling of spatial phenomena. GIS have been used in civil engineering applications for data handling, modeling spatially resident engineering phenomena, and result inter- pretation and presentation (16). Moreover, the ability to ef- ficiently integrate, store, and query spatially referenced data is probably the most compelling reason for using a GIS. Other PMS applications focus on the presentation of analysis results in map form or take advantage of the spa- tial operations that are included in current GIS software to support many pertinent decision processes. GIS-T Current DOT practices are shifting their business processes toward the use of integrated asset management systems for making strategic, agency-wide resource allocations and work programming decisions (8). For this reason, there is an increasing demand for means to integrate the great vari- ety of data collected and used by transportation agencies. Given the geographic distribution of the transportation as- sets, a GIS is one of the technologies of choice for facilitat- ing this process. Many agencies and organizations have supported these developments. AASHTO, along with other agencies, has sponsored annual GIS for Transportation Symposiums that offer forums for persons in government and private industry who are interested in the use of GIS-T Lake TI C (Registration Point) LINK (Line, Arc) NODE POLYGON LABEL POINT ANNOTATION

12 FIGURE 5 GIS-T as the merger of enhanced GIS and enhanced transportation information systems (18). opportunities to gather and share experiences, review state- of-the-art software, and learn more about this field. The proceedings for these symposiums are available electroni- cally (17). NCHRP has sponsored a series of research projects to define the basic structure of a GIS-T. NCHRP Report 359: Adaptation of Geographic Information Systems for Trans- portation (18) provided the framework for the adaptation of GIS-T (Figure 5). This project recommended a “corpo- rate” or enterprise-wide approach for information system planning and GIS development within a DOT, as well as a series of GIS enhancements relevant to its application for transportation management and operations. These include enhanced measurement tools; proximity analysis; raster processing; surface modeling; network analysis tools, such as dynamic segmentation and network overlay; and poly- gon overlay capabilities to link superimposed layers (18). Many of these capabilities have since been included in commercial GIS packages, as well as in other specialized highway management tools and database management sys- tems. NCHRP Project 20-27(2), Development of System and Application Architectures for Geographic Information Sys- tems in Transportation, defined a generic information ar- chitecture for the implementation of GIS-T and proposed a robust location referencing system data model (19,20). Furthermore, because the state DOTs are focusing increas- ingly on managing the entire life cycle of facilities and co- ordinating activities with other private and public organiza- tions, there is an increasing focus on referencing the data both spatially and temporally. NCHRP Report 460: Guide- lines for the Implementation of Multimodal Transportation Location Referencing Systems refined this model to ac- commodate the elements necessary to use, store, operate, and share multimodal, multidimensional, spatiotemporal transportation data (21). The following core functional requirements were identified as needs for an object- oriented location referencing system to support highway management and Intelligent Transportation Systems devel- opments: • A spatial referencing method that helps locate, place, and position processes, objects, and events in three dimensions to the roadway; • A temporal referencing system and datum to relate the database to the real world; • Transformations among linear, nonlinear, and tempo- ral referencing methods without a loss of spatial or temporal accuracy, precision, and resolution; • Multiple cartographic and spatial topological repre- sentations at different levels of generalizations of transportation objects; • Display and analysis of objects and events at multiple spatial and temporal resolutions; • Dynamic navigation of objects in near real time; • Regeneration of objects and network states over time and maintenance of the network event history; • Association of errors with the spatial temporal data; • Object-level metadata storage to guide the general user in interpreting the data; and • Identification of temporal relationships among ob- jects and events or temporal topology. Several of these functional requirements are important for pavement management. For example, it is important that a roadway segment can be presented as a centerline or as a two-dimensional or three-dimensional spatial object, depending on the scale being used. Similarly, the road seg- ment may be more appropriately represented by a node, link, or polygon for modeling purposes, depending on the application being used. The ability to handle different ref- erencing methods is needed to integrate data collected us- ing different referencing methods. Spatial and temporal considerations are important when considering perform- ance trends, work programming, and life-cycle cost analy- sis, among other applications. PAVEMENT MANAGEMENT SYSTEMS’ USES OF SPATIAL TECHNOLOGIES As with any business process, pavement management needs an efficient DSS to be effective. A DSS is a system GIS + + Enhancement TIS GIS-T

13 that provides managers with additional information to help them make better informed decisions as they allocate scarce departmental resources (22). This DSS may include procedures and tools for information retrieval and display, filtering and pattern recognition, extrapolation, inference, logical comparison, and complex mathematical modeling. GIS and other spatial technologies can facilitate and en- hance the preparation, analysis, presentation, and manage- ment of data used for supporting these decisions. One of the first GIS applications for highway manage- ment was the FHWA’s National Highway Network data- base, which was developed using 1:2,000,000 scale U.S. Geological Survey maps (23). A PMS was identified early on as one of the areas that could potentially reap great benefits from the use of GIS (24,25). After developing a prototype PMS–GIS system, Osman and Hayashi (26) identified the following advantages of using GIS for PMS: the possibility of automatically generating maps; enhanced analysis capabilities through powerful spatial queries; en- hanced data availability, quality, and integration; and easier consideration of other road assets in the decision process. There are at least four possible of spatial applications for supporting PMS: map generation and presentation, as- sisting with data collection, data integration and manage- ment, and geospatial analysis. In the early 1990s, Petzold and Freund (27) identified two main reasons for a highway agency to have a GIS: map/display and data integration. At its most basic level, a GIS allows data to be visualized quickly in many ways, on both graphic screens and plotted maps. It is possible to zoom in and out on a map display and to show the objects in the database color-coded by grouping or highlighted by selected attributes. A GIS can also be very effective in facilitating the integration of the large amounts of data that are collected and maintained by transportation agencies. A GIS can be a natural way to re- late highway databases because they are all spatially re- lated. However, current spatial analysis packages can do more for a transportation agency. They can rapidly answer questions about how data are spatially related or which data have common or related attributes, conduct network analysis, and perform dynamic segmentation, among other features. Figure 6 shows the status of the integration between GIS and other spatial analysis technologies and pavement man- agement within the agencies surveyed. The respondents were asked to indicate how they would best describe the level of integration between the PMS and GIS tools used. Most agencies indicated that their applications fall in the first category (i.e., the GIS is used mainly for preparing maps and graphic displays). In addition, some agencies are using GIS or other spatial technologies to integrate data and to manage the central enterprise-wide databases. A small percentage of the respondents also indicated that they are using a GIS as the main database for the PMS. It is interesting to note that most available commercial PMS software packages provide GIS interfaces but use other standard DBMS. Only one agency, the Wisconsin DOT, indicated that the spatial tools are fully integrated with their PMS. That PMS was developed using a GIS plat- form. The applications are presented in the following chap- ters: chapter three presents spatial applications for support- ing data collection and integration and chapter four reviews applications for map generation and spatial analysis. FIGURE 6 Status of the PMS–GIS integration. 0 5 10 15 20 25 GIS Used for Mapping GIS Manages PMS Database GIS Manages DOT Database(s) GIS Fully Integrated with the PMS N um be r of R es po nd en ts `