scholarly journals Sensor-Based Predictive Maintenance with Reduction of False Alarms—A Case Study in Heavy Industry

Sensors ◽  
2021 ◽  
Vol 22 (1) ◽  
pp. 226
Author(s):  
Marek Hermansa ◽  
Michał Kozielski ◽  
Marcin Michalak ◽  
Krzysztof Szczyrba ◽  
Łukasz Wróbel ◽  
...  
Keyword(s):  
Historical Data ◽  
Cold Start ◽  
Predictive Maintenance ◽  
Use Cases ◽  
False Alarms ◽  
Data Sets ◽  
Outlier Identification ◽  
Machine Operator ◽  
System Operating

In this paper, the problem of the identification of undesirable events is discussed. Such events can be poorly represented in the historical data, and it is predominantly impossible to learn from past examples. The discussed issue is considered in the work in the context of two use cases in which vibration and temperature measurements collected by wireless sensors are analysed. These use cases include crushers at a coal-fired power plant and gantries in a steelworks converter. The awareness, resulting from the cooperation with industry, of the need for a system that works in cold start conditions and does not flood the machine operator with alarms was the motivation for proposing a new predictive maintenance method. The proposed solution is based on the methods of outlier identification. These methods are applied to the collected data that was transformed into a multidimensional feature vector. The novelty of the proposed solution stems from the creation of a methodology for the reduction of false positive alarms, which was applied to a system identifying undesirable events. This methodology is based on the adaptation of the system to the analysed data, the interaction with the dispatcher, and the use of the XAI (eXplainable Artificial Intelligence) method. The experiments performed on several data sets showed that the proposed method reduced false alarms by 90.25% on average in relation to the performance of the stand-alone outlier detection method. The obtained results allowed for the implementation of the developed method to a system operating in a real industrial facility. The conducted research may be valuable for systems with a cold start problem where frequent alarms can lead to discouragement and disregard for the system by the user.

Experimenting on the past: a case study on changing analysability in English ly-adverbs

2017 ◽  
Vol 21 (2) ◽  
pp. 317-340 ◽  
Author(s):  
HENDRIK DE SMET ◽  
FREEK VAN DE VELDE
Keyword(s):  
Negative Correlation ◽  
Historical Data ◽  
Frequency Change ◽  
Data Sets ◽  
The Past ◽  
Analogous Composition ◽  
Using Data ◽  
Complex Words ◽  
Experimental Findings

While it is undoubtedly true that historical data do not lend themselves well to the reproduction of experimental findings, the availability of increasingly extensive data sets has brought some experimenting within practical reach. This means that certain predictions based on a combination of synchronic observations and uniformitarian thinking are now testable. Synchronic evidence shows a negative correlation between analysability in morphologically complex words and various measures of frequency. It is therefore expected that when the frequency of morphologically complex items changes, their analysability will change along with this. If analysability decreases, this should in turn be reflected in decreasing sensitivity to priming by items with analogous composition. The latter prediction is in principle testable on diachronic data, offering a way of verifying the diachronic effect of frequency change on analysability. In this spirit, the present article examines the relation between changing frequency and priming sensitivity, as a proxy to analysability. This is done for a sample of 250 English ly-adverbs, such as roughly, blindly, publicly, etc. over the period 1950–2005, using data from the Hansard Corpus. Some of the expected relations between frequency and analysability can be shown to hold, albeit with great variation across lexical items. At the same time, much of the variation in our measure of analysability cannot be accounted for by frequency or frequency change alone.

scholarly journals SALES TIME SERIES ANALYTICS USING DEEP Q-LEARNING

2020 ◽  
pp. 434-441
Author(s):  
Bohdan M. Pavlyshenko
Keyword(s):  
Time Series ◽  
Historical Data ◽  
Cold Start ◽  
Business Environment ◽  
Supervised Machine Learning ◽  
Learning Approach ◽  
Environment Modeling ◽  
Q Learning ◽  
Learning Agent

The article describes the use of deep Q-learning models in the problems of sales time series analytics. In contrast to supervised machine learning, which is a kind of passive learning, where historical data are used, Q-learning is a kind of active learning aimed at maximizing a reward by optimal sequence of actions. Model free Q-learning approach to optimal pricing strategies and supply-demand problems is considered in the work. The main idea of the study is to show that using deep Q-learning approach in time series analytics causes the sequence of actions to be optimized by maximizing the reward function when the environment for learning agent interaction can be modeled using the parametric model and in the case of using the model which is based on the historical data. In the pricing optimizing case study environment was modeled using sales dependence on extras price and randomly simulated demand. In the pricing optimizing case study, the environment was modeled using sales dependence on extra price and randomly simulated demand. In the supply-demand case study, it was proposed to use historical demand time series for environment modeling, agent states were represented by promo actions, previous demand values and weekly seasonality features. Obtained results show that using deep Q-learning, we can optimize the decision making process for price optimization and supply-demand problems. Environment modeling using parametric models and historical data can be used for the cold start of learning agent. On the next steps, after the cold start, the trained agent can be used in real business environment.

scholarly journals Penerapan Metode Clustering Dengan Algoritma K-Means Tindak Kejahatan Pencurian di Kabupaten Asahan

2021 ◽  
Vol 1 (1) ◽  
pp. 7-14
Author(s):  
Nur Afni Syahpitri Damanik ◽  
Irianto Irianto ◽  
Dahriansah Dahriansah
Keyword(s):  
Data Mining ◽  
Knowledge Discovery ◽  
Historical Data ◽  
Large Data ◽  
Large Data Sets ◽  
Annual Data ◽  
Data Sets ◽  
Process Data ◽  
Different Characteristics

Abstract:Theft is the illegal taking of property or belongings of another person without the permission of the owner. The most common crime problem in Asahan District is theft, so that the POLRES is still having trouble determining which areas are often the crime of theft. With this problem, we need to do a grouping for areas where theft often occurs, so the process used  is the data mining process. Data mining is one of the processes of Knowledge Discovery from Databases (KDD). KDD is an activity that includes collecting, using historical data to find regularities, patterns or relationships in large data sets. One of the techniques known in data mining is clustering technique. The K-Means method is a method for clustering techniques, K- Means is a method that partitions data into groups so that data with the same characteristics are entered into the same set of groups and data with different characteristics are grouped into other groups. The attributes used in grouping this data are annual data, namely 2015, 2016, 2017, 2018, 2019. A case study of 9 POLSEK in the Asahan. Keywords: Data Mining, Clustering, K-Means Algorithm, Theft Crimes Grouping.  Abstrak: Pencurian merupakan pengambilan properti atau barang milik orang lain secara tidak sah tanpa ijin dari pemilik. Masalah tindak kejahatan yang paling banyak terjadi di Kabupaten Asahan adalah tindak kejahatan pencurian sehingga pihak POLRES masih kesulitan untuk menentukan daerah mana saja yang sering terjadi tindak kejahatan pencuriaan. Dengan adanya masalah ini kita perlu melakukan pengelompokan untuk daerah mana saja yang sering terjadi tindak pencurian maka proses yang digunakan adalah proses data mining. Data mining adalah salah satu proses dari Knowledge Discovery from Databases (KDD). KDD adalah kegiatan yang meliputi pengumpulan, pemakaian data, historis untuk menemukan keteraturan, pola atau hubungan dalam set data besar. Salah satu teknik yang di kenal dalam data mining adalah teknik clustering. Metode K-Means merupakan metode untuk teknik clustering, K-Means adalah metode yang mempartisi data kedalam kelompok sehingga data berkarakteristik sama dimasukan kedalam set kelompok yang sama dan data yang berkerakteristik berbeda dikelompokkan ke dalam kelompok yang lain. Atribut yang di gunakan dalam pengelomokan data ini adalah data pertahun yaitu tahun 2015, 2016, 2017, 2018, 2019. Studi kasus pada 9 POLSEK yang ada di daerah kabupaten Asahan. Kata kunci: Data Mining, Clustering, Algoritma K-Means, Pengelompokan Tindak Kejahatan  Pencurian.

Teaching Computer Based Historical Data Analysis

1994 ◽  
Vol 6 (1) ◽  
pp. 52-58 ◽  
Author(s):  
Charles Anderson ◽  
Robert J. Morris
Keyword(s):  
Data Analysis ◽  
Social History ◽  
Historical Data ◽  
Good Practice ◽  
Computer Assisted ◽  
Historical Data Analysis ◽  
University Of Edinburgh ◽  
Computer Based ◽  
The University

A case study ofa third year course in the Department of Economic and Social History in the University of Edinburgh isusedto considerandhighlightaspects of good practice in the teaching of computer-assisted historical data analysis.

A Case Study on the Un-Used Capacity Assessment Using Time Driven Activity Based Costing for Magnetic Components

2020 ◽  
Vol 6 (1) ◽  
pp. 18-39
Author(s):  
Areena Zaini ◽  
Haryantie Kamil ◽  
Mohd Yazid Abu
Keyword(s):  
Historical Data ◽  
Manufacturing Sector ◽  
Type I ◽  
Activity Based Costing ◽  
Cost Rate ◽  
Continuous Innovation ◽  
Magnetic Components ◽  
Alignment Process ◽  
Capacity Cost

The Electrical & Electronic (E&E) company is one of Malaysia’s leading industries that has 24.5% in manufacturing sector production. With a continuous innovation of E&E company, the current costing being used is hardly to access the complete activities with variations required for each workstation to measure the un-used capacity in term of resources and cost. The objective of this work is to develop a new costing structure using time-driven activity-based costing (TDABC) at . This data collection was obtained at E&E company located at Kuantan, Pahang that focusing on magnetic component. The historical data was considered in 2018. TDABC is used to measure the un-used capacity by constructing the time equation and capacity cost rate. This work found three conditions of un-used capacity. Type I is pessimistic situation whereby according to winding toroid core, the un-used capacity of time and cost are -14820 hours and -MYR2.60 respectively. It means the system must sacrifice the time and cost more than actual apportionment. Type II is most likely situation whereby according to assembly process, the un-used capacity of time and cost are 7400 hours and MYR201575.45 respectively. It means the system minimize the time and cost which close to fully utilize from the actual apportionment. Type III is optimistic situation whereby according to alignment process, the un-used capacity of time and cost are 4120 hours and MYR289217.15 respectively. It means the system used small amount of cost and time from the actual apportionment.

scholarly journals Segment-Based Approach for Assessing Hazard Risk of Coastal Highways in Hawai‘i

2019 ◽  
Vol 2673 (1) ◽  
pp. 83-91 ◽  
Author(s):  
Harrison Togia ◽  
Oceana P. Francis ◽  
Karl Kim ◽  
Guohui Zhang
Keyword(s):  
Qualitative Method ◽  
Data Sets ◽  
Regional Environment ◽  
The Road ◽  
Hazard Exposure ◽  
Multiple Indicators ◽  
System Data ◽  
Rural Highway ◽  
Geographic Information System Data

Hazards to roadways and travelers can be drastically different because hazards are largely dependent on the regional environment and climate. This paper describes the development of a qualitative method for assessing infrastructure importance and hazard exposure for rural highway segments in Hawai‘i under different conditions. Multiple indicators of roadway importance are considered, including traffic volume, population served, accessibility, connectivity, reliability, land use, and roadway connection to critical infrastructures, such as hospitals and police stations. The method of evaluating roadway hazards and importance can be tailored to fit different regional hazard scenarios. It assimilates data from diverse sources to estimate risks of disruption. A case study for Highway HI83 in Hawai‘i, which is exposed to multiple hazards, is conducted. Weakening of the road by coastal erosion, inundation from sea level rise, and rockfall hazards require adaptation solutions. By analyzing the risk of disruption to highway segments, adaptation approaches can be prioritized. Using readily available geographic information system data sets for the exposure and impacts of potential hazards, this method could be adapted not only for emergency management but also for planning, design, and engineering of resilient highways.

Language change across a lifetime: A historical micro-perspective

2021 ◽  
Vol 7 (s2) ◽  
Author(s):  
Alexander Bergs
Keyword(s):  
Communities Of Practice ◽  
Social Factors ◽  
Language Use ◽  
Language Change ◽  
Historical Data ◽  
Data Sets ◽  
The Social ◽  
Challenges And Opportunities ◽  
Linguistic Behavior ◽  
Micro Analysis

Abstract This paper focuses on the micro-analysis of historical data, which allows us to investigate language use across the lifetime of individual speakers. Certain concepts, such as social network analysis or communities of practice, put individual speakers and their social embeddedness and dynamicity at the center of attention. This means that intra-speaker variation can be described and analyzed in quite some detail in certain historical data sets. The paper presents some exemplary empirical analyses of the diachronic linguistic behavior of individual speakers/writers in fifteenth to seventeenth century England. It discusses the social factors that influence this behavior, with an emphasis on the methodological and theoretical challenges and opportunities when investigating intra-speaker variation and change.

scholarly journals The Role of Industry 4.0 and BPMN in the Arise of Condition-Based and Predictive Maintenance: A Case Study in the Automotive Industry

2021 ◽  
Vol 11 (8) ◽  
pp. 3438
Author(s):  
Jorge Fernandes ◽  
João Reis ◽  
Nuno Melão ◽  
Leonor Teixeira ◽  
Marlene Amorim
Keyword(s):  
Business Process ◽  
Automotive Industry ◽  
Industry 4.0 ◽  
Process Model ◽  
Business Process Management ◽  
Process Management ◽  
Predictive Maintenance ◽  
The Evolution Of Industry

This article addresses the evolution of Industry 4.0 (I4.0) in the automotive industry, exploring its contribution to a shift in the maintenance paradigm. To this end, we firstly present the concepts of predictive maintenance (PdM), condition-based maintenance (CBM), and their applications to increase awareness of why and how these concepts are revolutionizing the automotive industry. Then, we introduce the business process management (BPM) and business process model and notation (BPMN) methodologies, as well as their relationship with maintenance. Finally, we present the case study of the Renault Cacia, which is developing and implementing the concepts mentioned above.

scholarly journals Predictive Maintenance in Building Facilities: A Machine Learning-Based Approach

Sensors ◽  
2021 ◽  
Vol 21 (4) ◽  
pp. 1044
Author(s):  
Yassine Bouabdallaoui ◽  
Zoubeir Lafhaj ◽  
Pascal Yim ◽  
Laure Ducoulombier ◽  
Belkacem Bennadji
Keyword(s):  
Machine Learning ◽  
Model Development ◽  
Predictive Maintenance ◽  
Data Availability ◽  
Machine Learning Techniques ◽  
Automation System ◽  
Building Maintenance ◽  
Collection Data ◽  
Iot Devices

The operation and maintenance of buildings has seen several advances in recent years. Multiple information and communication technology (ICT) solutions have been introduced to better manage building maintenance. However, maintenance practices in buildings remain less efficient and lead to significant energy waste. In this paper, a predictive maintenance framework based on machine learning techniques is proposed. This framework aims to provide guidelines to implement predictive maintenance for building installations. The framework is organised into five steps: data collection, data processing, model development, fault notification and model improvement. A sport facility was selected as a case study in this work to demonstrate the framework. Data were collected from different heating ventilation and air conditioning (HVAC) installations using Internet of Things (IoT) devices and a building automation system (BAS). Then, a deep learning model was used to predict failures. The case study showed the potential of this framework to predict failures. However, multiple obstacles and barriers were observed related to data availability and feedback collection. The overall results of this paper can help to provide guidelines for scientists and practitioners to implement predictive maintenance approaches in buildings.

Sign in / Sign up

Export Citation Format

Share Document