254 surface mine, any savings generated from improved road mainte- nance benefit the mining company directly as a reduced cost per ton of material hauled. A maintenance management system (MMS) approach provides a partial solution to mine road maintenance (1, 2), but for either complex mining road networks or highly variable construction and material qualities, the MMS becomes onerous. In addition to repeated remodeling of a large number of road segments, many of these segments may be used infrequently and as a result each re- iteration of the MMS program for the altered hauling scenario would have to account for the progressive degeneration of functionality on these infrequently trafficked segments. This situation would be problematic since in a dynamic mining environment—typically those mines in which production is managed by a centralized truck allocation system—there is no guarantee that the traffic vol- umes modeled in the MMS would be realized before a change was made to the system. The result would likely be the application of suboptimal road maintenance strategies with the attendant increase in total road-user costs. A further disadvantage of the system for complex networks is the necessity to communicate and regularly update the management strategy for road maintenance resources (grader, water car, support equipment, etc.) and the difficulty in accommodating rapid localized road deterioration due to structural failure, poor wearing course per- formance, effects of rain, spillage, and so forth. A real-time MMS was considered as a solution to overcoming the deficiencies of the existing MMS for mine haul roads: the road condition would be constantly reassessed in near real time through data generated from truck–pavement interaction. A similar approach was reported by Brown et al. (3) in which Canadian forest roads are evaluated through the use of truck vibration signatures (near real time but without fur- ther vehicle onboard data input with which to qualify the vibration records). Central to the development of a real-time condition-triggered maintenance (CTM) system is the ability to recognize road defects, in terms of type and size, from truck–pavement interaction. The aim of this study is to present a practical approach to road defect recon- struction by using the measured truck response. A modeling approach utilizing the concept of independent front suspension dynamic equi- librium is described as the basis for road defect recognition by mea- sured dynamic response data extraction and analysis. With measured suspension forces and the acceleration of the unsprung mass, the approach to tire force and road defect reconstruction is presented. The approach described can easily be generalized and applied to any vehicle type with similar suspension geometry as a basis for developing similar systems for public roads. Condition-Triggered Maintenance for Mine Haul Roads with Reconstructed-Vehicle Response to Haul Road Defects Daniel Hugo, Stephan P. Heyns, Roger J. Thompson, and Alex T. Visser The management of unpaved mine road networks—characterized by high axle loadings, low traffic volumes, variable materials and construction quality, and rapid rates of deterioration—is often inadequate. This situation results in either overmaintenance of the road or failure to rec- ognize significant deterioration, which both lead to the application of suboptimal road maintenance strategies with the attendant increase in total road-user costs. A real-time condition-triggered maintenance man- agement system was identified as a solution, in which onboard monitor- ing of vehicle dynamic response to road condition forms the basis of road defect recognition and maintenance response. A practical approach to road defect reconstruction using measured truck response is presented. Initially, the application context is introduced, then the field testing pro- gram is described, in which data sets were generated that served as the basis for mathematical modeling of the truck response. A modeling approach utilizing the concept of independent front suspension dynamic equilibrium is described as the basis for road defect recognition. On the basis of measured suspension forces and the acceleration of the unsprung mass, the approach to tire force and road defect reconstruction is presented. It is concluded that the methodology developed enables reconstruction of road defect geometries with an accuracy sufficient to allow specific types and dimensions of defects to be recognized for the purpose of road maintenance. By extending the methodology to public unpaved roads, maintenance could be applied as and where needed with a resultant reduction in authority cost and improvement in service provided for the road user. In surface mining operations, ultra-heavy trucks hauling payloads in excess of 290 t apply axle loads in excess of 1,900 kN to an unpaved mine haul road. Mine haul road networks are typically 10 to 40 km long and comprise a number of road segments, each with variable traffic volumes and construction and material qualities. These road networks have historically been maintained by relying heavily on local experience. Ever-increasing vehicle sizes have resulted in unpredictable road performance, inadequate road maintenance sched- uling, and excessive total road-user costs. Since truck haulage costs can account for up to 50% of the total operating costs incurred by a D. Hugo and S. P. Heyns, Dynamic Systems Group, Department of Mechanical and Aeronautical Engineering; R. J. Thompson, Department of Mining Engineering; and A. T. Visser, Department of Civil and Biosystems Engineering, University of Pretoria, Lynnwood Road, Hillcrest, Pretoria, South Africa, 0002. Corresponding author: R. J. Thompson, roger.thompson@up.ac.za. Transportation Research Record: Journal of the Transportation Research Board, No. 1989, Vol. 2, Transportation Research Board of the National Academies, Washington, D.C., 2007, pp. 254–260. DOI: 10.3141/1989-71