Search results for: train scheduling

Authors: Turgut Ayten, Tayfun Çay, Demet Ayten

Abstract:

In this research, nationalization method for obtaining land on the destination of Ankara-Konya High Speed Train in Turkey; Land consolidation for nationalization purpose as an alternative solution on obtaining land; a survey prepared for land owners whose lands were nationalized and institution officials who carries out the nationalization and land consolidation was applied, were investigated and the need for land consolidation for nationalization purpose is tried to be put forth. Study area is located in the Konya city- Kadınhanı district-Kolukısa and Sarikaya neighbourhood in Turkey and land consolidation results of the selected field which is on the destination of the high-speed train route were obtained. The data obtained was shared with the landowners in the research area, their choice between the nationalization method and land consolidation for nationalization method was questioned. In addition, the organization and institution officials who are accepted to used primarily by the state for obtaining land that are needed for the investments of state, and institution officials who make land consolidation were investigated on the issues of the efficiency of the methods they used and if they tried different methods.

Keywords: nationalization, land consolidation, land consolidation for nationalization

Procedia PDF Downloads 301

Authors: K. Kalaiarasan, B. Pandeeswari, A. Arockia John Francis

Abstract:

Wireless, Multi-hop networks consist of one or more intermediate nodes along the path that receive and forward packets via wireless links. The backpressure algorithm provides throughput optimal routing and scheduling decisions for multi-hop networks with dynamic traffic. Xpress, a cross-layer backpressure architecture was designed to reach the capacity of wireless multi-hop networks and it provides well coordination between layers of network by turning a mesh network into a wireless switch. Transmission over the network is scheduled using a throughput-optimal backpressure algorithm. But this architecture operates much below their capacity due to out-of-order packet delivery and variable packet size. In this paper, we present Xpress-T, a throughput optimal backpressure architecture with TCP support designed to reach maximum throughput of wireless multi-hop networks. Xpress-T operates at the IP layer, and therefore any transport protocol, including TCP, can run on top of Xpress-T. The proposed design not only avoids bottlenecks but also handles out-of-order packet delivery and variable packet size, optimally load-balances traffic across them when needed, improving fairness among competing flows. Our simulation results shows that Xpress-T gives 65% more throughput than Xpress.

Keywords: backpressure scheduling and routing, TCP, congestion control, wireless multihop network

Procedia PDF Downloads 500

Authors: Subodh Kumar, Amit Varde

Abstract:

Typical on-premises high-performance computing (HPC) environments consist of a fixed number and a fixed set of computing hardware. During the design of the HPC environment, the hardware components, including but not limited to CPU, Memory, GPU, and networking, are carefully chosen from select vendors for optimal performance. High capital cost for building the environment is a prime factor influencing the design environment. A class of software called “Job Schedulers” are critical to maximizing these resources and running multiple workloads to extract the maximum value for the high capital cost. In principle, schedulers work by preventing workloads and users from monopolizing the finite hardware resources by queuing jobs in a workload. A cloud-based HPC environment does not have the limitations of fixed (type of and quantity of) hardware resources. In theory, users and workloads could spin up any number and type of hardware resource. This paper discusses the limitations of using traditional scheduling algorithms for cloud-based HPC workloads. It proposes a new set of features, called “HPC optimizers,” for maximizing the benefits of the elasticity and scalability of the cloud with the goal of cost-performance optimization of the workload.

Keywords: high performance computing, HPC, cloud computing, optimization, schedulers

Procedia PDF Downloads 67

Authors: Güzi̇n Ti̇rkeş, Özge Teki̇n, Kerem Kurtuluş, Y. Yekta Yurtseven, Murat Baran

Abstract:

A scanning electron microscope (SEM) is a type of electron microscope mainly used in nanoscience and nanotechnology areas. Automatic image recognition and classification are among the general areas of application concerning SEM. In line with these usages, the present paper proposes a deep learning algorithm that classifies SEM images into nine categories by means of an online application to simplify the process. The NFFA-EUROPE - 100% SEM data set, containing approximately 21,000 images, was used to train and test the algorithm at 80% and 20%, respectively. Validation was carried out using a separate data set obtained from the Middle East Technical University (METU) in Turkey. To increase the accuracy in the results, the Inception ResNet-V2 model was used in view of the Fine-Tuning approach. By using a confusion matrix, it was observed that the coated-surface category has a negative effect on the accuracy of the results since it contains other categories in the data set, thereby confusing the model when detecting category-specific patterns. For this reason, the coated-surface category was removed from the train data set, hence increasing accuracy by up to 96.5%.

Keywords: convolutional neural networks, deep learning, image classification, scanning electron microscope

Procedia PDF Downloads 90

Authors: Vijay Shankar

Abstract:

In the era of water scarcity, effective use of water via irrigation requires good methods for determining crop water needs. Implementation of irrigation scheduling programs requires an accurate estimate of water use by the crop. Moisture depletion from the root zone represents the consequent crop evapotranspiration (ET). A numerical model for simulating soil water depletion in the root zone has been developed by taking into consideration soil physical properties, crop and climatic parameters. The governing differential equation for unsaturated flow of water in the soil is solved numerically using the fully implicit finite difference technique. The water uptake by plants is simulated by using three different sink functions. The non-linear model predictions are in good agreement with field data and thus it is possible to schedule irrigations more effectively. The present paper describes irrigation scheduling based on moisture depletion from the different layers of the root zone, obtained using different sink functions for three cash, oil and forage crops: cotton, safflower and barley, respectively. The soil is considered at a moisture level equal to field capacity prior to planting. Two soil moisture regimes are then imposed for irrigated treatment, one wherein irrigation is applied whenever soil moisture content is reduced to 50% of available soil water; and other wherein irrigation is applied whenever soil moisture content is reduced to 75% of available soil water. For both the soil moisture regimes it has been found that the model incorporating a non-linear sink function which provides best agreement of computed root zone moisture depletion with field data, is most effective in scheduling irrigations. Simulation runs with this moisture uptake function result in saving 27.3 to 45.5% & 18.7 to 37.5%, 12.5 to 25% % &16.7 to 33.3% and 16.7 to 33.3% & 20 to 40% irrigation water for cotton, safflower and barley respectively, under 50 & 75% moisture depletion regimes over other moisture uptake functions considered in the study. Simulation developed can be used for an optimized irrigation planning for different crops, choosing a suitable soil moisture regime depending upon the irrigation water availability and crop requirements.

Keywords: irrigation water, evapotranspiration, root uptake models, water scarcity

Procedia PDF Downloads 308

Authors: Shanaka Gunasekara, Maleen Jayasuriya, Nalin Harischandra, Lilantha Samaranayake, Gamini Dissanayake

Abstract:

One serious facet of the worsening Human-Elephant conflict (HEC) in nations such as Sri Lanka involves elephant-train collisions. Endangered Asian elephants are maimed or killed during such accidents, which also often result in orphaned or disabled elephants, contributing to the phenomenon of lone elephants. These lone elephants are found to be more likely to attack villages and showcase aggressive behaviour, which further exacerbates the overall HEC. Furthermore, Railway Services incur significant financial losses and disruptions to services annually due to such accidents. Most elephant-train collisions occur due to a lack of adequate reaction time. This is due to the significant stopping distance requirements of trains, as the full braking force needs to be avoided to minimise the risk of derailment. Thus, poor driver visibility at sharp turns, nighttime operation, and poor weather conditions are often contributing factors to this problem. Initial investigations also indicate that most collisions occur in localised “hotspots” where elephant pathways/corridors intersect with railway tracks that border grazing land and watering holes. Taking these factors into consideration, this work proposes the leveraging of recent developments in Convolutional Neural Network (CNN) technology to detect elephants using an RGB/infrared capable camera around known hotspots along the railway track. The CNN was trained using a curated dataset of elephants collected on field visits to elephant sanctuaries and wildlife parks in Sri Lanka. With this vision-based detection system at its core, a prototype unit of an early warning system was designed and tested. This weatherised and waterproofed unit consists of a Reolink security camera which provides a wide field of view and range, an Nvidia Jetson Xavier computing unit, a rechargeable battery, and a solar panel for self-sufficient functioning. The prototype unit was designed to be a low-cost, low-power and small footprint device that can be mounted on infrastructures such as poles or trees. If an elephant is detected, an early warning message is communicated to the train driver using the GSM network. A mobile app for this purpose was also designed to ensure that the warning is clearly communicated. A centralized control station manages and communicates all information through the train station network to ensure coordination among important stakeholders. Initial results indicate that detection accuracy is sufficient under varying lighting situations, provided comprehensive training datasets that represent a wide range of challenging conditions are available. The overall hardware prototype was shown to be robust and reliable. We envision a network of such units may help contribute to reducing the problem of elephant-train collisions and has the potential to act as an important surveillance mechanism in dealing with the broader issue of human-elephant conflicts.

Keywords: computer vision, deep learning, human-elephant conflict, wildlife early warning technology

Procedia PDF Downloads 198

Authors: John Pomerat, Aviv Segev

Abstract:

Recently, deep learning has had many theoretical breakthroughs. For deep learning to be successful in the industry, however, there need to be practical algorithms capable of handling many real-world hiccups preventing the immediate application of a learning algorithm. Although AI promises to revolutionize the healthcare industry, getting access to patient data in order to train learning algorithms has not been easy. One proposed solution to this is data- sharing. In this paper, we propose an alternative protocol, based on multi-party computation, to train deep learning models while maintaining both the privacy and security of training data. We examine three methods of training neural networks in this way: Transfer learning, average ensemble learning, and series network learning. We compare these methods to the equivalent model obtained through data-sharing across two different experiments. Additionally, we address the security concerns of this protocol. While the motivating example is healthcare, our findings regarding multi-party computation of neural network training are purely theoretical and have use-cases outside the domain of healthcare.

Keywords: neural network aggregation, multi-party computation, transfer learning, average ensemble learning

Procedia PDF Downloads 136

Authors: Mikiyas Adugna

Abstract:

Effort estimation is among the activities carried out in software development processes. An accurate model of estimation leads to project success. The method of agile effort estimation is a complex task because of the dynamic nature of software development. Researchers are still conducting studies on agile effort estimation to enhance prediction accuracy. Due to these reasons, we investigated and proposed a model on LASSO and Elastic Net regression to enhance estimation accuracy. The proposed model has major components: preprocessing, train-test split, training with default parameters, and cross-validation. During the preprocessing phase, the entire dataset is normalized. After normalization, a train-test split is performed on the dataset, setting training at 80% and testing set to 20%. We chose two different phases for training the two algorithms (Elastic Net and LASSO) regression following the train-test-split. In the first phase, the two algorithms are trained using their default parameters and evaluated on the testing data. In the second phase, the grid search technique (the grid is used to search for tuning and select optimum parameters) and 5-fold cross-validation to get the final trained model. Finally, the final trained model is evaluated using the testing set. The experimental work is applied to the agile story point dataset of 21 software projects collected from six firms. The results show that both Elastic Net and LASSO regression outperformed the compared ones. Compared to the proposed algorithms, LASSO regression achieved better predictive performance and has acquired PRED (8%) and PRED (25%) results of 100.0, MMRE of 0.0491, MMER of 0.0551, MdMRE of 0.0593, MdMER of 0.063, and MSE of 0.0007. The result implies LASSO regression algorithm trained model is the most acceptable, and higher estimation performance exists in the literature.

Keywords: agile software development, effort estimation, elastic net regression, LASSO

Procedia PDF Downloads 32

Authors: Zhenyu Zhang, Hsi-Hsien Wei

Abstract:

Highway network systems play a vital role in disaster response for disaster-damaged areas. Damaged bridges in such network systems can impede disaster response by disrupting transportation of rescue teams or humanitarian supplies. Therefore, emergency inspection and repair of bridges to quickly collect damage information of bridges and recover the functionality of highway networks is of paramount importance to disaster response. A widely used measure of a network’s capability to recover from disasters is resilience. To enhance highway network resilience, plenty of studies have developed various repair scheduling methods for the prioritization of bridge-repair tasks. These methods assume that repair activities are performed after the damage to a highway network is fully understood via inspection, although inspecting all bridges in a regional highway network may take days, leading to the significant delay in repairing bridges. In reality, emergency repair activities can be commenced as soon as the damage data of some bridges that are crucial to emergency response are obtained. Given that emergency bridge inspection and repair (EBIR) activities are executed simultaneously in the response phase, the real-time interactions between these activities can occur – the blockage of highways due to repair activities can affect inspection routes which in turn have an impact on emergency repair scheduling by providing real-time information on bridge damages. However, the impact of such interactions on the optimal emergency inspection routes (EIR) and emergency repair schedules (ERS) has not been discussed in prior studies. To overcome the aforementioned deficiencies, this study develops a routing and scheduling model for EBIR while accounting for real-time inspection-repair interactions to maximize highway network resilience. A stochastic, time-dependent integer program is proposed for the complex and real-time interacting EBIR problem given multiple inspection and repair teams at locations as set post-disaster. A hybrid genetic algorithm that integrates a heuristic approach into a traditional genetic algorithm to accelerate the evolution process is developed. Computational tests are performed using data from the 2008 Wenchuan earthquake, based on a regional highway network in Sichuan, China, consisting of 168 highway bridges on 36 highways connecting 25 cities/towns. The results show that the simultaneous implementation of bridge inspection and repair activities can significantly improve the highway network resilience. Moreover, the deployment of inspection and repair teams should match each other, and the network resilience will not be improved once the unilateral increase in inspection teams or repair teams exceeds a certain level. This study contributes to both knowledge and practice. First, the developed mathematical model makes it possible for capturing the impact of real-time inspection-repair interactions on inspection routing and repair scheduling and efficiently deriving optimal EIR and ERS on a large and complex highway network. Moreover, this study contributes to the organizational dimension of highway network resilience by providing optimal strategies for highway bridge management. With the decision support tool, disaster managers are able to identify the most critical bridges for disaster management and make decisions on proper inspection and repair strategies to improve highway network resilience.

Keywords: disaster management, emergency bridge inspection and repair, highway network, resilience, uncertainty

Procedia PDF Downloads 90

Authors: Kamyar Tolouei, Ehsan Moosavi

Abstract:

In open-pit mines, the long-term production scheduling optimization problem (LTPSOP) is a complicated problem that contains constraints, large datasets, and uncertainties. Uncertainty in the output is caused by several geological, economic, or technical factors. Due to its dimensions and NP-hard nature, it is usually difficult to find an ideal solution to the LTPSOP. The optimal schedule generally restricts the ore, metal, and waste tonnages, average grades, and cash flows of each period. Past decades have witnessed important measurements of long-term production scheduling and optimal algorithms since researchers have become highly cognizant of the issue. In fact, it is not possible to consider LTPSOP as a well-solved problem. Traditional production scheduling methods in open-pit mines apply an estimated orebody model to produce optimal schedules. The smoothing result of some geostatistical estimation procedures causes most of the mine schedules and production predictions to be unrealistic and imperfect. With the expansion of simulation procedures, the risks from grade uncertainty in ore reserves can be evaluated and organized through a set of equally probable orebody realizations. In this paper, to synthesize grade uncertainty into the strategic mine schedule, a stochastic integer programming framework is presented to LTPSOP. The objective function of the model is to maximize the net present value and minimize the risk of deviation from the production targets considering grade uncertainty simultaneously while satisfying all technical constraints and operational requirements. Instead of applying one estimated orebody model as input to optimize the production schedule, a set of equally probable orebody realizations are applied to synthesize grade uncertainty in the strategic mine schedule and to produce a more profitable and risk-based production schedule. A mixture of metaheuristic procedures and mathematical methods paves the way to achieve an appropriate solution. This paper introduced a hybrid model between the augmented Lagrangian relaxation (ALR) method and the metaheuristic algorithm, the Harris Hawks optimization (HHO), to solve the LTPSOP under grade uncertainty conditions. In this study, the HHO is experienced to update Lagrange coefficients. Besides, a machine learning method called Random Forest is applied to estimate gold grade in a mineral deposit. The Monte Carlo method is used as the simulation method with 20 realizations. The results specify that the progressive versions have been considerably developed in comparison with the traditional methods. The outcomes were also compared with the ALR-genetic algorithm and ALR-sub-gradient. To indicate the applicability of the model, a case study on an open-pit gold mining operation is implemented. The framework displays the capability to minimize risk and improvement in the expected net present value and financial profitability for LTPSOP. The framework could control geological risk more effectively than the traditional procedure considering grade uncertainty in the hybrid model framework.

Keywords: grade uncertainty, metaheuristic algorithms, open-pit mine, production scheduling optimization

Procedia PDF Downloads 77

Authors: Edward Holupka, John Rossman, Tye Morancy, Joseph Aronovitz, Irving Kaplan

Abstract:

A common modality for the treatment of early stage prostate cancer is the implantation of radioactive seeds directly into the prostate. The radioactive seeds are positioned inside the prostate to achieve optimal radiation dose coverage to the prostate. These radioactive seeds are positioned inside the prostate using Transrectal ultrasound imaging. Once all of the planned seeds have been implanted, two dimensional transaxial transrectal ultrasound images separated by 2 mm are obtained through out the prostate, beginning at the base of the prostate up to and including the apex. A common deep neural network, called DetectNet was trained to automatically determine the position of the implanted radioactive seeds within the prostate under ultrasound imaging. The results of the training using 950 training ultrasound images and 90 validation ultrasound images. The commonly used metrics for successful training were used to evaluate the efficacy and accuracy of the trained deep neural network and resulted in an loss_bbox (train) = 0.00, loss_coverage (train) = 1.89e-8, loss_bbox (validation) = 11.84, loss_coverage (validation) = 9.70, mAP (validation) = 66.87%, precision (validation) = 81.07%, and a recall (validation) = 82.29%, where train and validation refers to the training image set and validation refers to the validation training set. On the hardware platform used, the training expended 12.8 seconds per epoch. The network was trained for over 10,000 epochs. In addition, the seed locations as determined by the Deep Neural Network were compared to the seed locations as determined by a commercial software based on a one to three months after implant CT. The Deep Learning approach was within \strikeout off\uuline off\uwave off2.29\uuline default\uwave default mm of the seed locations determined by the commercial software. The Deep Learning approach to the determination of radioactive seed locations is robust, accurate, and fast and well within spatial agreement with the gold standard of CT determined seed coordinates.

Keywords: prostate, deep neural network, seed implant, ultrasound

Procedia PDF Downloads 170

Authors: Jérôme de Reffye, Marc Antoni

Abstract:

Following the development of the ERTMS system, we think it is interesting to develop another software-based track circuit system which would fit secondary railway lines with an easy-to-work implementation and a low sensitivity to rail-wheel impedance variations. We called this track circuit 'Track Railway by Automatic Circuits.' To be internationally implemented, this system must not have any mechanical component and must be compatible with existing track circuit systems. For example, the system is independent from the French 'Joints Isolants Collés' that isolate track sections from one another, and it is equally independent from component used in Germany called 'Counting Axles,' in French 'compteur d’essieux.' This track circuit is fully interoperable. Such universality is obtained by replacing the train detection mechanical system with a space-time filtering of train position. The various track sections are defined by the frequency of a continuous signal. The set of frequencies related to the track sections is a set of orthogonal functions in a Hilbert Space. Thus the failure probability of track sections separation is precisely calculated on the basis of signal-to-noise ratio. SNR is a function of the level of traction current conducted by rails. This is the reason why we developed a very powerful algorithm to reject noise and jamming to obtain an SNR compatible with the precision required for the track circuit and SIL 4 level. The SIL 4 level is thus reachable by an adjustment of the set of orthogonal functions. Our major contributions to railway engineering signalling science are i) Train space localization is precisely defined by a calibration system. The operation bypasses the GSM-R radio system of the ERTMS system. Moreover, the track circuit is naturally protected against radio-type jammers. After the calibration operation, the track circuit is autonomous. ii) A mathematical topology adapted to train space localization by following the train through a linear time filtering of the received signal. Track sections are numerically defined and can be modified with a software update. The system was numerically simulated, and results were beyond our expectations. We achieved a precision of one meter. Rail-ground and rail-wheel impedance sensitivity analysis gave excellent results. Results are now complete and ready to be published. This work was initialised as a research project of the French Railways developed by the Pi-Ramses Company under SNCF contract and required five years to obtain the results. This track circuit is already at Level 3 of the ERTMS system, and it will be much cheaper to implement and to work. The traffic regulation is based on variable length track sections. As the traffic growths, the maximum speed is reduced, and the track section lengths are decreasing. It is possible if the elementary track section is correctly defined for the minimum speed and if every track section is able to emit with variable frequencies.

Keywords: track section, track circuits, space-time crossing, adaptive track section, automatic railway signalling

Procedia PDF Downloads 309

Authors: Walid Moudani

Abstract:

The Crew Reserve Assignment Problem (CRAP) considers the assignment of the crew members to a set of reserve activities covering all the scheduled flights in order to ensure a continuous plan so that operations costs are minimized while its solution must meet hard constraints resulting from the safety regulations of Civil Aviation as well as from the airlines internal agreements. The problem considered in this study is of highest interest for airlines and may have important consequences on the service quality and on the economic return of the operations. In this communication, a new mathematical formulation for the CRAP is proposed which takes into account the regulations and the internal agreements. While current solutions make use of Artificial Intelligence techniques run on main frame computers, a low cost approach is proposed to provide on-line efficient solutions to face perturbed operating conditions. The proposed solution method uses a dynamic programming approach for the duties scheduling problem and when applied to the case of a medium airline while providing efficient solutions, shows good potential acceptability by the operations staff. This optimization scheme can then be considered as the core of an on-line Decision Support System for crew reserve assignment operations management.

Keywords: airlines operations management, combinatorial optimization, dynamic programming, crew scheduling

Procedia PDF Downloads 338

Authors: Richard Madanda, Paul Isaac Musasizi, Sandy Stevens Tickodri-Togboa, Doreen Orishaba, Victor Tumwine

Abstract:

New vehicle concepts targeting specific geographical markets are designed to satisfy a unique set of road and load requirements. The KIIRA-EV SMACK (KES) hybrid vehicle is designed in Uganda for the East African market. The engine and generator added to the KES electric power train serve both as the range extender and the power assist. In this paper, the design consideration taken to achieve the proper management of the on-board power from the batteries and engine-generator based on the specific drive cycle are presented. To harness the fuel- efficiency benefits of the power train, a specific control philosophy operating the engine and generator at the most efficient speed- torque and speed-power regions is presented. By using a suitable model developed in MATLAB using Simulink and Stateflow, preliminary results show that the steady-state response of the vehicle for a particular hypothetical drive cycle mimicking the expected drive conditions in the city and highway traffic is sufficient.

Keywords: control strategy, drive cycle, hybrid vehicle, simulation

Procedia PDF Downloads 351

Authors: Thomas Roche, Erica Wilson, Elizabeth Goode

Abstract:

Universities across the globe are considering how to effect meaningful change in their higher education (HE) delivery in the face of increasingly diverse student cohorts and shifting student learning preferences. This paper reports on a descriptive case study of whole-of-institution curriculum reform at one regional Australian university, where more traditional 13-week semesters were replaced with a 6-week immersive block model drawing on active learning pedagogy. Based on a synthesis of literature in best practice HE pedagogy and principles, the case study draws on student performance data and senior management staff interviews (N = 5) to outline the key changes necessary for successful HE transformation to deliver increased student pass rates and retention. The findings from this case study indicate that an institutional transformation to an immersive block model requires both a considered change in institutional policy and process as well as the appropriate resourcing of roles, governance committees, technical solutions, and, importantly, communities of practice. Implications for practice at higher education institutions considering reforming their curriculum model are also discussed.

Keywords: student retention, immersive scheduling, block model, curriculum reform, active learning, higher education pedagogy, higher education policy

Procedia PDF Downloads 40

Authors: Lin Yunuo, Seow Xing Quan, Burra Venkata Durga Kumar

Abstract:

In this modern era, the cloud operating system is the world trend applied in various industries such as business, healthy, etc. In order to deal with the large capacity of requirements in cloud computing, research come up with multi-tenant cloud computing to maximize the benefits of server providers and clients. However, there are still issues in multi-tenant cloud computing especially regarding resource allocation. Issues such as inefficient resource utilization, large latency, lack of scalability and elasticity and poor data isolation had caused inefficient resource allocation in multi-tenant cloud computing. Without a doubt, these issues prevent multitenancy reaches its best condition. In fact, there are multiple studies conducted to determine the optimal resource allocation to solve these problems these days. This article will briefly introduce the cloud operating system, Multi-tenant cloud computing and resource allocation in cloud computing. It then discusses resource allocation in multi-tenant cloud computing and the current challenges it faces. According to the issue ‘ineffective resource utilization’, we will discuss an efficient dynamic scheduling technique for multitenancy, namely Multi-tenant Dynamic Resource Scheduling Model (MTDRSM). Moreover, there also have some recommendations to improve the shortcoming of this model in this paper’s final section.

Keywords: cloud computing, cloud operation system, multitenancy, resource allocation, utilization of cloud resources

Procedia PDF Downloads 63

Authors: Alexander Goncharenko, Andrey Denisov, Sergey Alyamkin, Evgeny Terentev

Abstract:

The neural network quantization is highly desired procedure to perform before running neural networks on mobile devices. Quantization without fine-tuning leads to accuracy drop of the model, whereas commonly used training with quantization is done on the full set of the labeled data and therefore is both time- and resource-consuming. Real life applications require simplification and acceleration of quantization procedure that will maintain accuracy of full-precision neural network, especially for modern mobile neural network architectures like Mobilenet-v1, MobileNet-v2 and MNAS. Here we present a method to significantly optimize training with quantization procedure by introducing the trained scale factors for discretization thresholds that are separate for each filter. Using the proposed technique, we quantize the modern mobile architectures of neural networks with the set of train data of only ∼ 10% of the total ImageNet 2012 sample. Such reduction of train dataset size and small number of trainable parameters allow to fine-tune the network for several hours while maintaining the high accuracy of quantized model (accuracy drop was less than 0.5%). Ready-for-use models and code are available in the GitHub repository.

Keywords: distillation, machine learning, neural networks, quantization

Procedia PDF Downloads 293

Authors: Min Kyung An

Abstract:

In Wireless Sensor Networks which consist of tiny wireless sensor nodes with limited battery power, one of the most fundamental applications is data aggregation which collects nearby environmental conditions and aggregates the data to a designated destination, called a sink node. Important issues concerning the data aggregation are time efficiency and energy consumption due to its limited energy, and therefore, the related problem, named Minimum Latency Aggregation Scheduling (MLAS), has been the focus of many researchers. Its objective is to compute the minimum latency schedule, that is, to compute a schedule with the minimum number of timeslots, such that the sink node can receive the aggregated data from all the other nodes without any collision or interference. For the problem, the two interference models, the graph model and the more realistic physical interference model known as Signal-to-Interference-Noise-Ratio (SINR), have been adopted with different power models, uniform-power and non-uniform power (with power control or without power control), and different antenna models, omni-directional antenna and directional antenna models. In this survey article, as the problem has proven to be NP-hard, we present and compare several state-of-the-art approximation algorithms in various models on the basis of latency as its performance measure.

Keywords: data aggregation, convergecast, gathering, approximation, interference, omni-directional, directional

Procedia PDF Downloads 200

Authors: Hyo-In Koh, Anders Nordborg, Alex Sievi, Chun-Kwon Park

Abstract:

The noise source for the interior room of the high speed train is constituted by the rolling contact between the wheel and the rail, aerodynamic noise and structure-borne sound generated through the vibrations of bogie, connection points to the carbody. Air-borne sound is radiated through the panels and structures into the interior room of the trains. The high-speed lines are constructed with slab track systems and many tunnels. The interior noise level and the frequency characteristics vary according to types of the track structure and the infrastructure. In this paper the main sound sources and the transfer paths are studied to find out the contribution characteristics of the sources to the interior noise of a high-speed rail vehicle. For the identification of the acoustic power of each parts of the rolling noise sources a calculation model of wheel/rail noise is developed and used. For the analysis of the transmission of the sources to the interior noise noise and vibration are measured during the operation of the vehicle. According to operation speeds, the mainly contributed sources and the paths could be analyzed. Results of the calculations on the source generation and the results of the measurement with a high-speed train are shown and discussed.

Keywords: rail vehicle, high-speed, interior noise, noise source

Procedia PDF Downloads 376

Authors: S. M. T. Mustafa, D. Raes, M. Huysmans

Abstract:

Proper utilization of water resource is very important in agro-based Bangladesh. Irrigation schedule based on local environmental conditions, soil type and water availability will allow a sustainable use of water resources in agriculture. In this study, the FAO crop water model (AquaCrop) was used to simulate the different water and fertilizer management strategies in different location of Bangladesh to obtain a management guideline for the farmer. Model was calibrated and validated for wheat (Triticum aestivum L.). The statistical indices between the observed and simulated grain yields obtained were very good with R2, RMSE, and EF values of 0.92, 0.33, and 0.83, respectively for model calibration and 0.92, 0.68 and 0.77, respectively for model validations. Stem elongation (jointing) to booting and flowering stage were identified as most water sensitive for wheat. Deficit irrigation on water sensitive stage could increase the grain yield for increasing soil fertility levels both for loamy and sandy type soils. Deficit irrigation strategies provides higher water productivity than full irrigation strategies and increase the yield stability (reduce the standard deviation). The practical deficit irrigation schedule for wheat for four different stations and two different soils were designed. Farmer can produce more crops by using deficit irrigation schedule under water stress condition. Practical application and validation of proposed strategies will make them more credible.

Keywords: crop-water model, deficit irrigation, irrigation scheduling, wheat

Procedia PDF Downloads 405

Authors: Mathias Linden, André Schneider, Harald F. O. von Korflesch

Abstract:

The risk of train accidents is an ongoing concern for railway organizations, governments, insurance companies and other depended sectors. Safety technologies are installed to reduce and to prevent potential damages of train accidents. Since the budgetary for the safety of railway organizations is limited, it is necessary not only to achieve a high availability and high safety standard but also to be cost effective. Therefore, an economic assessment of safety technologies is fundamental to create an accurate risk analysis. In order to conduct an economical assessment of a railway safety technology and a quantification of the costs of the accident causes, the Global Railway Accident Database & Evaluation (GRADE) has been developed. The aim of this paper is to describe the structure of this accident database and to show how it can be used for risk analyses. A number of risk analysis methods, such as the probabilistic safety assessment method (PSA), was used to demonstrate this accident database’s different possibilities of risk analysis. In conclusion, it can be noted that these analyses would not be as accurate without GRADE. The information gathered in the accident database was not available in this way before. Our findings are relevant for railway operators, safety technology suppliers, assurances, governments and other concerned railway organizations.

Keywords: accident causes, accident costs, accident database, global railway accident database & evaluation, GRADE, probabilistic safety assessment, PSA, railway accidents, risk analysis

Procedia PDF Downloads 334

Authors: R. N. Begum, Ambalika Sharma, G. K. Singh

Abstract:

Identity theft has serious ramifications beyond data and personal information loss. This necessitates the implementation of robust and efficient user identification systems. Therefore, automatic biometric recognition systems are the need of the hour, and ECG-based systems are unquestionably the best choice due to their appealing inherent characteristics. The CNNs are the recent state-of-the-art techniques for ECG-based user identification systems. However, the results obtained are significantly below standards, and the situation worsens as the number of users and types of heartbeats in the dataset grows. As a result, this study proposes a highly accurate and resilient ECG-based person identification system using CNN's dense learning framework. The proposed research explores explicitly the calibre of dense CNNs in the field of ECG-based human recognition. The study tests four different configurations of dense CNN which are trained on a dataset of recordings collected from eight popular ECG databases. With the highest FAR of 0.04 percent and the highest FRR of 5%, the best performing network achieved an identification accuracy of 99.94 percent. The best network is also tested with various train/test split ratios. The findings show that DenseNets are not only extremely reliable but also highly efficient. Thus, they might also be implemented in real-time ECG-based human recognition systems.

Keywords: Biometrics, Dense Networks, Identification Rate, Train/Test split ratio

Procedia PDF Downloads 139

Authors: Tommaso Adamo, Gianpaolo Ghiani, Antonio Domenico Grieco, Emanuela Guerriero

Abstract:

Batch production plants provide a wide range of scheduling problems. In pharmaceutical industries a batch process is usually described by a recipe, consisting of an ordering of tasks to produce the desired product. In this research work we focused on pharmaceutical production processes requiring the culture of a microorganism population (i.e. bacteria, yeasts or antibiotics). Several sources of uncertainty may influence the yield of the culture processes, including (i) low performance and quality of the cultured microorganism population or (ii) microbial contamination. For these reasons, robustness is a valuable property for the considered application context. In particular, a robust schedule will not collapse immediately when a cell of microorganisms has to be thrown away due to a microbial contamination. Indeed, a robust schedule should change locally in small proportions and the overall performance measure (i.e. makespan, lateness) should change a little if at all. In this research work we formulated a constraint programming optimization (COP) model for the robust planning of antibiotics production. We developed a discrete-time model with a multi-criteria objective, ordering the different criteria and performing a lexicographic optimization. A feasible solution of the proposed COP model is a schedule of a given set of tasks onto available resources. The schedule has to satisfy tasks precedence constraints, resource capacity constraints and time constraints. In particular time constraints model tasks duedates and resource availability time windows constraints. To improve the schedule robustness, we modeled the concept of (a, b) super-solutions, where (a, b) are input parameters of the COP model. An (a, b) super-solution is one in which if a variables (i.e. the completion times of a culture tasks) lose their values (i.e. cultures are contaminated), the solution can be repaired by assigning these variables values with a new values (i.e. the completion times of a backup culture tasks) and at most b other variables (i.e. delaying the completion of at most b other tasks). The efficiency and applicability of the proposed model is demonstrated by solving instances taken from Sanofi Aventis, a French pharmaceutical company. Computational results showed that the determined super-solutions are near-optimal.

Keywords: constraint programming, super-solutions, robust scheduling, batch process, pharmaceutical industries

Procedia PDF Downloads 592

Authors: Li Guo, Jianhong Wang, Dian Huang, Shengzhong Feng

Abstract:

Computing networks as a new network architecture has shown great promise in boosting the utilization of different resources, such as computing, caching, and communications. To maximise the efficiency of resource orchestration in computing network control systems (CNCSs), this work proposes a dynamic orchestration strategy of a different resource based on task requirements from computing power requestors (CPRs). Specifically, computing power providers (CPPs) in CNCSs could share information with each other through communication channels on the basis of blockchain technology, especially their current idle resources. This dynamic process is modeled as a cooperative game in which CPPs have the same target of maximising long-term rewards by improving the resource utilization ratio. Meanwhile, the task requirements from CPRs, including size, deadline, and calculation, are simultaneously considered in this paper. According to task requirements, the proposed orchestration strategy could schedule the best-fitting resource in CNCSs, achieving the maximum long-term rewards of CPPs and the best quality of experience (QoE) of CRRs at the same time. Based on the EdgeCloudSim simulation platform, the efficiency of the proposed strategy is achieved from both sides of CPRs and CPPs. Besides, experimental results show that the proposed strategy outperforms the other comparisons in all cases.

Keywords: computing network control systems, resource orchestration, dynamic scheduling, blockchain, cooperative game

Procedia PDF Downloads 79

Authors: Juzhong Tan, William Kerr

Abstract:

Roasting is one critical procedure in chocolate processing, where special favors are developed, moisture content is decreased, and better processing properties are developed. Therefore, determination of roasting degree of cocoa bean is important for chocolate manufacturers to ensure the quality of chocolate products, and it also decides the commercial value of cocoa beans collected from cocoa farmers. The roasting degree of cocoa beans currently relies on human specialists, who sometimes are biased, and chemical analysis, which take long time and are inaccessible to many manufacturers and farmers. In this study, a self-made electronic nose system consists of gas sensors (TGS 800 and 2000 series) was used to detecting the gas generated by cocoa beans with a different roasting degree (0min, 20min, 30min, and 40min) and the signals collected by gas sensors were used to train a three-layers ANN. Chemical analysis of the graded beans was operated by traditional GC-MS system and the contents of volatile chemical compounds were used to train another ANN as a reference to electronic nosed signals trained ANN. Both trained ANN were used to predict cocoa beans with a different roasting degree for validation. The best accuracy of grading achieved by electronic nose signals trained ANN (using signals from TGS 813 826 820 880 830 2620 2602 2610) turned out to be 96.7%, however, the GC trained ANN got the accuracy of 83.8%.

Keywords: artificial neutron network, cocoa bean, electronic nose, roasting

Procedia PDF Downloads 208

Authors: Chandra Prayaga, Aaron Wade, Lakshmi Prayaga, Gopi Shankar Mallu

Abstract:

This paper presents the use of machine learning algorithms to predict the success of students in an introductory physics course. Data having 140 rows pertaining to the performance of two batches of students was used. The lack of sufficient data to train robust machine learning models was compensated for by generating synthetic data similar to the real data. CTGAN and CTGAN with Gaussian Copula (Gaussian) were used to generate synthetic data, with the real data as input. To check the similarity between the real data and each synthetic dataset, pair plots were made. The synthetic data was used to train machine learning models using the PyCaret package. For the CTGAN data, the Ada Boost Classifier (ADA) was found to be the ML model with the best fit, whereas the CTGAN with Gaussian Copula yielded Logistic Regression (LR) as the best model. Both models were then tested for accuracy with the real data. ROC-AUC analysis was performed for all the ten classes of the target variable (Grades A, A-, B+, B, B-, C+, C, C-, D, F). The ADA model with CTGAN data showed a mean AUC score of 0.4377, but the LR model with the Gaussian data showed a mean AUC score of 0.6149. ROC-AUC plots were obtained for each Grade value separately. The LR model with Gaussian data showed consistently better AUC scores compared to the ADA model with CTGAN data, except in two cases of the Grade value, C- and A-.

Keywords: machine learning, student success, physics course, grades, synthetic data, CTGAN, gaussian copula CTGAN

Procedia PDF Downloads 22

Authors: Thiago Viana

Abstract:

Eventually train accidents occur on railways and for some specific cases it is necessary to use a train rescue with a crane positioned under a platform wagon. These tumbled machines are collected and sent to the machine shop or scrap yard. In one of these cranes that were being used to rescue a wagon, occurred a fall of hoist due to fracture of two large pins. The two pins were collected and sent for failure analysis. This work investigates the main cause and the secondary causes for the initiation of the fatigue crack. All standard failure analysis procedures were applied, with careful evaluation of the characteristics of the material, fractured surfaces and, mainly, metallographic tests using an optical microscope to compare the geometry of the peaks and valleys of the thread of the pins and their respective seats. By metallographic analysis, it was concluded that the fatigue cracks were started from a notch (stress concentration) in the valley of the threads of the pin applied to the right side of the crane (pin 1). In this, it was verified that the peaks of the threads of the pin seat did not have proper geometry, with sharp edges being present that caused such notches. The visual analysis showed that fracture of the pin on the left side of the crane (pin 2) was brittle type, being a consequence of the fracture of the first one. Recommendations for this and other railway cranes have been made, such as nondestructive testing, stress calculation, design review, quality control and suitability of the mechanical forming process of the seat threads and pin threads.

Keywords: crane, fracture, pin, railway

Procedia PDF Downloads 83

Authors: Taramani Yadav, Gajender Kumar, R.K. Yadav, H.S. Jat

Abstract:

Soil Salinity and irrigation water salinity is critical threat to enhance agricultural food production to full fill the demand of billion plus people worldwide. Salt affected soils covers 6.73 Mha in India and ~1000 Mha area around the world. Irrigation scheduling of saline water is the way to ensure food security in salt affected areas. Research experiment was conducted at ICAR-Central Soil Salinity Research Institute, Experimental Farm, Nain, Haryana, India with 36 treatment combinations in double split plot design. Three sets of treatments consisted of (i) three regimes of irrigation viz., 60, 80 and 100% (I1, I2 and I3, respectively) of crop ETc (crop evapotranspiration at identified respective stages) in main plot; (ii) four levels of irrigation water salinity (sub plot treatments) viz., 2, 4, 8 and 12 dS m-1 (iii) applications of two PBRs along with control (without PBRs) i.e. salicylic acid (G1; 1 mM) and thiourea (G2; 500 ppm) as sub-sub plot treatments. Grain yield of wheat (Triticum aestivum) was increased with less amount of high salt loaded irrigation water at the same level of salinity (2 dS m-1), the trend was I3>I2>I1 at 2 dS m-1 with 8.10 and 17.07% increase at 80 and 100% ETc, respectively compared to 60% ETc. But contrary results were obtained by increasing amount of irrigation water at same level of highest salinity (12 dS m-1) showing following trend; I1>I2>I3 at 12 dS m-1 with 9.35 and 12.26% increase at 80 and 60% ETc compared to 100% ETc. Enhancement in grain yield of wheat (Triticum aestivum) is not need to increase amount of irrigation water under saline condition, with salty irrigation water less amount of irrigation water gave the maximum wheat (Triticum aestivum) grain yield.

Keywords: Irrigation, Salinity, Wheat, Yield

Procedia PDF Downloads 137

Authors: Natalie M. Wride, Xueyu Geng

Abstract:

Subgrade soils of railway infrastructure are subjected to a significant number of load applications over their design life. The use of slab track on existing and future proposed rail links requires a reduced maintenance and repair regime for the embankment subgrade, due to restricted access to the subgrade soils for remediation caused by cyclic deformation. It is, therefore, important to study the deformation behaviour of soft cohesive subsoils induced as a result of long term cyclic loading. In this study, a series of oedometer tests and cyclic triaxial tests (10,000 cycles) have been undertaken to investigate the undrained deformation behaviour of soft kaolin. X-ray Computer Tomography (CT) scanning of the samples has been performed to determine the change in porosity and soil structure density from the sample microstructure as a result of the laboratory testing regime undertaken. Combined with the examination of excess pore pressures and strains obtained from the cyclic triaxial tests, the results are compared with an existing analytical solution for long term settlement considering repeated low amplitude loading. Modifications to the analytical solution are presented based on the laboratory analysis that shows good agreement with further test data.

Keywords: creep, cyclic loading, deformation, long term settlement, train loading

Procedia PDF Downloads 273

Authors: Emilia K. H. Danielsson-Waters, Malaz Elsaddig, Kevin Jones

Abstract:

Unfortunately, it is well known that neonatal deaths are a common and potentially preventable occurrence across the world. Neonatal resuscitation is a simple and inexpensive intervention that can effectively reduce this rate, and can be taught and implemented globally. This project is a follow-on from one in 2012, which found that neonatal resuscitation simulation was valuable for education, but would be better improved by being delivered by local staff. Methods: This study involved auditing the neonatal admission and death records within a rural Ugandan hospital, alongside implementing a Train-The-Trainer teaching scheme to teach Neonatal Resuscitation. One local doctor was trained for simulating neonatal resuscitation, whom subsequently taught an additional 14 staff members in one-afternoon session. Participants were asked to complete questionnaires to assess their knowledge and confidence pre- and post-simulation, and a survey to identify barriers and drivers to simulation. Results: The results found that the neonatal mortality rate in this hospital was 25% between July 2016- July 2017, with birth asphyxia, prematurity and sepsis being the most common causes. Barriers to simulation that were identified predominantly included a lack of time, facilities and opportunity, yet all members stated simulation was beneficial for improving skills and confidence. The simulation session received incredibly positive qualitative feedback, and also a 0.58-point increase in knowledge (p=0.197) and 0.73-point increase in confidence (0.079). Conclusion: This research shows that it is possible to create a teaching scheme in a rural hospital, however, many barriers are in place for its sustainability, and a larger sample size with a more sensitive scale is required to achieve statistical significance. This is undeniably important, because teaching neonatal resuscitation can have a direct impact on neonatal mortality. Subsequently, recommendations include that efforts should be put in place to create a sustainable training scheme, for example, by employing a resuscitation officer. Moreover, neonatal resuscitation teaching should be conducted more frequently in hospitals, and conducted in a wider geographical context, including within the community, in order to achieve its full effect.

Keywords: neonatal resuscitation, sustainable medical education, train-the-trainer, Uganda

Procedia PDF Downloads 122