Field Failure Data Research Articles

Analysis and Diagnosis of SLA Violations in a Production SaaS Cloud

A software-as-a-service (SaaS) needs to provide its intended service as per its stated service-level agreements (SLAs). While SLA violations in a SaaS platform have been reported, not much work has been done to empirically characterize failures of SaaS. In this paper, we study SLA violations of a production SaaS platform, diagnose the causes, unearth several critical failure modes, and then, suggest various solution approaches to increase the availability of the platform as perceived by the end user. Our approach combines field failure data analysis (FFDA) and fault injection. Our study is based on 283 days of operational logs of the platform. During this time, the platform received business workload from 42 customers spread over 22 countries. We have first developed a set of home-grown FFDA tools to analyze the log, and second implemented a fault injector to automatically inject several runtime errors in the application code written in .NET/C#, and then, collate the injection results. We summarize our finding as: first, system failures have caused 93% of all SLA violations; second, our fault injector has been able to recreate a few cases of bursts of SLA violations that could not be diagnosed from the logs; and third, the fault injection mechanism could recreate several error propagation paths leading to data corruptions that the failure data analysis could not reveal. Finally, the paper presents some system-level implication of this study and how the joint use of fault injection and log analysis may help in improving the reliability of the measured platform.

Reliability based design of fluid power pitch systems for wind turbines

AbstractThis paper presents a qualitative design tool for evaluation of the risk for fluid power pitch systems. The design tool is developed with special attention to industry standard failure analysis methods and is aimed at the early phase of system design. Firstly, the concept of Fault Tree Analysis is used for systematic description of fault propagation linking failure modes to system effects. The methodology is conducted solely on a circuit diagram and functional behavior. The Failure Mode and Effects Criticality Analysis is subsequently employed to determine the failure mode risk via the Risk Priority Number. The Failure Mode and Effect Criticality Analysis is based on past research concerning failure analysis of wind turbine drive trains. Guidelines are given to select the severity, occurrence and detection score that make up the risk priority number. The usability of the method is shown in a case study of a fluid power pitch system applied to wind turbines. The results show a good agreement to recent field failure data for offshore turbines where the dominating failure modes are related to valves, accumulators and leakage. The results are further used for making design improvements to lower the overall risk of the pitch system. Copyright © 2017 John Wiley & Sons, Ltd.

Assessment of the Reliability of the Indian Tsunami Early Warning System

AbstractThis paper analyses the reliability of the Indian Tsunami Early Warning System (ITEWS), comprising a 24 × 7 manned and automated center capable of monitoring the seismic, open sea water level and coastal tide levels and disseminating tsunami bulletins with the aid of proven prerun scenario models during a tsunamigenic earthquake. Since its inception in 2007, the ITEWS has undergone technological maturity with reliability as the prime objective. The system is expected to be in operation throughout the year and alerting the entire Indian Ocean rim countries in the event of a tsunami. Based on International Electrotechnical Commission (IEC) 61508 standards and field failure data, quantitative reliability modeling is done for the subsystems, and it is found that the seismic network, tsunami buoy network, and distress information dissemination systems conform to Safety Integrity Level SIL4, while tide gauge stations conform to SIL4 with a maintenance interval of 45 days. In case of the tsunami buoy network, the failure of one tsunami buoy degrades the network to SIL3 and needs to be restored within 8 months. The study provides confidence on ITEWS's reliable support to tsunami early warning.

Key Technologies of the Independent and Innovative Communications-Based Train Control Systems in China

In the past 10 years, the independent and innovative communications-based train control system has experienced breakthroughs in key technologies by conducting pilot experiments, field tests, and demonstration projects. For the first time, the design theories and technical aspects of moving block systems, which cover the entire life cycle of fail-safe systems, are set up. The models and algorithms are proposed to control the safe operation of trains. The safety technologies for short-interval train sequencing are adopted to realize the 90 s headway between consecutive trains. The optimized automatic train operation strategies are used to save energy and allow trains to safely reach their destinations. The highly dependable and bi-directional train-to-ground communication technologies are introduced for reliable transmission of safety related train-to-ground information in a complex environment which is compatible with multiple media, such as free waves, leaky waveguides, and leaky feeders. The simulation, tests, and verification technologies based on the minimum system and test cases were adopted to conduct the following: principle demonstration, functional tests, integrated tests, field failure data replay, analysis and disposal, system upgrade, and maintenance.

Reliability estimation considering usage rate profile and warranty claims

Providing more realistic reliability prediction based on small proportion of failed population or test data has always been a challenging task. Manufacturers rely heavily on reliability prediction for designing warranty plan. Furthermore, to predict warranty claims for the remaining warranty period, it is important to have more realistic reliability assessment by considering a larger proportion of the population or the maximum possible information on the remaining population. However, generally this information is not readily available and is very difficult to gather on the scattered population. In this work, we propose to use customer usage rate profile to generate censored usage data on the remaining population that do not have any failure and warranty claims yet. We intend to use field data available such as warranty claims, field failures, recall data, and maintenance data to develop usage rate profile and subsequently estimate censored usage time. Finally, reliability estimation methodology is developed considering both censored data and field failure data to provide more reasonable reliability prediction for the remaining warranty period. The proposed methodology is demonstrated considering real-life data from a big manufacturing company.

Reliability Analysis of the CNC System Based on Field Failure Data in Operating Environments

Reliability is a measure of how well a product will perform under a certain set of conditions for a specified amount of time especially in the field environments. In this paper, a reliability study of a computer numerical control (CNC) system is described. For this analysis, field failure data from a shop manufacturing factory collected over the course of a year on approximately 20 CNC machine tools during their operating period were analyzed. Based on the field failure data, the two-parameter exponential distribution was found to be applicable to describe the time between failures of the CNC system from among many distributions including Weibull, gamma, two-parameter exponential, normal, and logistic using the chi-squared test. In this paper, we discuss the reliability estimation of the CNC system based on the collected field failure data from a manufacturing factory using the maximum likelihood estimate (MLE) and uniform minimum variance estimate (UMVUE) methods. We also discuss the confidence intervals of the mean residual lifetime and reliability function. The result shows that the UMVUE method can provide much better and more accurate results in estimating the reliability of the CNC system than the MLE. This finding, on the one hand, seems to be obvious because the UMVUE is not only an unbiased estimator but also sufficient statistic with the smallest variance; on the other hand, it is not straightforward to obtain the UMVUE of any complex function, which is the reliability function in this case. This is a very important finding and is very encouraging because it indicates that the reliability analysis of the CNC system based on the UMVUE can be more than compensated by the ability of the complexity of parameter estimation method to better evaluate and predict the reliability of the CNC system. Hence, we believe that it is worth the effort to derive those parameter functions using UMVUE method. Copyright © 2015 John Wiley & Sons, Ltd.

FMECA를 통한 전차선로 가선시스템의 신뢰도 분석에 관한 연구

The reliability of catenary system is very important for uninterrupted train operation as it supplies electric power to train without redundant facilities. This paper provides a systematic approach to the reliability analysis of the catenary system based on FMECA procedures defined by global standards such as MIL Std 1692a and IEC 60812. Field failure data collected from the operation and maintenance of high-speed railway catenary system for 9 years are used to derive critical failure modes and to evaluate the criticality of the failure modes. Evaluation of the criticality are carried out by quantitative procedures defined by MIL Std 1692a and by criticality matrix defined by IEC 60812. FMECA results suggests that three critical failure modes should be checked carefully during maintenance work, which include strand break of dropper and voltage equalizing wire, power supply failure of feeder. Maintenance procedure of catenary system in order of importance is suggested too. These results can be applied to maintenance planning and design of catenary system to improve the reliability of electric railway system.

Uncertainty Importance Measure of Individual Components in Multi-State Systems

Traditional reliability importance measures have been successfully extended from binary-state models to multi-state models. The calculation of these measures typically relies on the true reliabilities of components. In reality, however, the true values of component reliabilities are usually unknown, and they are generally approximated by their estimates generated from testing or field failure data. The accuracy of the estimates is limited by the available data. Research on uncertainty importance measures (UIMs) has emerged on this account to rank components based on their potentials to reduce the uncertainty about the estimated system reliability. The UIMs of components for binary-state models are well studied, but there is a lack of studies dedicated to multi-state models. In this paper, the reliability estimator and the corresponding uncertainty (characterized by the variance estimator) are derived for multi-state systems with structures such as serial, parallel, bridge, and their more complex combinations. The derivation process utilizes multinomial reliability testing and the universal generating function method. With the help of the derived estimators, we extend uncertainty importance research to multi-state models through a newly defined variance-based measure. Examples are provided to demonstrate the proposed ideas.

Optimal Redundancy and Maintenance Strategy Decisions for Offshore Wind Power Converters

Analysis of field failure data collected from various wind farm databases indicates that the power converters are among the most critical components in offshore wind turbines, since they suffer from a high failure rate. One efficient approach to enhance the reliability and availability of the wind power systems is through using a redundant converter design, in which a set of power converters is placed together in parallel. The main advantage of a multiple parallel converter system is that the failure of one converter will not necessarily lead to the failure of the entire system. It may however increase the wind turbine's acquisition cost, volume, and weight. In this paper, we propose an approach of joint redundancy and maintenance strategy optimization for offshore wind power converters, aiming to simultaneously determine the "optimal allocation of redundant converters" and the "optimal threshold number of converters that are allowed to fail before sending a maintenance crew to the offshore platform". The optimal solution under various system-level constraints (such as reliability, weight, and the available space in nacelle) is derived and the conditions required to make using a redundant system beneficial are discussed. The proposed design is applied to an offshore wind turbine system and its performance is evaluated using a Monte-Carlo simulation technique. Finally, the results are compared with the conventional power converter system and a sensitivity analysis is conducted in order to make the proposed approach applicable for the next generation of wind turbines.

Remaining Useful Life Estimation Method for the Turbine Blade of a Civil Aircraft Engine Based on the QAR and Field Failure Data

Remaining Useful Life Estimation Method for the Turbine Blade of a Civil Aircraft Engine Based on the QAR and Field Failure Data

A practical approach for reliability and maintainability analysis of repairable systems: a case study of polypropylene production for food packaging applications

The polypropylene (PP) industry is an important branch of industry with rapid growth and prospects. The production process is complex and requires the continuous operation of the line without failure of equipment to avoid quality problems and maintain the performance of the system. In this study, reliability and maintainability analysis was conducted for PP industry by applying statistical techniques on field failure data. Data collection from the line and their analysis were valid over a long time of 12 months. It was found out that (a) the real operating time of the PP production line (PL) is 92.21%, whereas the rest 7.79% is under repair, (b) the line reliability can be improved if it is focused firstly on workstation auxiliary supplies (WS.8), and then on packing (WS.6), and (c) the maintainability can be improved on the reactor (WS.2), extruder (WS.5), and the entire production line as well.

Design for Reliability of Complex System: Case Study of Horizontal Drilling Equipment with Limited Failure Data

Reliability is an important phase in durable system designs, specifically in the early phase of the product development. In this paper, a new methodology is proposed for complex systems’ design for reliability. Specific test and field failure data scarcity is evaluated here as a challenge to implement design for reliability of a new product. In the developed approach, modeling and simulation of the system are accomplished by using reliability block diagram (RBD) method. The generic data are corrected to account for the design and environment effects on the application. The integral methodology evaluates reliability of the system and assesses the importance of each component. In addition, the availability of the system was evaluated using Monte Carlo simulation. Available design alternatives with different components are analyzed for reliability optimization. Evaluating reliability of complex systems in competitive design attempts is one of the applications of this method. The advantage of this method is that it is applicable in early design phase where there is only limited failure data available. As a case study, horizontal drilling equipment is used for assessment of the proposed method. Benchmarking of the results with a system with more available failure and maintenance data verifies the effectiveness and performance quality of presented method.

Reliability Defects Identification of Serial Production Systems: Application to a Piston Production Line

Failure and repair data are urgently needed for reliability and maintainability analysis of production lines. A piston production line is traced, and field failure data during a period of 15 months are collected. The statistical analysis of the failure and repair data and the determination of the theoretical distributions parameters that best fit to failure data were carried out, based on which, reliability of the piston production line is evaluated from a global perspective. Furthermore, a failure losses-based importance measure method for repairable systems was proposed and used for reliability defects identification of the piston production line from a local perspective, the primary failure modes and equipments that were prone to failures were ranked with this method. This analysis will be essential and valuable to identify the reliability problems for current and further designing projects of production lines for piston or other important parts of motor engine.

Effect of preventive maintenance intervals on reliability and maintenance costs of wind turbine gearboxes

AbstractAs many of the installed wind turbines (WTs) get older or approach their design life, there will be a drive to keep extending the lives of the main components especially the gearbox. The challenge of operations and maintenance will potentially be even more as there will be a need to keep the cost to a minimum. Similarly, as years of experience of operating WTs accumulate, knowledge about the behaviour and failure of subsystems is gained as well. Also with good documentation and repository of historical operational, performance and failure data, future decisions of operations and maintenance can be taken on the basis of insights from past experience. This paper presents an approach for implementing preventive maintenance (PM) by using historical failure data to determine the optimal PM interval required to maintain desired reliability of a typical module or subassembly. This paper builds upon previous research in the area of WT gearbox reliability analysis and prediction, taking it further by examining the relationships between the frequency of a PM task and the reliability, availability and maintenance costs. The approach presented demonstrates how historical in‐service failure data can be used in PM task selection based on the minimum maintenance cost and maximum availability. Available historical field failure data of the high speed module of a Vestas 2MW WT gearbox is used to validate the approach and show its practicality. The results of this study are then presented—indicating that choosing the right PM interval based on the minimum unit maintenance cost and maximum availability also improves WT gearbox reliability. Copyright © 2014 John Wiley & Sons, Ltd.

On analysis of incomplete field failure data

Many commercial products are sold with warranties and indirectly through dealers. The manufacturer-retailer distribution mechanism results in serious missing data problems in field return data, as the sales date for an unreturned unit is generally unknown to the manufacturer. This study considers a general setting for field failure data with unknown sales dates and a warranty limit. A stochastic expectation-maximization (SEM) algorithm is developed to estimate the distributions of the sales lag (time between shipment to a retailer and sale to a customer) and the lifetime of the product under study. Extensive simulations are used to evaluate the performance of the SEM algorithm and to compare with the imputation method proposed by Ghosh [Ann. Appl. Stat. 4 (2010) 1976-1999]. Three real examples illustrate the methodology proposed in this paper.

Survival Analysis of Automobile Components Using Mutually Exclusive Forests

An ability to predict the mileage at failure of components in a complicated system, particularly in automobiles, is a challenging task. In the current work, a methodology for estimating the distribution of failure and survival rate of automobile components affected by multiple factors is presented. A novel adaptation of an ensemble recursive partitioning and tree-based learning method, mutually exclusive forest, is introduced. The proposed method is capable of handling a high dimensional dataset and maximizing the extracted information to estimate the distribution of mileage at failure of automobile components. Each tree in the proposed mutually exclusive forest uses a mutually exclusive set of factors in each of its constituent decision trees to classify the failure data. Information across the trees is combined to obtain the failure rate distribution of an automobile component with respect to mileage. A case study, based on real-world field failure data and censored data of automobile components, is presented to evaluate the proposed algorithm. Results show similar results to the C-Forest approach in terms of prediction quality, while generating models with significantly lower space that are easier to interpret.

A new reliability analysis method for horizontal machining centre based on proportion intensity model and nonlinear mixed-effects model

The reasonable reliability model is needed to best approximate the engineering reality and capture the physical intrinsic underlying mechanisms of the repairable systems. Traditional reliability models are inadequate in modelling the field failure data, as they ignore individual copy-specific random effects and correlation among repeated measurements. This paper presents an integrated method based on modified proportional intensity model and nonlinear mixed-effect model. This method can capture the intrinsic characteristic of engineering systems and evaluate quantitatively the effects of deterioration and repair on the reliability of equipment. Especially, the analytical solutions of reliability and the failure rate of equipment can be obtained by this method. Then, a case study is presented based on field failure data for 14 horizontal machining centres (HMC) over one year. The results demonstrate the effectiveness of this integrated method to analyse the reliability of multi-copy repairable systems.

How do heterogeneities in operating environments affect field failure predictions and test planning?

The main objective of accelerated life tests (ALTs) is to predict fraction failings of products in the field. However, there are often discrepancies between the predicted fraction failing from the lab testing data and that from the field failure data, due to the yet unobserved heterogeneities in usage and operating conditions. Most previous research on ALT planning and data analysis ignores the discrepancies, resulting in inferior test plans and biased predictions. In this paper we model the heterogeneous environments together with their effects on the product failures as a frailty term to link the lab failure time distribution and field failure time distribution of a product. We show that in the presence of the heterogeneous operating conditions, the hazard rate function of the field failure time distribution exhibits a range of shapes. Statistical inference procedure for the frailty models is developed when both the ALT data and the field failure data are available. Based on the frailty models, optimal ALT plans aimed at predicting the field failure time distribution are obtained. The developed methods are demonstrated through a real life example.

RELIABILITY IMPROVEMENT METHODOLOGY FOR CONSUMER PRODUCTS - A CASE STUDY

The objective of every industry is to bring world-class quality products to the market in shortest possible time with least expenses. To meet this objective manufacturer’s always remains non compliant with Reliability requirement of the products due to which field failure rate increases and in turn increases the warranty cost. In this work, an attempt has been made to improve the field reliability for consumer product through a proposed step by step Reliability Improvement Methodology. The proposed methodology includes analysis of field failure data, failure analysis for identifying root cause, experimental investigation to confirm root causes, design modification recommendations and improvement validation through accelerated testing. Further, this analysis helps the manufacturers in warranty extension decision making for the existing products and in the improvement of the next generation product’s performance which directly increases the brand image and profits of an organization.

Availability-based optimal maintenance policies for repairable systems in military aviation by identification of dominant failure modes

Modeling of imperfect repair through perfect renewal process uses an “as good as new” repair assumption and nonhomogeneous Poisson process uses an “ABAO” repair assumption. In practice, repair actions do not result in such extreme situations but in a complex transitional one, that is, general repair. This article discusses generalized renewal process for an aero engine as repairable component. Maximum likelihood estimators for the reliability parameters are estimated using generalized renewal process, for field failure data of an aero engine. The current practice designates repairable components, as high failure rate components based on intuition, experience and the number of unscheduled failures at repair depots. A methodology is developed to designate high failure rate components based on availability by taking into consideration the dominant failure modes of the aero engine. Then, a comparison is made with a “Black Box” approach. The present maintenance policy is then reviewed by reducing the present time between overhauls for the high failure rate components. We observe a noteworthy enhancement in all the performance parameters in the suggested maintenance policy. We also observe that percent improvement achieved in all performance parameters on reducing the next overhaul cycle time, with failure modes analysis is more than when failure modes are not considered.