Prioritizing forest road maintenance using spatial multi-criteria decision making and the Unsurfaced Road Condition Index

Introduction: Prioritizing maintenance needs in forest road maintenance operations is one of the most challenging aspects of road management. On the one hand, available budgets are often insufficient relative to maintenance costs, and on the other hand, the unpredictable conditions of forest environments create a high risk of new road problems occurring and disrupting planned schedules. This study aimed to prioritize forest road maintenance operations and compare the efficiency of two common approaches—Spatial Multi-Criteria Decision Making and the Unsurfaced Road Condition Index—for prioritizing maintenance needs.
Method: The study area was the second district of the Kheyrud Educational and Research Forest, where 11,720 m of the main road network was divided into 50 segments based on the spatial locations of culverts. Field data for both methods were collected simultaneously. In the multi-criteria approach, the effective criteria were identified using the Delphi method and then weighted through the Analytic Hierarchy Process (AHP). The criteria were integrated in GIS to produce a road maintenance suitability map. In the URCI method, the type and density of seven major road problems were measured in each segment, and the condition index was calculated after completing the required computations. The Wilcoxon signed-rank test was used for the pairwise statistical comparison of the two methods, while Cohen’s Kappa coefficient was employed to evaluate the level of agreement between the results of the two approaches.
Results: After completing three rounds of the Delphi process, expert opinions converged and the criteria for prioritizing road maintenance were identified. The results of the multi-criteria process showed that drainage had the highest weight in prioritizing maintenance operations, accounting for approximately 30% of the total weight. Only 22% of the road segments were classified in the high and very high maintenance-priority classes, indicating that most of the network was in relatively acceptable condition. The URCI results revealed a high surfacing quality, with 94% of the segments classified as excellent or very good. The Wilcoxon test indicated a statistically significant difference between the overall results of the two methods. Although descriptive results showed agreement between the two methods in 34% of the segments, this agreement could be attributed to chance. Cohen’s Kappa analysis further demonstrated that the agreement between the segment classifications of the two methods was weak and statistically insignificant.
Conclusion: Overall, the URCI method primarily evaluates existing and observable road problems, whereas the multi-criteria approach, depending on the selected criteria, emphasizes the potential and likelihood of future deterioration. Consequently, the influence of previous maintenance operations is less evident in the multi-criteria approach, which may explain the discrepancy between the results of the two methods. The URCI method can therefore serve as an accurate basis for assessing current road conditions and supporting short-term planning and on-demand maintenance strategies, while the Spatial Multi-Criteria process is more suitable for preventative maintenance strategies, optimal allocation of limited resources, and the consideration of multiple technical and environmental objectives.