Abstract: Early detection of anomalies in data centers is important to reduce downtimes and the costs of periodic maintenance. However, there is little research on this topic and even fewer on the fusion of sensor data for the detection of abnormal events. The goal of this paper is to propose a method for anomaly detection in data centers by combining sensor data (temperature, humidity, power) and deep learning models. The model described in the paper uses one autoencoder per sensor to reconstruct the inputs. The auto-encoders contain Long-Short Term Memory (LSTM) layers and are trained using the normal samples of the relevant sensors selected by correlation analysis. The difference signal between the input and its reconstruction is then used to classify the samples using feature extraction and a random forest classifier. The data measured by the sensors of a data center between January 2019 and May 2020 are used to train the model, while the data between June 2020 and May 2021 are used to assess it. Performances of the model are assessed a posteriori through F1-score by comparing detected anomalies with the data center’s history. The proposed model outperforms the state-of-the-art reconstruction method, which uses only one autoencoder taking multivariate sequences and detects an anomaly with a threshold on the reconstruction error, with an F1-score of 83.60% compared to 24.16%.
Abstract: Current server systems are responsible for critical applications that run in different infrastructures, such as the cloud, physical machines, and virtual machines. A common challenge that these systems face are the various hardware faults that may occur due to the high load, among other reasons, which translates to errors resulting in malfunctions or even server downtime. The most important hardware parts, that are causing most of the errors, are the CPU, RAM, and the hard drive - HDD. In this work, we investigate selected CPU, RAM, and HDD errors, observed or simulated in kernel ring buffer log files from GNU/Linux servers. Moreover, a severity characterization is given for each error type. Understanding these errors is crucial for the efficient analysis of kernel logs that are usually utilized for monitoring servers and diagnosing faults. In addition, to support the previous analysis, we present possible ways of simulating hardware errors in RAM and HDD, aiming to facilitate the testing of methods for detecting and tackling the above issues in a server running on GNU/Linux.
Abstract: Repairable items inventory system is a management tool used to incorporate all information concerning inventory levels and movements for repaired and new items. This paper presents development of an effective simulation model for managing the inventory of repairable items for a production system where production lines send their faulty items to a repair shop considering the stochastic failure behavior and repair times. The developed model imitates the process of handling the on-hand inventory of repaired items and the replenishment of the inventory of new items using Economic Order Quantity and Reorder Point ordering policy in a flexible and risk-free environment. We demonstrate the appropriateness and effectiveness of the proposed simulation model using an illustrative case problem. The developed simulation model can be used as a reliable tool for estimating a healthy on-hand inventory of new and repaired items, backordered items, and downtime due to unavailability of repaired items, and validating and examining Economic Order Quantity and Reorder Point ordering policy, which would further be compared with other ordering strategies as future work.
Abstract: Structural frequency response testing is accurate in identifying the dynamic characteristic of a machinery structure. In practical perspective, conventional structural frequency response testing such as experimental modal analysis with impulse technique (also known as “impulse testing”) has limitation especially on its long acquisition time. The high acquisition time is mainly due to the redundancy procedure where the engineer has to repeatedly perform the test in 3 directions, namely the axial-, horizontal- and vertical-axis, in order to comprehensively define the dynamic behavior of a 3D structure. This is unfavorable to numerous industries where the downtime cost is high. This study proposes to reduce the testing time by using oblique impact. Theoretically, a single oblique impact can induce significant vibration responses and vibration modes in all the 3 directions. Hence, the acquisition time with the implementation of the oblique impulse technique can be reduced by a factor of three (i.e. for a 3D dynamic system). This study initiates an experimental investigation of impulse testing with oblique excitation. A motor-driven test rig has been used for the testing purpose. Its dynamic characteristic has been identified using the impulse testing with the conventional normal impact and the proposed oblique impact respectively. The results show that the proposed oblique impulse testing is able to obtain all the desired natural frequencies in all 3 directions and thus providing a feasible solution for a fast and time effective way of conducting the impulse testing.
Abstract: Mining industry is known for its appetite to spend sizeable capital on mine equipment. However, in the current scenario, the mining industry is challenged by daunting factors of non-uniform geological conditions, uneven ore grade, uncontrollable and volatile mineral commodity prices and the ever increasing quest to optimize the capital and operational costs. Thus, the role of equipment reliability and maintenance planning inherits a significant role in augmenting the equipment availability for the operation and in turn boosting the mine productivity. This paper presents the Risk Based Maintenance (RBM) planning conducted on mine loading equipment namely Load Haul Dumpers (LHDs) at Vedanta Resources Ltd subsidiary Hindustan Zinc Limited operated Sindesar Khurd Mines, an underground zinc and lead mine situated in Dariba, Rajasthan, India. The mining equipment at the location is maintained by the Original Equipment Manufacturers (OEMs) namely Sandvik and Atlas Copco, who carry out the maintenance and inspection operations for the equipment. Based on the downtime data extracted for the equipment fleet over the period of 6 months spanning from 1st January 2017 until 30th June 2017, it was revealed that significant contribution of three downtime issues related to namely Engine, Hydraulics, and Transmission to be common among all the loading equipment fleet and substantiated by Pareto Analysis. Further scrutiny through Bubble Matrix Analysis of the given factors revealed the major influence of selective factors namely Overheating, No Load Taken (NTL) issues, Gear Changing issues and Hose Puncture and leakage issues. Utilizing the equipment wise analysis of all the downtime factors obtained, spares consumed, and the alarm logs extracted from the machines, technical design changes in the equipment and pre shift critical alarms checklist were proposed for the equipment maintenance. The given analysis is beneficial to allow OEMs or mine management to focus on the critical issues hampering the reliability of mine equipment and design necessary maintenance strategies to mitigate them.
Abstract: This paper deals with development of Computerized Maintenance Management System (CMMS) for a fertilizer plant. The software is advanced, easy to use, less complex, less expensive and also less time consuming. It consists of number of modules like detailed information of equipment, maintenance procedures, work order and employees detail. The objectives of CMMS are to reduce overall downtime, overall yearly maintenance cost and occurrence of failures of the equipment and to get day-by-day maintenance plan and strategy. In this regard, the behavioral chart for urea prilling unit at Fertilizer plant has been developed in form of Root Cause Analysis (RCA). Besides this, a maintenance program has also been proposed and used for the purpose of maintenance planning of the urea prilling unit. The outcome of software has been consulted with the concerned plant individuals and found to be extremely favorable for improving the performance level of the concerned plant.
Abstract: Background: The thread lift technique has become popular because it is less invasive, requires a shorter operation, less downtime, and results in fewer postoperative complications. The advantage of the technique is that the thread can be inserted under the skin without the need for long incisions. Currently, there are a lot of thread lift techniques with respect to the specific types of thread used on specific areas, such as the mid-face, lower face, or neck area. Objective: To review the thread lift technique for specific areas according to type of thread, patient selection, and how to match the most appropriate to the patient. Materials and Methods: A literature review technique was conducted by searching PubMed and MEDLINE, then compiled and summarized. Result: We have divided our protocols into two sections: Protocols for short suture, and protocols for long suture techniques. We also created 3D pictures for each technique to enhance understanding and application in a clinical setting. Conclusion: There are advantages and disadvantages to short suture and long suture techniques. The best outcome for each patient depends on appropriate patient selection and determining the most suitable technique for the defect and area of patient concern.
Abstract: Nowadays, to decrease the number of downtimes in the industries such as metal mining, petroleum and chemical industries, predictive maintenance is crucial. In order to have efficient predictive maintenance, knowing the performance of critical equipment of production line such as pumps and hydro-cyclones under variable operating parameters, selecting best indicators of this equipment health situations, best locations for instrumentation, and also measuring of these indicators are very important. In this paper, computer aided engineering (CAE) tools are implemented to study some important elements of copper process line, namely slurry pumps and cyclone to predict the performance of these components under different working conditions. These modeling and simulations can be used in predicting, for example, the damage tolerance of the main shaft of the slurry pump or wear rate and location of cyclone wall or pump case and impeller. Also, the simulations can suggest best-measuring parameters, measuring intervals, and their locations.
Abstract: During aircraft maintenance scheduling, operator calculates the budget of the maintenance. Usually, this calculation includes only the costs that are directly related to the maintenance process such as cost of labor, material, and equipment. In some cases, overhead cost is also included. However, in some of those, downtime cost is neglected claiming that grounding is a natural fact of maintenance; therefore, it is not considered as part of the analytical decision-making process. Based on the normalized data, we introduce downtime cost with its monetary value and add its seasonal character. We envision that the rest of the model, which works together with the downtime cost, could be checked with the real life cases, through the review of MRO cost and airline spending in the particular and scheduled maintenance events.
Abstract: In this paper static scheme of under-frequency based load shedding is considered for chemical and petrochemical industries with islanded distribution networks relying heavily on the primary commodity to ensure minimum production loss, plant downtime or critical equipment shutdown. A simplistic methodology is proposed for in-house implementation of this scheme using underfrequency relays and a step by step guide is provided including the techniques to calculate maximum percentage overloads, frequency decay rates, time based frequency response and frequency based time response of the system. Case study of FFL electrical system is utilized, presenting the actual system parameters and employed load shedding settings following the similar series of steps. The arbitrary settings are then verified for worst overload conditions (loss of a generation source in this case) and comprehensive system response is then investigated.
Abstract: Operations, maintenance and reliability of wind
turbines have received much attention over the years due to the rapid
expansion of wind farms. This paper explores early fault diagnosis
technique for a 5MW wind turbine system subjected to multiple
faults, where genetic optimization algorithm is employed to make the
residual sensitive to the faults, but robust against disturbances. The
proposed technique has a potential to reduce the downtime mostly
caused by the breakdown of components and exploit the productivity
consistency by providing timely fault alarms. Simulation results show
the effectiveness of the robust fault detection methods used under
Matlab/Simulink/Gatool environment.
Abstract: ESPRIT-TLS method appears a good choice for high
resolution fault detection in induction machines. It has a very high
effectiveness in the frequency and amplitude identification.
Contrariwise, it presents a high computation complexity which
affects its implementation in real time fault diagnosis. To avoid this
problem, a Fast-ESPRIT algorithm that combined the IIR band-pass
filtering technique, the decimation technique and the original
ESPRIT-TLS method was employed to enhance extracting accurately
frequencies and their magnitudes from the wind stator current with
less computation cost. The proposed algorithm has been applied to
verify the wind turbine machine need in the implementation of an online,
fast, and proactive condition monitoring. This type of remote
and periodic maintenance provides an acceptable machine lifetime,
minimize its downtimes and maximize its productivity. The
developed technique has evaluated by computer simulations under
many fault scenarios. Study results prove the performance of Fast-
ESPRIT offering rapid and high resolution harmonics recognizing
with minimum computation time and less memory cost.
Abstract: This paper proposes an effective maintenance method by considering the implementation of the Computerized Maintenance Management System (CMMS) software to plan a maintenance activity in a manufacturing industry. Globally, maintenance is a very important activity in the manufacturing sector to prolong the life span of equipment and machinery; it is also applicable to all household items. It is obvious and well known that apart from giving long life to equipment, it reduces the substantial financial losses for repairs and save the production downtime. In some cases, appropriate maintenance of plant equipment and machinery reduces the tendencies of injuries to personnel in the job floor. But before the maintenance process can be carried out, proper and effective work order planning and scheduling must be in place in other to achieve the set goals and objectives of a maintenance shop. Brief reviews of common planning tools which include the Computerized Maintenance Management System (CMMS) are presented. An interesting outline of analyses on planning and scheduling for effective job planning in a typical manufacturing industry using the CMMS is also presented and discussed. Finally, the steps to adhere to in making job planning effective in a manufacturing industry are also highlighted.
Abstract: Applying a rigorous process to optimize the elements
of a supply-chain network resulted in reduction of the waiting time
for a service provider and customer. Different sources of downtime
of hydraulic pressure controller/calibrator (HPC) were causing
interruptions in the operations. The process examined all the issues to
drive greater efficiencies. The issues included inherent design issues
with HPC pump, contamination of the HPC with impurities, and the
lead time required for annual calibration in the USA.
HPC is used for mandatory testing/verification of formation
tester/pressure measurement/logging-while drilling tools by oilfield
service providers, including Halliburton.
After market study andanalysis, it was concluded that the current
HPC model is best suited in the oilfield industry. To use theexisting
HPC model effectively, design andcontamination issues were
addressed through design and process improvements. An optimum
network is proposed after comparing different supply-chain models
for calibration lead-time reduction.
Abstract: Rolling element bearings are widely used in industry,
especially where high load capacity is required. The diagnosis of
their conditions is essential matter for downtime reduction and saving
cost of maintenance. Therefore, an intensive analysis of frequency
spectrum of their faults must be carried out in order to determine the
main reason of the fault. This paper focus on a beating phenomena
observed in the waveform (time domain) of a cylindrical rolling
element bearing. The beating frequencies were not related to any
sources nearby the machine nor any other malfunctions (unbalance,
misalignment ...etc). More investigation on the spike energy and the
frequency spectrum indicated a problem with races of the bearing.
Multi-harmonics of the fundamental defects frequencies were
observed. Two of them were close to each other in magnitude those
were the source of the beating phenomena.
Abstract: This paper explores the plant maintenance management system that has been used by giant oil and gas company in Malaysia. The system also called as PMMS used to manage the upstream operations for more than 100 plants of the case study company. Moreover, from the observations, focus group discussion with PMMS personnel and application through simulation (SAP R/3), the paper reviews the step-by-step approach and the elements that required for the PMMS. The findings show that the PMMS integrates the overall business strategy in upstream operations that consist of asset management, work management and performance management. In addition, PMMS roles are to help operations personnel organize and plan their daily activities, to improve productivity and reduce equipment downtime and to help operations management analyze the facilities and create performance, and to provide and maintain the operational effectiveness of the facilities.
Abstract: The manufacture of large-scale precision aerospace
components using CNC requires a highly effective maintenance
strategy to ensure that the required accuracy can be achieved over
many hours of production. This paper reviews a strategy for a
maintenance management system based on Failure Mode Avoidance,
which uses advanced techniques and technologies to underpin a
predictive maintenance strategy. It is shown how condition
monitoring (CM) is important to predict potential failures in high
precision machining facilities and achieve intelligent and integrated
maintenance management. There are two distinct ways in which CM
can be applied. One is to monitor key process parameters and
observe trends which may indicate a gradual deterioration of
accuracy in the product. The other is the use of CM techniques to
monitor high status machine parameters enables trends to be
observed which can be corrected before machine failure and
downtime occurs.
It is concluded that the key to developing a flexible and intelligent
maintenance framework in any precision manufacturing operation is
the ability to evaluate reliably and routinely machine tool condition
using condition monitoring techniques within a framework of Failure
Mode Avoidance.
Abstract: This paper provides a replacement policy for warranty products with different failure rate from the consumer-s viewpoint. Assume that the product is replaced once within a finite planning horizon, and the failure rate of the second product is lower than the failure rate of the first product. Within warranty period (WP), the failed product is corrected by minimal repair without any cost to the consumers. After WP, the failed product is repaired with a fixed repair cost to the consumers. However, each failure incurs a fixed downtime cost to the consumers over a finite planning horizon. In this paper, we derive the model of the expected total disbursement cost within a finite planning horizon and some properties of the optimal replacement policy under some reasonable conditions are obtained. Finally, numerical examples are given to illustrate the features of the optimal replacement policy under various maintenance costs.
Abstract: The Programmable Logic Controller (PLC) plays a
vital role in automation and process control. Grafcet is used for
representing the control logic, and traditional programming
languages are used for describing the pure algorithms. Grafcet is used
for dividing the process to be automated in elementary sequences that
can be easily implemented. Each sequence represent a step that has
associated actions programmed using textual or graphical languages
after case. The programming task is simplified by using a set of
subroutines that are used in several steps. The paper presents an
example of implementation for a punching machine for sheets and
plates. The use the graphical languages the programming of a
complex sequential process is a necessary solution. The state of
Grafcet can be used for debugging and malfunction determination.
The use of the method combined with a set of knowledge acquisition
for process application reduces the downtime of the machine and
improve the productivity.