This study was designed to develop a risk assessment model for controlling cost overruns in road maintenance projects, using roads in Songea District as a case study. It addressed the persistent problem of financial inefficiencies and unexpected cost escalations that frequently undermine the success of infrastructure maintenance in Tanzania. The study sought to systematically identify the key risk factors contributing to cost overruns and establish a model to predict and mitigate these risks in future projects. The research was motivated by the critical role of road infrastructure in economic development and the increasing challenges faced by agencies like TARURA and TANROADS in managing project budgets amidst growing demand and environmental volatility. Employing a quantitative research design, the study used structured questionnaires to collect data on six major risk categories: financial risks, environmental and climatic risks, project management risks, stakeholder risks, maintenance method risks, and resource-related risks. Multiple linear regression analysis was used to quantify their relationship with cost overrun, and model validation was conducted using five years of historical data from TARURA. The findings revealed that financial and environmental risks exert the greatest influence on cost overruns, followed by project management and resource-related risks, while maintenance methods and stakeholder concerns had minimal influence. The developed model demonstrated consistent performance, showing that approximately 22.71% of cost overruns could be mitigated through its application. The study concludes that integrating risk assessment models into road maintenance planning can significantly enhance budget control, promote infrastructure sustainability, and improve strategic decision-making.
Building similarity graph...
Analyzing shared references across papers
Loading...
Godfrey Cliford Mngale
Joseph Mkilania
Jubily Musagasa
International Journal of Advanced Research
Dar es Salaam Institute of Technology
Building similarity graph...
Analyzing shared references across papers
Loading...
Mngale et al. (Wed,) studied this question.
www.synapsesocial.com/papers/68f199d1de32064e504dd605 — DOI: https://doi.org/10.37284/ijar.8.2.3847