Development of a GIS-Based Road Inventory and Condition Model for Enhanced Maintenance Management: A Case Study of Tarura Kilimanjaro Region, Tanzania

Abstract

Effective road infrastructure management remains a critical challenge in developing countries, where traditional paper-based inventory systems hinder optimal maintenance planning and resource allocation. This study developed and validated a Geographic Information System (GIS)-based road inventory and condition model for the Tanzania Rural and Urban Roads Agency (TARURA) in Kilimanjaro Region, addressing the gap in spatial data integration for rural and urban road networks. Using a mixed-methods research design, the study employed Relative Importance Index analysis to identify critical road inventory components from ten technical factors, followed by multiple regression modeling to develop a predictive condition assessment tool. Data were collected from 97 respondents and 98 kilometres of road sections across seven districts comprising 4,674 kilometres of road network. The RII analysis revealed six significant factors: Surface Roughness (0.654), Rutting (0.652), Cracking (0.651), Traffic Characteristics (0.648), Roadside Features (0.647), and Drainage Structures (0.604). The developed model demonstrated strong predictive capability with R=0.880 and R²=0.774, indicating that 77.4% of road condition variation can be explained by these factors. Field validation across diverse road sections confirmed model accuracy, with conditions ranging from Very Severe (25%) to Very Good (99%). The integration of statistical modeling with GIS capabilities enables TARURA to visualise deterioration patterns spatially, optimise maintenance interventions, and allocate limited resources more effectively across the 4,674-kilometre network, where currently only 29% of roads are in good condition