• The KIPS Transactions:PartD
• /
• v.8D no.4
• /
• pp.387-392
• /
• 2001

This Many software projects collect grouped failure data (failures in some failure interval or in variable time interval) rather than individual failure times or failure count data during the testing or operational phase. This paper presents the neural network (NN) modeling for grouped failure data that is able to predict cumulative failures in the variable future time. The two variant models of cascade-correlation learning (CasCor) algorithm are presented. Suggested models are compared with other well-known NN models and statistical software reliability growth models (SRGMs). Experimental results show that the suggested models show better predictability.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.8
• /
• pp.3086-3101
• /
• 2021

To supply precise marketing and differentiated service for the electric power service department, it is very important to predict the customers with high sensitivity of electric power failure. To solve this problem, we propose a novel grouped 𝑙_1/2 sparsity constrained logistic regression method for sensitivity assessment of electric power failure. Different from the 𝑙₁ norm and k-support norm, the proposed grouped 𝑙_1/2 sparsity constrained logistic regression method simultaneously imposes the inter-class information and tighter approximation to the nonconvex 𝑙₀ sparsity to exploit multiple correlated attributions for prediction. Firstly, the attributes or factors for predicting the customer sensitivity of power failure are selected from customer sheets, such as customer information, electric consuming information, electrical bill, 95598 work sheet, power failure events, etc. Secondly, all these samples with attributes are clustered into several categories, and samples in the same category are assumed to be sharing similar properties. Then, 𝑙_1/2 norm constrained logistic regression model is built to predict the customer's sensitivity of power failure. Alternating direction of multipliers (ADMM) algorithm is finally employed to solve the problem by splitting it into several sub-problems effectively. Experimental results on power electrical dataset with about one million customer data from a province validate that the proposed method has a good prediction accuracy.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.12
• /
• pp.3821-3828
• /
• 2000

Many software projects collect grouped failure data (failures in some failure interval or in variable time interval) rather than individual failure times or failure count data during the testing or operational phase. This paper presents the neural network (NN) modeling that is dble to predict cumulative failures in the variable future time for grouped failure data. ANN's predictive ability can be affected by what it learns and in its ledming sequence. Eleven training regimes that represents the input-output of NN are considered. The best training regimes dre selected rJdsed on the next' step dvemge reldtive prediction error (AE) and normalized AE (NAE). The suggested NN models are compared with other well-known KN models and statistical software reliability growth models (SHGlvls) in order to evaluate performance, Experimental results show that the NN model with variable time interval information is necessary in order to predict cumulative failures in the variable future time interval.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.4A
• /
• pp.399-414
• /
• 2004

In this paper, an evolution of software reliability engineering in a large-scale software project is summarized. The considered software consists of many components, called functional blocks in software of switching system. These functional blocks are served as the unit of coding and test, and the software is continuously updated by adding new functional blocks. We are mainly concerned with the analysis of the effects of these software components in software reliability and reliability evolution. We analyze the static characteristics of the software related to software reliability using collected failure data during system test. We also discussed a pattern which represents a local and global growth of the software reliability as version evolves. To find the pattern of system software, we apply the S-shaped model to a collection of failure data sets of each evolutionary version and the Goel-Okumoto(G-O) model to a grouped overall failure data set. We expect this pattern analysis will be helpful to plan and manage necessary human/resources fur a new similar software project which is developed under the same developing circumstances by estimating the total software failures with respect to its size and time.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.64 no.4
• /
• pp.251-255
• /
• 2015

Chiller system plays an important role of maintaining room temperature constantly by supplying chilled water to Heating, Ventilating and Air Conditioning(HVAC)or area room cooler equipment during plant normal operation or accident condition. Chiller failures are one of the most frequently occurring equipment failures. If the types of chiller failures are analyzed and grouped thoroughly, it would be helpful to make chiller maintenance strategy at the plants. That would enhance equipment reliability of chiller in the end. In this paper, chiller failure data during three years were analyzed and categorized by specific failure code. In addition, the various proposals to improve equipment reliability of chiller were suggested such as Preventive Maintenance Optimization(PMO) strategy and performance monitoring reinforcement and so on.

• Journal of Biosystems Engineering
• /
• v.42 no.2
• /
• pp.75-79
• /
• 2017

The use of tractors to carry out agricultural work has played an important role in mechanizing the agricultural sector. A repairable mechanical system (such as an agricultural tractor) is subject to deterioration or failure. In this study, a regression model was used to predict the failure rate of a John Deere 3140 tractor. The machine failure pattern was carefully studied, and key factors affecting the failure rate were identified in five regions of the Khouzestan province. Through a questionnaire, data was obtained from farm records. This data was grouped into six sub-groups, according to the annual use hours (AUH) and the manner in which the tractors were stored. Results showed that AUH and storage policies affected failure rate slightly. With an increase in the age of the tractors, the failure rate in the tractors used for 1050-2000 hours annually and stored outdoors was higher than those used for 200-1000 hours annually and stored in sheds. When the tractors were of the same age, the slope of the curve in the 200-1000 annual use hours increased gradually and then rapidly, but failure rate in the 1050-2000 annual use hours was high from the beginning, and subsequent increase in this value was almost uniform. As a result, it can be said that with an increase in the annual use hours, the failure and breakdown rate in John Deere 3140 tractors rapidly increases, but maintenance conditions only slightly affect the failure and breakdown rate.

• Electronics and Telecommunications Trends
• /
• v.14 no.4 s.58
• /
• pp.1-9
• /
• 1999

In this paper, we summarize the lessons learned from the applications of the software reliability engineering to a large-scale software project. The considered software is the software system of the TDX-10 ISDN switching system. The considered software consists of many components, called functional blocks. These functional blocks serve as the unit of coding and test. The software is continuing to be developed by adding new functional blocks. We are mainly concerned with the analysis of the effects of these software components to software reliability and with the analysis of the reliability evolution. We analyze the static characteristics of the software related to software reliability using failure data collected during system test. We also discussed a pattern which represents a local and global growth of the software reliability as version evolves. To find the pattern of software of the TDX-10 ISDN system, we apply the S-shaped model to a collection of failure data sets of each evolutionary version and the Goel-Okumoto (G-O) model to a grouped overall failure data set. We expect this pattern analysis will be helpful to plan and manage necessary human/resources for a new similar software project which is developed under the same developing circumstances by estimating the total software failures with respect to its size and time.

• Communications for Statistical Applications and Methods
• /
• v.14 no.2
• /
• pp.309-316
• /
• 2007

In this paper, we decompose the whole likelihood based on grouped data into conditional likelihoods and study the approximate contribution of additional inspection to the efficiency. We also combine the conditional maximum likelihood estimators to construct an approximate maximum likelihood estimator. For an exponential distribution, we see that a large inspection size does not increase the efficiency much if the failure rate is small, and the maximum likelihood estimator can be approximated with a linear function of inspection times.

• Hip & pelvis
• /
• v.35 no.3
• /
• pp.175-182
• /
• 2023

Purpose: This study aims to compare patients in whom fixation failure occurred via cut-out (CO) or cut-through (CT) in order to determine patient factors and radiographic parameters that may be predictive of each mechanism. Materials and Methods: This retrospective cohort study includes 18 patients with intertrochanteric (IT) hip fractures (AO/OTA classification 31A1.3) who underwent treatment using a single lag screw design intramedullary nail in whom fixation failure occurred within one year. All patients were reviewed for demographics and radiographic parameters including tip-to-apex distance (TAD), posteromedial calcar continuity, neck-shaft angle, lateral wall thickness, and others. Patients were grouped into cohorts based on the mechanism of failure, either lag screw CO or CT, and a comparison was performed. Results: No differences in demographics, injury details, fracture classifications, or radiographic parameters were observed between CO/CT cohorts. Of note, a similar rate of post-reduction TAD>25 mm (P=0.936) was observed between groups. A higher rate of DEXA (dual energy X-ray absorptiometry) confirmed osteoporosis (25.0% vs. 60.0%) was observed in the CT group, but without significance. Conclusion: The mechanism of CT failure during intramedullary nail fixation of an IT fracture did not show an association with clinical data including patient demographics, reduction accuracy, or radiographic parameters. As reported in previous biomechanical studies, the main predictive factor for patients in whom early failure might occur via the CT effect mechanism may be related to bone quality; however, conduct of larger studies will be required in order to determine whether there is a difference in bone quality.

• Restorative Dentistry and Endodontics
• /
• v.37 no.2
• /
• pp.74-78
• /
• 2012

Objectives: This study aimed to evaluate the relationship between the cyclic fatigue of a K3 file and internal stress intentionally induced until the activation of the autostop function of the torque-controlled motor. Materials and Methods: K3 (Sybron Endo) .04 and .06 taper, size 25, 30, 35, 40 and 45 were used in this study. To give the internal stress, the K3 files were put into the .02 taper Endo-Training-Bloc (Dentsply Maillefer) until the activation of the auto-stop function of the torque-controlled motor. The rotation speed was 300 rpm and torque value was 1.0 $N{\cdot}cm$. K3 were grouped by the number of induced internal stress and randomly distributed to 4 experimental groups (n = 10, Stress 0 [control], Stress 1, Stress 2 and Stress 3). For measuring the cyclic fatigue failure, the K3 files were worked against a sloped glass block and time for file separation was recorded. Data was statistically analyzed Statistical analyses were performed using two-way ANOVA and Duncan post-hoc test at p < 0.05 level. Results: Except .04 taper size 30 in Stress 1 group, there were statistically significant differences in time for file separation between control and all experimental groups. K3 with .04 taper showed higher cyclic fatigue resistance than those of .06 taper. Conclusion: In the limitation of this study, the cyclic fatigue of the K3 file was influenced by the accumulated internal stress from use until the auto-stop function was activated by the torque-controlled motor. Therefore, clinicians should avoid the reuse of the K3 file that has undergone auto-stops.