Evaluating Deterioration Levels of Utility Poles Using Participatory Sensing and Fuzzy Inference

A public infrastructure management system should be able to identify the physical condition of infrastructure assets continuously for maintenance and repairs at the right time. However, it is a daunting task to timely monitor numerous existing infrastructure assets with limited monitoring resources. As an alternative data collection method, participatory sensing has increasingly gained attention to collect large amounts of data over large areas with little incremental costs based on citizen's mobile devices. However, it is hard for citizens to assess the condition of infrastructure assets in detail due to the lack of expertise and subjectivity in manual assessment. To address these issues, this study investigates an infrastructure condition assessment method using fuzzy inference to facilitate the evaluation for infrastructure assets by citizens. Fuzzy inference, founded on fuzzy sets and logic, is adopted to emulate the expert reasoning process of evaluating the condition of infrastructure assets. To validate the proposed method, case studies were conducted to evaluate utility poles' condition in urban areas. The fuzzy inference system was able to reduce the condition evaluation error by 20.57%.