Mn/DOT demonstrated a particularly strong link between pavement management and the Department's long-term planning activities. Minnesota is not unique in the fact that it has several metropolitan areas that are heavily populated and some very rural areas with that are sparsely populated. This demographic has had a significant influence on the Department's investment decisions, including decisions regarding how much to invest in rural roads. Since the District personnel are typically closer to politicians than central office personnel, they have a strong political base to support project and treatment decisions that is not available in the central office. Therefore, the Department is constantly trying to balance its desire for consistent standards and performance targets with the autonomous nature of the eight Districts. The development of a District plan was a deliberate effort to develop more consistency in the Department's planning and programming activities.
At Mn/DOT, pavement management information is an important foundation for the Department's long-term planning activities. In fact, pavement management has been used as a model for long-term planning and the bridge unit has been instructed to develop and implement the same types of forecasting tools currently available in pavement management. An overview of Mn/DOT's planning and programming process is provided in figure 2. Both the 20-year plans and the 10-year plans (including the first 4 years of the plan, which are updated annually) are based on outputs from the pavement management analysis conducted by the central office.
Investment levels to support the strategic plan are established to achieve specific performance targets established at the policy level for 1) safeguarding what exists, 2) making the transportation network operate better, and 3) making Mn/DOT work better (e.g., through improved efficiencies or better decisions). Within each of these strategic areas, specific policies are established with goals, strategies, and performance targets established for each. For pavement preservation, the performance target is set so 70 percent, or more, of the road network classified as Principal Arterial is in good condition (as shown previously in table 3). This is representative of the condition level that has been maintained over time and, since the public has been reasonably satisfied with this condition level, the Department feels it is a reasonable goal. A "reality check" is applied to ensure that the trunk system maintained by the State does not fall below the condition of the county road network or in neighboring states. This benchmark system has been very useful in preventing decision makers from lowering the performance target.
Figure 2. Mn/DOT's planning and programming process.
Mn/DOT reports that the strategic plan is driving its pavement improvement program, but it has taken time to "turn the ship" to align with this philosophy since a number of decision makers remember the availability of funds to support large expansion programs and haven't accepted the reality that current funding levels will not support the same types of programs. An important part of Mn/DOT's strategic plan is pavement preservation and Districts are told to put money in preservation activities first before other demands and the preservation program is fully funded to support the philosophy. In addition, each District is given a specific performance target to achieve and District Engineers are held accountable for meeting these targets. As a result of these performance-based District plans, Mn/DOT is better able to report "needs" to the legislature, and the available resources are better matched to key performance issues.
The level of acceptance for the information provided by pavement management is admirable. Mn/DOT reports that the Pavement Management Unit has regularly promoted the concepts to the decision makers so the principles are well understood. Pavement management has also made a point of garnering the support of the Material Engineers in each District and getting consensus on any changes that are made to the analysis models. Once the Material Engineers are on board, pavement management seeks the support of the District Engineers and by the time they reach the Planning and Programming Division, they have defused any questions or concerns in the analysis results. As a result there is a high degree of confidence in the pavement management system and an acceptance of the information provided for planning purposes.
UDOT is also facing the challenge associated with balancing limited resources with demands for capacity and preservation needs. The agency saw a surge in funding for high-profile projects associated with the State's hosting the Winter Olympics in 2002. However, provisions were never made for the maintenance and operation of these new facilities and so the agency has been placing more of a focus on pavement preservation in recent years.
Pavement management information is used to develop a 20-year program, which is translated into multiple 10-year programs for long-term planning. The information is also used in developing 5-year programs, with four years fully funded and the fifth year updated annually. Politics influences some of the projects that are funded, but the Asset Management Division makes regular presentations to the Transportation Commission to convey the impact of cost increases on the program, the current and projected network conditions, and the funding needs to achieve performance targets. Pavement conditions are reported to the Commission in terms of their Ride Index. Interstate conditions are currently above the condition targets, although the rest of the network is below the targeted condition. It is an ongoing challenge for Asset Management to determine what message should be conveyed to the Transportation Commission and how best to present the message. However, by keeping their decision process very transparent, the Department has been able to build credibility with the Commission over time.
Pavement management information has become increasingly important to both Mn/DOT and UDOT as competition for funding continues to increase and the cost of raw materials continues to rise much faster than the rate of inflation. Without the availability of the results of objective trend analyses and "what if" scenarios, political influences on project and treatment selection tends to more heavily influence the process than when funding is sufficient to address both political and agency needs. The role of pavement management on the project and treatment selection process in each of the host states is described further.
Mn/DOT is an example of a decentralized state, meaning that the Districts have a significant amount of autonomy in the project and treatment selection process. This has had a significant influence on the role of pavement management in supporting the decision process. In general, pavement management uses its HPMA program to predict pavement performance and to determine what types of treatments are needed in each year of the analysis. Although the Districts have a significant influence on the final selection of projects and treatments, the Pavement Management Unit has established checks and balances to ensure that the appropriate treatment is being placed to address any deficiencies that are identified. The components of the analysis are described separately.
There are two types of RQI performance models used in the HPMA analysis: site specific models and default models. The site specific curves are preferred, since they show the deterioration patterns of each individual section, as shown in figure 3. Each year, after the pavement condition surveys are completed, the Pavement Management Engineer reviews the performance models for each individual section to look for anomalies in the data or to determine where maintenance treatments have been performed. In addition, default curves are developed for each pavement type based on statewide historical data. Default models are used when site specific models exceed agency-established rules for the expected performance associated with different treatment types. For instance, if an overlay is expected to perform adequately for 5 to 10 years and the section-specific performance models shows 8 years of performance, then the section-specific curve is used. However, if the section-specific curve predicted performance of 15 years, the default model would be used. The performance of approximately one-third of the network relies on the section-specific curves.
Figure 3. Example of a site-specific performance curve.
In addition to modeling performance in terms of an RQI, performance models for individual distress are also developed since Mn/DOT uses individual distress quantities as a factor in recommending appropriate treatments. Both site-specific and default distress models are developed, in a similar fashion to the RQI models. The availability of distress models that estimate the percentage of each severity of distress present is an important factor in Mn/DOT's ability to incorporate preventive maintenance treatments into its pavement management analysis.
Periodically, Mn/DOT uses feedback from the field to update its default performance models using an external modeling tool called TableCurve 2D. Perhaps the most obvious outcome of this feedback loop can be seen in the performance curves used after a treatment has been performed. For example, as a result of its field investigations Mn/DOT has adjusted its RQI models to start at values less than a perfect score of 5.0.
Mn/DOT's pavement management software is used to evaluate preventive maintenance, rehabilitation, and reconstruction alternatives. The treatments listed in table 5 are currently considered in the analysis. Each activity is defined as a construction activity, rehabilitation activity, global maintenance activity, or localized maintenance activity. The type of activity impacts the predicted performance once the treatment has been applied. The HPMA model allows Mn/DOT to reset performance (following the recommendation of a treatment) using an equation, by setting a relative percentage improvement, by holding the condition for a period of time, or by reducing the amount of distress observed. The type of treatment dictates the approach used to reset conditions. For example, an equation that resets the indices to a perfect score can be used for reconstruction projects such as cold in-place recycling, where the original performance of the pavement has little impact on the performance of the treatment. However, for preventive maintenance treatments, where the pre-existing condition is very important, a relative improvement is used. Mn/DOT holds the condition of pavement sections where crack sealing is applied. Once the hold period is over, the pavement then reverts back to the original rate of deterioration. The distress reduction option is used with localized maintenance treatments such as patching.
The HPMA software has a tool to create decision trees that allows Mn/DOT the flexibility to modify the rules as policies and practices change. Every two to three years, representatives from the Pavement Management Unit spend a day in the field with the District Materials Engineer to review the types of treatments that are appropriate for randomly-selected sites. The results are compared to the rules used in the pavement management software to help calibrate the treatment rules to actual practice. In addition, this process helps build credibility in the system and results in better acceptance of the recommendations from the pavement management system. Several sets of decision trees have been developed so that different scenarios can be evaluated quickly. Mn/DOT is one of the few states that have developed decision trees for its preventive maintenance treatments in addition to rehabilitation and reconstruction treatments.
As a decentralized state, the Districts are heavily involved in the selection of projects and treatments. In a typical analysis, District-selected projects are imported into the pavement management system, along with estimated budget allocations, and performance results are analyzed in terms of the RQI, SR, and/or PQI. Where performance targets are not met with the resulting program, adjustments are made or additional funding needs are estimated. Preventive maintenance projects are programmed separately since the Statewide Transportation Improvement Program (STIP) lists a funding level for preventive maintenance rather than list specific projects. Recommendations for preventive maintenance treatments are provided to the Districts using the pavement management decision trees, and the Districts select the final set of projects that will be funded using the preservation funding. The Office of Materials and Road Research must agree that any projects funded with the pavement preservation funding are good candidates to help ensure that the funding is being used for its intended purpose.
In addition to the analysis conducted to develop the STIP, a 20-year maintenance and rehabilitation analysis is also conducted to support the agency's long-term planning and programming activities. In the long-term analysis, the optimal set of projects are selected based on a cost effectiveness ratio that takes into account the additional life associated with a treatment, the length of the project, and a weighting factor (to determine effectiveness) divided by the cost of the treatment. An optimization can be run to determine either the best use of available funding or the amount of funding needed to achieve certain performance targets.
There are two factors that influence the project and treatment selection process used by UDOT. First, the Department maintains a database that defines a planned set of strategies for every section, using time-based treatment strategies for different pavement types. While this database in no way dictates the treatments that will be applied, it provides Region personnel with guidelines that reflect the typical timing when different types of treatments are applied. As actual treatments are performed, the database is updated. However, the database is difficult to access and so it provides limited benefit outside the Regions. There are plans to replace this database with a new Pavements module as part of UDOT's new maintenance management system implementation.
The primary source of pavement management recommendations is the optimization analysis conducted using the pavement management system. Details about each of the analysis components are provided in the following three subsections (i.e., 1. Performance Models, 2. Treatment Rules, and 3. Analysis Approach). A steering committee comprised of Pavement Management staff from the central office and the Region Pavement Management Engineers was involved in the original development of the treatment rules and continues to be involved in any changes that are made to the models. This involvement of Region personnel has had a significant impact on the level of acceptance of the recommendations that are generated and has provided a solid basis for understanding the operation of the pavement management system.
Pavement performance models have been developed based primarily on engineering judgment for pavement families that contain pavement sections with similar rates of deterioration. Pavement families are defined based on pavement surface type (gravel, PCC, and HMA) and traffic (including interstate, high speed routes [> 50 mph], medium speed routes [40 to 50 mph], and low-speed routes [<40 mph]. In total, UDOT has nine pavement families. Deterioration models have been developed for each condition index within each family. For example, there are four performance models for the high-speed asphalt family representing the ride, rutting, cracking, and wheel-path cracking indexes.
A variety of treatment types are considered in the pavement management analysis, as shown in table 6. The Department continues to work on refining the rules for selecting each treatment, with current efforts focused on improving the PCC treatment rules.
The three different types of seal coats considered on HMA pavements were added within the last year to enable pavement management to better estimate project costs. The appropriate type of seal is selected based on project location (in terms of an urban or rural environment), functional class, and traffic volume. In general, seal coats are applied to pavements in good condition with indexes between 70 and 100. A high seal is applied to pavement sections with traffic in excess of 15,000 vehicles per day and a low seal is applied on pavement sections with less than 7,000 vehicles per day. Medium seals are used on pavement sections with moderate traffic volumes that fall between the other ranges. The decision process for selecting the appropriate seal is shown in figure 4.
Minor rehabilitation activities are generally recommended when pavement condition indexes are between 50 and 70 and the rehabilitation/reconstruction treatment is triggered when the pavement condition indexes fall below 50. In the past, rehabilitation and reconstruction treatments were triggered separately, but because of the low number of reconstruction projects being funded, they were recently combined into a single treatment in the analysis. Treatment costs are estimated using both recent bid prices and Region input. The pavement management system allows for an inflation rate to be applied to future costs and provides for costs to be differentiated for rural and urban situations.
Reset values for each treatment are based on an estimate of time for the road to return to the condition at the time the treatment was applied. Separate reset values have been established for each treatment and each index. For example, a low seal resets the conditions in five years while a medium seal might reset the conditions in seven years.
The pavement management software is used to conduct at least three types of analyses. For example, an iterative process is used to determine the recommended level of funding based on the projected conditions under each scenario. This type of analysis is conducted by inputting different budget levels into the analysis and evaluating the overall distribution of network conditions achieved. By comparing the results from several budget levels, a recommended funding level can be determined to meet system level goals and strategies.
Figure 4. UDOT's decision process for selecting the preferred seal.
Once funding levels are set, the pavement management analysis is used to set Region budgets from an assessment of needs in each Region. After Regional budgets are set, five years of candidate projects are recommended for funding using the outputs from the pavement management system. Regions either accept or justify the selection of other projects for the program, and the central office fits the proposed projects to funding availability by eligibility and makes the final allocations of funds to each Region. The Regions are responsible for managing their programs within the funding allocated to them. Depending on the Region, the projects selected by the Regions usually closely match those recommended through the pavement management analysis. Questions in data quality has limited one Region's match to about 50 percent, but most of the other Regions report a match closer to 70 or 80 percent.
The pavement management analysis results are used to develop project recommendations for the Orange Book, which includes pavement preventive maintenance projects and simple resurfacing projects intended to address functional improvements only, the Purple Book, intended to address minor rehabilitation, and the Blue Book, which funds major rehabilitation and reconstruction projects. Projects of all three types (Orange Book, Purple Book and Blue Book) can be funded using either state or federal funds, or may be funded by a combination of the two sources.
Information from the pavement management system is provided to the Regions to use as guidance in selecting projects and treatments that make the best use of available funds. To help aid the buy-in of Region personnel in the recommendations from the pavement management system, UDOT has offered 1-day training sessions, conducted field visits with Region personnel to review treatment recommendations, and involved the Regions in the refinements to the pavement management models. UDOT now reports that Regions are coordinating their program with pavement management and the project cost estimates are now more in line with the actual costs in the field.
Because of the limited funding levels available for pavement preservation in recent years, UDOT is developing a process for identifying certain routes as "Maintenance Only" sections in recognition of the fact that many low-volume rural routes were not a high enough priority to be funded for rehabilitation or reconstruction. Under this approach, these sections will be maintained using only patching and chip seals and the pavement management software will not recommend any other treatments. Although the strategy for incorporating these sections into the pavement management analysis has not been finalized, initial estimates indicate that as much as 20 percent of the system could fall within this category due to limited funding availability.