Machine vision for the automatic classification of images acquired from Non-destructive tests

This project is based on Artificial Intelligence (A.I) and Digital Image processing (I.P) for automatic condition monitoring of sleepers in the railway track. Rail inspection is a very important task in railway maintenance for traffic safety issues and in preventing dangerous situations. Monitoring railway track infrastructure is an important aspect in which the periodical inspection of rail rolling plane is required.Up to the present days the inspection of the railroad is operated manually by trained personnel. A human operator walks along the railway track searching for sleeper anomalies. This monitoring way is not more acceptable for its slowness and subjectivity. Hence, it is desired to automate such intuitive human skills for the development of more robust and reliable testing methods. Images of wooden sleepers have been used as data for my project. The aim of this project is to present a vision based technique for inspecting railway sleepers (wooden planks under the railway track) by automatic interpretation of Non Destructive Test (NDT) data using A.I. techniques in determining the results of inspection.

MACHINE VISION FOR AUTOMATICVISUAL INSPECTION OF WOODENRAILWAY SLEEPERS USING UNSUPERVISED NEURAL NETWORKS

The motivation for this thesis work is the need for improving reliability of equipment and quality of service to railway passengers as well as a requirement for cost-effective and efficient condition maintenance management for rail transportation. This thesis work develops a fusion of various machine vision analysis methods to achieve high performance in automation of wooden rail track inspection.The condition monitoring in rail transport is done manually by a human operator where people rely on inference systems and assumptions to develop conclusions. The use of conditional monitoring allows maintenance to be scheduled, or other actions to be taken to avoid the consequences of failure, before the failure occurs. Manual or automated condition monitoring of materials in fields of public transportation like railway, aerial navigation, traffic safety, etc, where safety is of prior importance needs non-destructive testing (NDT).In general, wooden railway sleeper inspection is done manually by a human operator, by moving along the rail sleeper and gathering information by visual and sound analysis for examining the presence of cracks. Human inspectors working on lines visually inspect wooden rails to judge the quality of rail sleeper. In this project work the machine vision system is developed based on the manual visual analysis system, which uses digital cameras and image processing software to perform similar manual inspections. As the manual inspection requires much effort and is expected to be error prone sometimes and also appears difficult to discriminate even for a human operator by the frequent changes in inspected material. The machine vision system developed classifies the condition of material by examining individual pixels of images, processing them and attempting to develop conclusions with the assistance of knowledge bases and features.A pattern recognition approach is developed based on the methodological knowledge from manual procedure. The pattern recognition approach for this thesis work was developed and achieved by a non destructive testing method to identify the flaws in manually done condition monitoring of sleepers.In this method, a test vehicle is designed to capture sleeper images similar to visual inspection by human operator and the raw data for pattern recognition approach is provided from the captured images of the wooden sleepers. The data from the NDT method were further processed and appropriate features were extracted.The collection of data by the NDT method is to achieve high accuracy in reliable classification results. A key idea is to use the non supervised classifier based on the features extracted from the method to discriminate the condition of wooden sleepers in to either good or bad. Self organising map is used as classifier for the wooden sleeper classification.In order to achieve greater integration, the data collected by the machine vision system was made to interface with one another by a strategy called fusion. Data fusion was looked in at two different levels namely sensor-level fusion, feature- level fusion. As the goal was to reduce the accuracy of the human error on the rail sleeper classification as good or bad the results obtained by the feature-level fusion compared to that of the results of actual classification were satisfactory.

Influência da soldagem por caldeamento na tenacidade à fratura e na resistência à fadiga de trilhos ferroviários

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Criterios de calidad del servicio prestado en el ámbito de la línea de atraque de las terminales portuarias de contenedores

Esta Tesis Doctoral evalúa empíricamente la calidad del servicio en la línea de atraque de las terminales portuarias de contenedores. La metodología propuesta utiliza indicadores de calidad y se basa en el concepto de la segmentación de servicio. Para ello se desarrolla la teoría del Control Estadístico de Procesos (CEP) y se utilizan los gráficos de control para clasificar el servicio en las terminales de contenedores. El propósito de la Tesis es proporcionar una metodología basada en el Control Estadístico de Procesos (CEP) para evaluar la calidad del servicio y detectar las escalas de un servicio regular de línea que se encuentran fuera de control. Por tanto, la metodología se puede utilizar para detectar eventos que son indicativos de cambio real del proceso en la línea de atraque. Esta detección puede ser compleja pues las características de los procesos en terminales de contenedores son variables. La metodología proporciona mediante los gráficos de control criterios estadísticamente objetivos de cambio. Cuando el cambio es detectado y considerado bueno sus causas deben ser identificadas y posiblemente convertirse en la nueva forma de trabajar, cuando el cambio es negativo, entonces sus causas deberían ser identificadas y eliminadas. La Tesis Doctoral está organizado de la siguiente manera: La primera parte es la introducción, e incluye los capítulos 1 al 4, la segunda parte presenta el Estado del Arte (capítulo 5) y algunos de los estudios que han inspirado esta investigación, la tercera parte se centra en la metodología utilizada (capítulo 6) y su aplicación sobre un caso de estudio (capítulo 7). Finalmente, en la cuarta parte se presentan las conclusiones y se proponen algunas de las nuevas líneas de investigación que quedan abiertas (capítulo 8). This Thesis empirically evaluates the quality of service in the berthing face of container terminals. The methodology proposed is focused on quality of service indicators and is based on the concept of service segmentation. The Statistical Process Control (SPC) theory and the control charts are used to classify container terminal service. The aim of this Thesis is to provide a methodology based on Statistical Process Control that can evaluate the quality of service and also can detect scales of shipping liner service that are out of control. Therefore, the methodology can be used to detect events that are indicative of real change in the berthing process of container vessels. The methodology proposed allows simple detection of events that are indicative of actual process change in container terminals. This detection is complex because the characteristics of the processes are variable in the container terminals; the control chart provides statistically objective criteria of change. When change is detected and considered good its cause should be identified and possibly become the new way of working, where the change is bad then its cause should be identified and eliminated. This Thesis is organized as follows: The first part is the introduction (includes Chapters 1 to 4), the second part presents the State of the Art (Chapter 5) and some of the studies that have inspired this research. The third part focuses on the methodology used (Chapter 6) and its application to a case study (Chapter 7). Finally, Part 4 presents the conclusions and suggests future research (Chapter 8).

Análise crítica e métodos para avaliação da competição portuária entre terminais de contêineres.

Uma compreensão aprofundada da dinâmica de competição portuária é particularmente importante dado o contexto atual do setor, que orienta à outorga de novos portos e terminais no Brasil, à luz da Nova Lei dos Portos, Lei Nº 12.815 de 2013. A avaliação dos reais impactos decorrentes do aumento de capacidade portuária em cada região será atividade fundamental para que, por um lado, o poder público oriente a alocação efetiva de recursos, sem prejudicar a operação dos complexos existentes; e para que a iniciativa privada, por sua vez, possa compreender os impactos dos possíveis novos empreendimentos sobre as suas operações e delinear estratégias comerciais compatíveis com o novo cenário competitivo. A partir de extensa revisão bibliográfica e da aplicação de técnicas a casos específicos, o presente trabalho detalha a dinâmica competitiva entre terminais de contêineres e avalia criticamente seis métodos utilizados para identificar a existência de competição: correlação de market share, comparação de taxas de ocupação, sobreposição de escalas marítimas, comparação de custos logísticos terrestres, representatividade da região de influência contestável e existência de poder de mercado sobre a região de influência. Dos seis métodos analisados, dois apresentam conclusões fulminantes para a questão, embora sua aplicação demande grande volume de informações; um é assertivo em condições normais de distribuição geográfica de cargas; dois apresentam condições necessárias, porém não suficientes para a identificação de competição; e um deve ser aplicado com ressalvas, uma vez que pode levar a conclusões equivocadas.

Mathematical Optimization for the Train Timetabling Problem

AMS Subj. Classification: 90C57; 90C10;

EVALUATION OF POTENTIAL TRANSLOAD FACILITY LOCATIONS IN THE UPPER PENINSULA (UP) OF MICHIGAN

Shippers want to improve their transportation efficiency and rail transportation has the potential to provide an economical alternative to trucking, but it also has potential drawbacks. The pressure to optimize transportation supply chain logistics has resulted in growing interest in multimodal alternatives, such as a combination of truck and rail transportation, but the comparison of multimodal and modal alternatives can be complicated. Shippers in Michigan’s Upper Peninsula (UP) face similar challenges. Adding to the challenge is the distance from major markets and the absence of available facilities for transloading activities. This study reviewed three potential locations for a transload facility (Nestoria, Ishpeming, and Amasa) where truck shipments could be transferred to rail and vice versa. These locations were evaluated on the basis of transportation costs for shippers when compared to the use of single mode transportation by truck to Wisconsin, Chicago, Minneapolis, and Sault Ste. Marie. In addition to shipping costs, the study also evaluated the potential impact of future carbon emission penalties on the shipping cost and the effects of changing fuel prices on shipping cost. The study used data obtained from TRANSEARCH database (2009) and found that although there were slight differences between percent savings for the three locations, any of them could provide potential benefits for movements to Chicago and Minneapolis, as long as final destination could be accessed by rail for delivery. Short haul movements of less than 200 miles (Wisconsin and Sault Ste. Marie) were not cost effective for multimodal transport. The study also found that for every dollar increase in fuel price, cost savings from multimodal option increased by three to five percent, but the inclusion of emission costs would only add one to two percent additional savings. Under a specific case study that addressed shipments by Northern Hardwoods, the most distant locations in Wisconsin would also provide cost savings, partially due to the possibility of using Michigan trucks with higher carrying capacity for the initial movement from the facility to transload location. In addition, Minneapolis movements were found to provide savings for Northern Hardwoods, even without final rail access.

Bottlenecks detection of track allocation schemes at rail stations by Petri nets

Robustness of the track allocation problem is rarely addressed in literatures and the obtained track allocation schemes (TAS) embody some bottlenecks. Therefore, an approach to detect bottlenecks is needed to support local optimization. First a TAS is transformed to an executable model by Petri nets. Then disturbances analysis is performed using the model and the indicators of the total trains' departure delays are collected to detect bottlenecks when each train suffers a disturbance. Finally, the results of the tests based on a rail hub linking six lines and a TAS about thirty minutes show that the minimum buffer time is 21 seconds and there are two bottlenecks where the buffer times are 57 and 44 seconds respectively, and it indicates that the bottlenecks do not certainly locate at the area where there is minimum buffer time. The proposed approach can further support selection of multi schemes and robustness optimization.

Towards the adoption of low-cost rail level crossing warning devices in regional areas of Australia : a review of current technologies and reliability issues

This paper discusses major obstacles for the adoption of low cost level crossing warning devices (LCLCWDs) in Australia and reviews those trialed in Australia and internationally. The argument for the use of LCLCWDs is that for a given investment, more passive level crossings can be treated, therefore increasing safety benefits across the rail network. This approach, in theory, reduces risk across the network by utilizing a combination of low-cost and conventional level crossing interventions, similar to what is done in the road environment. This paper concludes that in order to determine if this approach can produce better safety outcomes than the current approach, involving the incremental upgrade of level crossings with conventional interventions, it is necessary to perform rigorous risk assessments and cost-benefit analyses of LCLCWDs. Further research is also needed to determine how best to differentiate less reliable LCCLWDs from conventional warning devices through the use of different warning signs and signals. This paper presents a strategy for progressing research and development of LCLCWDs and details how the Cooperative Research Centre (CRC) for Rail Innovation is fulfilling this strategy through the current and future affordable level crossing projects.

Estimating the latent demand for rail transit : a case study in Perth, Western Australia

This paper presents a study on estimating the latent demand for rail transit in Australian context. Based on travel mode-choice modelling, a two-stage analysis approach is proposed, namely market population identification and mode share estimation. A case study is conducted on Midland-Fremantle rail transit corridor in Perth, Western Australia. The required data mainly include journey-to-work trip data from Australian Bureau of Statistics Census 2006 and work-purpose mode-choice model in Perth Strategic Transport Evaluation Model. The market profile is analysed, such as catchment areas, market population, mode shares, mode specific trip distributions and average trip distances. A numerical simulation is performed to test the sensitivity of the transit ridership to the change of fuel price. A corridor-level transit demand function of fuel price is thus obtained and its characteristics of elasticity are discussed. This study explores a viable approach to developing a decision-support tool for the assessment of short-term impacts of policy and operational adjustments on corridor-level demand for rail transit.

Modeling train delays in urban networks

The reliability of urban passenger trains is a critical performance measure for passenger satisfaction and ultimately market share. A delay to one train in a peak period can have a severe effect on the schedule adherence of other trains. This paper presents an analytically based model to quantify the expected positive delay for individual passenger trains and track links in an urban rail network. The model specifically addresses direct delay to trains, knock-on delays to other trains, and delays at scheduled connections. A solution to the resultant system of equations is found using an iterative refinement algorithm. Model validation, which is carried out using a real-life suburban train network consisting of 157 trains, shows the model estimates to be on average within 8% of those obtained from a large scale simulation. Also discussed, is the application of the model to assess the consequences of increased scheduled slack time as well as investment strategies designed to reduce delay.

Traffic safety at road-rail level crossings using a driving simulator and traffic simulation

A number of Intelligent Transportation Systems (ITS) were used with an advanced driving simulator to assess its influence on driving behavior. Three types of ITS interventions namely, Video in-vehicle (ITS1), Audio in-vehicle (ITS2), and On-road flashing marker (ITS3) were tested. Then, the results from the driving simulator were used as inputs for a developed model using a traffic micro-simulation (Vissim 5.4) in order to assess the safety interventions. Using a driving simulator, 58 participants were required to drive through a number of active and passive crossings with and without an ITS device and in the presence or absence of an approaching train. The effect of driver behavior changing in terms of speed and compliance rate was greater at passive crossings than at active crossings. The difference in speed of drivers approaching ITS devices was very small which indicates that ITS helps drivers encounter the crossings in a safer way. Since the current traffic simulation was not able to replicate a dynamic speed change or a probability of stopping that varies based on different ITS safety devices, some modifications of the current traffic simulation were conducted. The results showed that exposure to ITS devices at active crossings did not influence the drivers’ behavior significantly according to the traffic performance indicators used, such as delay time, number of stops, speed, and stopped delay. On the other hand, the results of traffic simulation for passive crossings, where low traffic volumes and low train headway normally occur, showed that ITS devices improved overall traffic performance.

The p-hub Model with Hub-catchment Areas, Existing Hubs, and Simulation: A Case Study of Serbian Intermodal Terminals

This paper addresses the problem of optimally locating intermodal freight terminals in Serbia. To solve this problem and determine the effects of the resulting scenarios, two modeling approaches were combined. The first approach is based on multiple-assignment hub-network design, and the second is based on simulation. The multiple-assignment p-hub network location model was used to determine the optimal location of intermodal terminals. Simulation was used as a tool to estimate intermodal transport flow volumes, due to the unreliability and unavailability of specific statistical data, and as a method for quantitatively analyzing the economic, time, and environmental effects of different scenarios of intermodal terminal development. The results presented here represent a summary, with some extension, of the research realized in the IMOD-X project (Intermodal Solutions for Competitive Transport in Serbia).