Effective risk assessment and prioritization are crucial for maintaining the integrity of pipeline systems in the oil and gas industry. Pipeline system failure can result in severe consequences. The risk-based inspection (RBI) is a commonly used method for assessing risks with API 581 and API 580 as the standards. Qualitative RBI relies on expert judgment to estimate the consequence of failure (CoF) and the probability of failure (PoF), and its categorical risk classification introduces subjectivity and uncertainty. Many studies have explored fuzzy logic, including classical and intuitionistic fuzzy sets, to reduce uncertainty in expert judgments. In this work, we propose a linguistic-based approach that integrates neutrosophic sets with the risk-based inspection method within the API framework. Expert judgments are gathered in linguistic terms and converted into triangular neutrosophic numbers. These values are transformed into crisp numbers using the de-neutrosophication concept which are then used to calculate risk levels, rank priorities, and present the results in a classified risk matrix. The proposed approach is compared with the qualitative RBI method to examine the ranking consistency, differences, and potential advantages in handling uncertainty. This method offers a new alternative framework for evaluating pipeline thinning risks and refining inspection prioritization.