Files in this item



application/pdfLOVETT-DISSERTATION-2017.pdf (4MB)Restricted to U of Illinois
(no description provided)PDF


Title:Railroad decision support tools for track maintenance
Author(s):Lovett, Alexander Hale
Director of Research:Barkan, Christopher P. L.
Doctoral Committee Chair(s):Barkan, Christopher P. L.
Doctoral Committee Member(s):Ouyang , Yanfeng; Anand , Gopesh; Grimes, G Avery; Dick, C Tyler
Department / Program:Civil & Environmental Eng
Discipline:Civil Engineering
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Track maintenance, Indirect costs, Train delay, Maintenance aggregation
Abstract:The North American rail system requires billions of dollars annually to be maintained in proper working order. Therefore, it is critical that maintenance is performed on the right components, at the right time, and in the right location. Application of decision support tools that use objective analysis methods can result in more efficient and effective maintenance plans. This requires quantifying both direct costs associated with the performance of maintenance and the indirect costs of train delay, disruption risk, and equipment routing. To thoroughly assess these costs, an integrated approach is needed that incorporates degradation modeling, project selection, and maintenance scheduling for the entire track structure. Planning track maintenance in this way allows for the effects of changing maintenance timing to be seen explicitly through the disruption risk while considering equipment and other constraints. Managers can then combine the output from the decision support tools with their practical experience to account for location- or situation-specific characteristics that are not easily quantifiable. This dissertation presents new methods for determining the indirect costs associated with both planned and unplanned disruptions. Train delay cost models were developed that consider train operating characteristics such as terminal dwell and trainset configurations. These costs were combined with a train delay calculator adapted from the highway domain to determine the operational impact to trains during both disrupted and recovery operations. Degradation models were also developed or modified to estimate unplanned disruption risk for slow orders and acute disruptions such as rail breaks and derailments. Combined, these new methods allow for the costs of unplanned disruptions to be estimated and accounted for when planning track maintenance. A maintenance plan costing model was developed that incorporates the direct and indirect costs associated with a proposed maintenance plan. The model determines the complete cost of the plan based on capital maintenance timing, level of maintenance aggregation, and detour use. Incorporation of maintenance aggregation allows for the efficiencies of performing multiple maintenance activities simultaneously on long work windows to be explicitly considered. Alternative maintenance plans that adjust a base schedule to use maintenance aggregation can be compared to determine if the reduced direct and delay costs outweigh the additional indirect costs. Since the best way to modify a plan to reduce costs is not always obvious and can be tedious to determine manually, an optimization model was developed and solved using simulated annealing. While optimality is not guaranteed when using simulated annealing, it was shown to provide lower cost maintenance plans.
Issue Date:2017-10-13
Rights Information:Copyright 2017 Alexander H. Lovett
Date Available in IDEALS:2018-03-13
Date Deposited:2017-12

This item appears in the following Collection(s)

Item Statistics