Project: Data-driven condition assessment of gravel roads for sustainable maintenance

Project information

Project manager
Keegan Mbiyana
Other project members
Mirka Kans, Jaime Campos and Lars Håkansson, Linnaeus University
Participating organizations
Linnaeus University
Financier
Linnaeus University
Timetable
March 2021 - Feb 2026
Subject
Maintenance and Infrastructure (Department of Mechanical Engineering, Faculty of Technology)

More about the project

Project Overview: Improving Gravel Road Maintenance through Advanced Sensor Technology and Data-Driven Methods

Introduction: Gravel roads form a significant portion of the global road network, playing a crucial role in economic development by facilitating the movement of people and goods and promoting cultural and social interactions. These roads are especially vital in connecting rural areas to urban centres, providing essential access for residents and entrepreneurs. However, maintaining gravel roads at an acceptable level of service requires substantial investment, and resources are often limited. These roads are prone to dust, potholes, corrugations, rutting, and loose gravel, leading to faster deterioration than paved roads. Therefore, efficient and effective maintenance planning is necessary to ensure safe and reliable transportation.

Project Aim: This project aims to enhance the efficiency and effectiveness of gravel road condition assessments to achieve sustainable maintenance. By leveraging advancements in sensor technology, we seek to develop innovative methods for condition assessment, explore data-driven techniques for condition classification, failure diagnosis, and prognosis, and support maintenance decision-making.

Challenges in Current Gravel Road Maintenance: Current condition assessment methods primarily rely on visual inspections and manual evaluations. These approaches are unreliable, susceptible to human error, and time-consuming. Maintenance activities are often predetermined based on historical data rather than real-time road conditions, leading to inefficient and ineffective maintenance practices. Additionally, methods for assessing paved roads cannot be directly applied to gravel roads due to differences in failure propagation, degradation mechanisms, and environmental conditions.

Proposed Data-Driven Approach: Our project proposes a holistic, data-driven approach for gravel road maintenance to address these challenges. This approach involves systematically collecting, processing, analysing, and interpreting condition data, utilising the opportunities presented by sensor technology.

1. Objective Condition Data Collection:

• Deploy advanced sensors to collect real-time data on gravel road conditions.
• Focus on relevant parameters such as surface roughness, dust levels, and the drainage condition.

2. Data Processing and Analysis:

• Develop algorithms to process and analyse the collected data.
• Use signal processing techniques, including machine learning, to classify road conditions, diagnose failures, and predict future deterioration.

3. Decision Support System:

• Create a decision support system that uses objective data to guide maintenance planning.
• Ensure decisions are based on accurate, real-time condition data rather than subjective evaluations.

Technological Integration: The successful implementation of this data-driven approach relies heavily on the quality and relevance of the collected data. High-quality data from sensor technologies such as IoT devices, GPS, and remote sensing will be crucial. The integration of these technologies will enable continuous monitoring and provide a comprehensive understanding of road conditions, leading to timely and effective maintenance interventions.

Social and Epistemic Justice: Our project also emphasises the importance of social and epistemic justice. By involving local communities in decision-making, we ensure that their knowledge and needs are considered. This inclusive approach addresses systemic inequalities and promotes sustainable development. Effective maintenance strategies improve infrastructure and enhance access to resources, economic opportunities, and social services in rural areas.

Conclusion: This project represents a significant step towards sustainable gravel road maintenance by adopting advanced sensor technologies and data-driven methods. By ensuring that maintenance decisions are based on accurate and objective data, we can enhance the efficiency and effectiveness of maintenance activities. Integrating social and epistemic justice into our approach can create more resilient and equitable development outcomes for underserved communities. This innovative approach will ultimately contribute to safer, more reliable gravel roads that support economic and social development.