Technology-Based Condition Assessment Framework for Railway Infrastructure

铁路基础设施的状况，如轨道，镇流器和枕木，应该始终监测和分析，以确保乘客和货物的乘坐安全和乘坐质量。由于铁路基础设施的组成部分种类繁多，因此很难对其状况进行评估和监控。现有的大多数状态评估模型仅限于铁路基础设施的特定组成部分和特定的检测技术/技术。因此，应该开发一种全面的状态评估工具，使用不同的检查技术来覆盖铁路基础设施组件，以确保公共服务的安全和质量。本研究旨在开发一种基于技术的铁路基础设施状况评估框架，以考虑不同部件的结构和几何缺陷。通过文献综述和专家意见，找出并研究各组成部分的缺陷。两个主要输入将用于开发模型：(1)组件，缺陷及其类别的重要性的相对权重和(2)缺陷严重性。为了获得相对权重，将采用分析网络过程（ANP）技术，该技术考虑了组件之间的相互依赖关系及其缺陷。模糊逻辑将用于统一不同的缺陷/标准，并将语言状况评估等级量表转化为严重程度的数值分数。利用加权和均值对权重和严重程度进行综合，确定铁路基础设施的状况，并对铁路基础设施的整体状况进行评价。本研究所需的数据将从ExpGlobal Inc.、铁路状况分类手册、文献和向加拿大各地的专业人员分发的调查问卷中收集。这种融合的成果也将在一个使用EXCEL的用户友好的自动化工具中呈现。将选择Exp Global Inc.的几个案例研究来证明所提出的模型/系统的概念并验证其结果。所提出的模型有助于资产管理公司最大限度地减少铁路状况评估的不准确性，并优先考虑维护、修复和预算分配。

The condition of railway infrastructure, such as rails, ballasts and sleepers, should always be monitored andanalyzed to ensure ride safety and ride quality for both passengers and freight. Railway infrastructure is hard toassess and monitor in terms of its condition due to various types of infrastructure components. The majority ofthe existing condition assessment models are limited to a specific component of railway infrastructure andparticular inspection technique / technology. Therefore, a comprehensive condition assessment tool should bedeveloped to cover the railway infrastructure components using different inspection techniques to ensure thesafety and quality of public service. This research aims at developing a technology-based condition assessmentframework for railway infrastructure considering structural and geometrical defects in different components.The defects of each component will be identified and studied through literature review and experts opinion inthe field. Two main inputs will be used to develop the model: (1) relative weights of importance ofcomponents, defects and their categories and (2) defects severities. To obtain the relative weights, the AnalyticNetwork Process (ANP) technique will be adopted, which considers the inter-dependencies among componentsand their defects. Fuzzy logic will be used to unify different defects / criteria and to translate the linguisticcondition assessment grading scale to a numerical score for severities. Furthermore, the weighted sum meanwill be used to integrate both the weights and severities to determine the condition and to evaluate the overallcondition of the railway infrastructure. The required data for the present research will be collected from ExpGlobal Inc., railway condition classification manuals, literature and questionnaires distributed to professionalsacross Canada. The fruit of this fusion will also be presented in a user-friendly automated tool using EXCEL.Several case studies from Exp Global Inc. will be selected to prove the concept of the proposed model/systemand validate its results. The proposed model helps asset managers in minimizing the inaccuracy of railwaycondition assessment and prioritizing maintenance, rehabilitation and budget allocation.