Monday 26 April 2010

Development of a Prediction Model and Prioritization of the Expressway Maintenance Management System

Japan International Cooperation Agency (JICA) established the guideline of inspection and maintenance management system for the expressway system of Thailand since 1994 (JICA, 1994). However, ETA did not implement the inspection and maintenance system during that period due to limited resources. The Expressway inspection and maintenance section of ETA started to implement only the inspection database system since 1999 by following the JICA guideline.

Preserving the Expressway condition to be at serviceable level is one of the main responsibilities of ETA. Previously, Expressway in Thailand was maintained, but not managed with future planning of maintenance. Expressway Management System (EMS) could be introduced by using both the concept of Pavement Management System (PMS) and Bridge Management System (BMS).

ETA has been performing the maintenance policy by when need basis. To develop an effective expressway management system ETA should have future maintenance plan. The prediction model is needed for future maintenance policy. The ETA needs to set the minimum acceptable level of the expressway condition to determine the time when the maintenance should be performed. Especially while resources are scarce, it is more crucial that the prediction of deterioration must be effective to propose the most appropriate plan of maintenance alternatives. Also it needs to be prioritized of deteriorated expressway sections to maximize the available resources.

Pavement management activities must consider the factors that influence costs of maintenance, in addition to the performance improvement of the expressway system. The major costs of each alternative treatment includes i) cost of the actual work (materials, transport, placement, etc.), ii) vehicle operating costs (VOC), iii) cost of user delays (including extra vehicle operating costs during maintenance), iv) accident cost due to traffic hazards or interruptions associated with the maintenance, v) Environmental damages (Air or water pollution, noise, etc) and these costs are calculated in the present worth value (Haas et al., 1994). When a maintenance strategy is selected for a section in a year, the marginal cost effectiveness is calculated for the other candidate sections for ranking the candidate sections in each year with budget limitation.

Mr. Mohammad Shahidul Islam made a study which main objective of this study is to determine a maintenance prioritization policy based on the concept of single-year optimization for the Thailand Expressway System

However, his study focuses on the development of deterioration model of each expressway section. And on the prioritization of roadway sections by single-year optimization based on a heuristic, marginal cost effectiveness method. This technique for the prioritization of expressway sections is much simpler, flexible and operationally efficient, and provided results close enough to the actual optimization for all practical purposes (Haas et al., 1997).

·His secondary objectives were to: (1) determine the structural weight factors of the structural elements of the Expressway system; (2) determine the numerical condition ratings corresponding to the alphabetical damage ratings for developing the prediction model; (3) determine the Expressway Performance Index (EPI) for each section of the expressway system; (4) study the characteristics of each type of prediction models and determine the appropriate prediction model for Thailand Expressway System; and (5) determine the effectiveness of each treatment alternative for each expressway section and year combination.

Conclusions

In this study, the deterioration model has been developed for each section. An overall model has been also developed for each expressway system but for each route of the First Stage of Expressway system. The deterioration models have been developed by using the Expressway Performance Index (EPI), which considers the condition of elements together with the structural and functional importance by using the structural weight factors.

The structural weight factors have been developed from the ETA maintenance experts’ opinion based on the Analytical Hierarchy Process (AHP). In the AHP, those weighting factors were used to develop the overall structural weight factors (W.F.), which comparison matrices had the consistency rating of 0.1 or less. It is concluded that the sub-structural elements for the bridge sections and pavement elements for the land sections have the greater importance than the other types of elements.

In addition to the questionnaire survey for structural weight factors, another questionnaire survey was performed on the ETA maintenance engineers for assigning a numerical value to the alphabetic condition ratings of the expressway elements, which have been practicing by ETA. Based on the average and the modal values, the used numerical condition ratings for ‘B’ and ‘C’ are 3 and 7 respectively. The lowest value for the worst condition and the highest value for the no damage condition were used as zero and 10 respectively (i.e. A=0 and D=10).

The model validation was performed based on the coefficient of determination (R2 Statistic) and the correlation test between the actual EPI and the predicted EPI for each expressway system and each route of the First Stage of Expressway system. It was found that the coefficient of determination (R2 Statistic) of each sectional model is close to one which is reasonably acceptable. It was also found that the overall model developed for the Ram-Indra Atnarong Expressway is the best, which coefficient of determination is 0.8113 and the Spearman’s correlation coefficient (rs) between actual EPI and predicted EPI is 0.939.

In the decision module, the single-year optimization based on the Marginal Cost Effectiveness (MCE) single-year optimization has been performed to select the expressway sections for maintenance with a limited budget. In this study 100 Million THB per year was used as the maintenance budget which is the average yearly maintenance budget of the Expressway and Rapid Transit Authority of Thailand (ETA). All the maintenance methods were grouped into five maintenance packages for the maintenance decision analysis. These are 1) do nothing, 2) preventive maintenance, 3) minor corrective maintenance, 4) major corrective maintenance and 5) replacement of deteriorated elements. The effectiveness of each combination of section, maintenance alternative and the implementation year was calculated by using the respective sectional model. However, the overall model of the corresponding expressway system was used for those sections which were eliminated by the data manipulation and filtration process from the model development.

The MCE prioritization result shows that 661 sections of the total 506 candidate sections in each year would be maintained within the 5-year programming period with a limited budget. The average annual maintenance budget of the Expressway and Rapid Transit Authority of Thailand (ETA) is 100 Million THB. The annual effectiveness has been increased with year as the needs of candidate sections increases. The cost-effectiveness ratio (CE) has been also increased, in the first year of the programming period the CE is 0.0312 and the CE has been increased to 0.1745 at the fifth year. The total effectiveness would be achieved by maintaining the 661 candidate sections with the four maintenance options is 52,162,235 with the total cost of 499,760,000.00 THB and the total cost-effectiveness ratio (CE) is the 0.1044.

From the MCE prioritization result, it can be concluded that the preventive maintenance and the minor corrective maintenance are the more cost-effective than the others for the early stage of the programming period. However, for the late stage of the programming period, the major corrective maintenance and the rehabilitation are the more cost-effective than the first two maintenance options.

His thesis abstract is copied and posted.

ABSTRACT

Maintenance management system plays an important role for infrastructures after the infrastructure facilities are constructed. The expressway system is one of the infrastructures that provides road network for road users in Thailand. An expressway road will not last for an extended period of time without application of some type of maintenance, rehabilitation, or reconstruction. Expressway Maintenance Management System (EMMS) is a process for maintaining and preserving expressway at a certain level of performance in the most cost-effective manner. The future plan for maintenance of expressway system is highly affected by the future condition and the budget allocation of the Expressway and Rapid Transit Authority of Thailand (ETA).

Due to several constraints such as budget, time, manpower, equipment and other resources, it is not possible for the agency to maintain all the road sections or structural components that require maintenance within the same time period. In such cases, prioritization has to be set on which sections or components of the expressway system are to be repaired, what maintenance policy will be applied and when the repairs should be done. The prioritization of expressway sections and selection of maintenance alternatives should be such that the available resource utilization can be maximized.

In this research, the structural weight factors (W.F.) have been developed for each element of the expressway system. The structural weight factors have been developed based on the Analytical Hierarchy Process. The numerical condition ratings have been also developed for each alphabetical damage rating of the expressway elements, which have been practicing by ETA for the inspection process.
In this research, the prediction models for each 200-meter section of the expressway systems have been developed in order to develop an effective Expressway Maintenance Management System (EMMS). The single-year prioritization of the expressway sections and selection of maintenance options has been done based on the Marginal Cost-Effectiveness (MCE) for the five year programming period.

No comments: