Fuzzy Markov development for buried transportation bridges

Inspection interval of safety valves is generally at least once a year specified by the Safety Technical Supervision Regulations for Safety Valves (TSG ZF001-2006) in China, while the maintenance interval of production equipment in oil refining and chemical enterprises is usually 3–5 years. There is a prominent problem that inspection interval and maintenance interval is not consistent. This paper reveals the main failure modes, opening rate as well as maintenance replacement rate by analyzing the operation condition of 24720 safety valves of 322 typical units in oil refining and chemical industry. Moreover, the engineering application data of 44692 safety valves carried out with Risk-Based Inspection (RBI) method in the past eight years are statistically analyzed to reveal the risk distribution and risk level of safety valves. Based on the failure and risk level big data of safety valves, this paper proposes the safety valves inspection method to solve the contradiction between inspection interval and maintenance interval, which maybe further promote the legal compliance and long-term safe operation of production units in oil refining and chemical enterprises. The advantages and limitations of this method are summarized by sharing the engineering application cases in a petrochemical company about Safety valves inspection based on big data of failure and risk.

410O - Conditional Probability of Survival in Patients with Glioblastoma Multiforme in the Temozolomide Treatment Era

The capabilities of smart meters and the potential vulnerabilities they introduce make the Advanced Metering Infrastructure (AMI) a critical components of the Smart Grid. A virus propagating among meters and massively issuing remote disconnect commands could have catastrophic consequences. To understand and assess the risk posed by a given set of meter devices getting compromised, we introduce a novel modeling approach based on fuzzy Markov chain. Markov modeling is one of the main approaches used to analyze the reliability of critical systems. Markov models are usually analyzed using mathematical approaches or Monte-Carlo Simulation. For large and complex systems, traditional Markov models suffer from the state explosion problem and quickly become intractable. The key idea of our approach to solve this issue in the context of the smart grid is to leverage the hierarchical topology of AMI to build a high-level model where components represent Neighborhood Area Networks (NAN) instead of individual meters. The fuzzy Markov formalism enables our solution to take into account gradual component degradation. We propose a Fuzzy Markov Model (FMM) for reliability analysis of a Smart Grid. The reliability of each NAN is represented by a number in [0, 1] which denotes the state of that component in the FMM.We define the transition probabilities between states of the fuzzy Markov model based on the fuzzy time-to-failure value for each NAN. We utilize the Fuzzy Monte Carlo Simulation (FMCS) approach to calculate the fuzzy time-to-failure value of each NAN. Our proposed method enables us to assess the reliability of the entire AMI at the meter level while keeping the model manageable.

The disease outcome for patients with cancer is typically described in terms of estimated survival from diagnosis. Conditional probability offers more relevant information regarding survival for patients once they have survived for some time. We report conditional survival probabilities on the basis of 498 patients with glioblastoma multiforme receiving radiation and chemotherapy. For 1-year survivors, we evaluated variables that may inform subsequent survival. Motivated by the trend in data, we also evaluated the assumption of constant hazard. Patients enrolled onto seven phase II protocols between 1975 and 2007 were included. Conditional survival probabilities and 95% CIs were calculated. The Cox proportional hazards model was used to evaluate prognostic values of age, Karnofsky performance score (KPS), and prior progression 1-year post diagnosis. To assess the constant hazard assumption, we used a likelihood-ratio test to compare the Weibull and exponential distributions. The probabilities of surviving an additional year given survival to 1, 2, 3, and 4 years were 35%, 49%, 69%, and 93%, respectively. For patients who survived for 1 year, lower KPS and progression were significantly predictive of shorter survival (both P < .001), but age was not (hazard ratio, 1.22 for a 10-year increase; P = .25). The Weibull distribution fits the data significantly better than exponential (P = .02), suggesting nonconstant hazard. Conditional probabilities provide encouraging information regarding life expectancy to survivors of glioblastoma multiforme. Our data also showed that the constant hazard assumption may be violated in modern brain tumor trials. For single-arm trials, we advise using individual patient data from historical data sets for efficacy comparisons.

This article presents a risk‐based methodology to estimate shutdown inspection and maintenance interval considering system availability. Most inspection and maintenance activities are performed when the plant/unit is in the operational state. However, some inspection and maintenance activities require the plant to be in a nonoperational or shutdown state. In most cases, operating companies adopt a shutdown schedule based on the original equipment manufacturer's (OEM) suggested recommended periods. However, this may not be the best strategy as OEM recommended duration is general and may not reflect the current state of operation. The proposed methodology is unique in the sense that it identifies a shutdown interval by identifying the critical equipment in terms of risk considering availability and safety of the operating unit. It optimizes process plant shutdown interval to minimize the risk (in dollar terms). The Markov process is used to establish the state diagram to calculate system availability. The proposed methodology is comprised of three steps namely, risk‐based equipment selection, shutdown availability modeling of a complex system using the Markov process, and risk‐based shutdown inspection and maintenance interval modeling. It can be applied to process plants such as those for liquefied natural gas processing, petrochemicals, and refineries. The key elements for the success of the proposed methodology are the plant‐specific data and identification of critical equipment. © 2014 American Institute of Chemical Engineers Process Saf Prog 34: 267–279, 2015

With standard treatment for glioblastoma (GBM) consisting of surgery followed by radiotherapy (RT) with concurrent and adjuvant temozolomide (TMZ), median survival is ~14.6 months. This is not as informative to patients who have survived for some time. Conditional probability of survival may offer more relevant survival estimates. Outcomes/conditional probability of survival and post-progression survival (PPS) estimates were retrospectively reviewed in the TMZ treatment era of 882 consecutive patients with a diagnosis of GBM from January 2004 to August 2010. Median age of entire cohort was 62 years including 62 % males. Baseline performance status (PS) was 0-1 in 67, 23 % had frontal lobe tumors, 58 % received concurrent RT/TMZ ± adjuvant TMZ. Survival (OS) was similar for those with frontal lobe tumors versus other locations (P = 0.25). OS for patients receiving standard RT/TMZ ± TMZ was 14.2 months. Age, PS, extent of surgery, therapy post-surgery had significant effects on OS. OS for entire cohort at 1, 2, 3, 4, 5 years was 43.4, 17.9, 10.4, 8.4, 7.2 % respectively. Conditional probability of survival of an additional year given survival to 1, 2, 3, 4, 5 years was 41.4, 58, 80.7, 85.7, 81.5 % respectively. Conditional probability of survival for those patients receiving concurrent RT/TMZ ± adjuvant TMZ was similar. Patients who progress >18 months after their initial treatment for GBM had significantly greater 2 and 5 year PPS as well as OS. Conditional probabilities of survival may provide more meaningful life expectancy predictions for survivors of GBM than conventional survival outcomes.

Setting inspection intervals based on an accurate prediction of fatigue crack sizes is essential for sustaining the integrity of aeronautical structures. However, the fatigue crack growth and its prognosis are affected by various uncertainties, which makes the current inspection strategy with fixed intervals challenging in managing the aircraft with diverse damage states in a fleet. In this study, an intelligent crack inspection strategy is proposed based on a digital twin, in which a reduced-order fracture mechanics simulation methodology, a validated fatigue crack growth model, and the historical crack length inspection results are integrated into a dynamic Bayesian network. The proposed strategy uses two connected probabilistic processes, which conduct the diagnosis/prognosis and calculate the inspection intervals, respectively, to adaptively set the inspection intervals according to the updating of the digital twin model. The proposed inspection strategy is demonstrated by the various crack growth histories of a helicopter component and benchmarked against several baselines. The results show that the probability of failure can be kept below the threshold, even though the initial crack size and the crack growth parameters are underestimated in the prior distribution. Further applications on more realistic aircraft structures will be carried out in the future.

A decision framework is introduced to assist municipal engineers and planners to optimize decisions regarding the renewal of large infrastructure assets such as water transmission pipes, trunk sewers, or other assets with high costs of failure, inspection, and condition assessment. The proposed decision framework identifies a need for immediate intervention or, alternatively, enables optimization of the scheduling of the next inspection and condition assessment. The deterioration of the asset is modeled as a semi-Markov process and is thereby discretized into condition states. The waiting times in each state are assumed to be random variables with “known” probability distributions. If pertinent data are scarce (as is typical in most municipalities) these probability distributions can be initially derived based on expert opinion. These distributions will then be continually updated as observed deterioration data are collected over time. Monte Carlo simulation is used to calculate the distributions of the cumulative waiting times. Conditional survival probabilities are used to compile age-dependent transition probability matrices in the various states. The expected discounted total cost associated with an asset (including cost of intervention, inspection, and failure) is computed as a function of time. The time to schedule the next inspection/condition assessment is when the total expected discounted cost is minimum. Immediate intervention should be planned if the time of minimum cost is less than a threshold period (2 to 3 years) away. A computer program is prepared for demonstration and proof of concept. The decision framework lends itself to a computer application fairly easily. Although usable in its current form, this paper identifies some issues that require as yet unavailable data as well as more research in order to develop the framework into a comprehensive application tool.

Conditional probability of survival in patients with biliary atresia after Kasai portoenterostomy: a Chinese population-based study

Large gas turbine combustion chambers, being arranged outside of the unit, exhibit quite a lot of advantages with respect to combustion. Moreover, they are characterized by a long life of all components. Thus, in case of such gas turbine units the maintenance and inspection intervals are relatively large being not determined by the combustion chamber or combustion chamber components. There are not many failures. They may easily be recognized at their initial stage and can be eliminated quickly as the inside is accessible via a manhole. This in turn has a positive effect on overall maintenance and service cost. Besides, this easy accessibility allows for a direct examination of the turbine inner casing and the first turbine stages in case of maintenanced works. Experiences are based on the operation of more than 100 gas turbines of such a kind, whereby several have been run at peak load with more than 5000 starts, others at base load with more than 100,000 operating hours.

BP has been operating in the Caspian Sea and associated area (AGT Region - Azerbaijan, Georgia and Turkey) since the early 1990's and today the area is one of the world's leading hydrocarbon provinces. The assets consist of 7 offshore platforms, one of the world's largest terminals and export pipelines. The topic of this paper is a description of a practical implementation of an assessment procedure for determining optimum inspection frequencies (endorsement interval) for pressure relief devices protecting pressure system equipment. We describe the development process, the resulting solution and work processes, the challenges related to the data requirements and dependency of the maintenance and inspection organizations to record and collate such data. Within the AGT Region there are in excess of 5,800 pressure relief devices. Pressure relief devices provide a key layer of protection for the process safety of oil and gas fields and installations and need to operate on demand as and when required. All need to be inspected, tested and recertified on a regular basis. The objective of the exercise from the outset was to increase the process safety protection and reliability on the installations. Flow charts were developed to ensure the methodology was clear, simple, repeatable and auditable. This methodology was then combined with knowledge and expertise on the failure mechanisms of safety devices and the consequences of an uncontrolled release to atmosphere to determine an optimal endorsement interval and verify that the device was functioning properly thereby preventing the over-pressuring (or under pressuring) of equipment during abnormal operating or emergency conditions or leakage past the seat during service. The output from the work showed that around 70% of the pressure relief devices could have their maintenance and inspection intervals increased, by up to 6 years. The work was initiated when the oil price was in excess of 100US$ per barrel and on BP facilities that are top of the league in terms of production efficiency (greater than 90%). The primary objective was to improve the safety, reliability and integrity of the installations, however the resulting benefits of performing the assessments aligned to the ability to increase inspection intervals are to significantly reduce maintenance costs, reduce offshore and onshore interventions and increase availability and production efficiency. The team also reviewed and enhanced the management procedure for the testing, repair and recertification of pressure relief devices, to provide additional guidance to the maintenance teams in the areas of data collection and recording and witnessing pressure relief device testing and re-certification.. To the authors knowledge this is the first practical implementation of the assessment procedure that embraces the total combination of asset integrity information and work process data sources within a whole oil and gas producing region.

BackgroundLeft ventricular assist devices (LVADs) improve outcomes in patients with end‐stage heart failure and are increasingly implanted for destination therapy. We describe dynamic estimates of event‐free survival with conditional survival probabilities in a destination therapy LVAD population.Methods and ResultsWe studied 8245 adult patients in INTERMACS (Interagency Registry for Mechanically Assisted Circulatory Support) implanted with a continuous‐flow destination therapy LVAD. The composite primary end point was death, device exchange or removal, or heart transplantation. Conditional survival probabilities were calculated and stratified by implantation characteristics and nonfatal adverse events experienced within the first year after implant. Probabilities of surviving an additional 1 to 3 years were numerically higher after longer prior event‐free survival. INTERMACS profile 1, extracorporeal membrane oxygenation support, prior or concomitant surgery, and dialysis within 48 hours of implantation were associated with significantly lower event‐free survival in the first year but did not impact event‐free survival beyond then. For patients who experienced a nonfatal adverse event within the first year, subsequent 1‐year conditional survival was lower than in the absence of that event for stroke (65% [95% CI, 57%–73%] versus 75% [95% CI, 73%–77%]; P<0.001), device‐related infection (64% [95% CI 57%–71%] versus 76% [95% CI, 74%–78%]; P<0.001), and pump thrombosis or malfunction (64% [95% CI, 57%–70%] versus 76% [95% CI, 74%–78%]; P<0.001).ConclusionsConditional survival in patients with destination therapy LVADs improves over time, even for patients with unfavorable implantation characteristics. However, LVAD‐related complications including stroke, device‐related infection, and pump thrombosis or malfunction have an enduring negative influence on dynamic estimates of long‐term prognosis.

This paper considers the problem of finding the location and propagation path of chemical source using multiple intensity sensors. Source origin localization is treated as the sequence estimation problem over a fuzzy hidden Markov model. The main advantage of using fuzzy hidden Markov model is its inherent nonstationary behavior with time varying state transition probabilities which characterizes the uncertainties in advection and diffusion of the plume source. Unfortunately, estimation algorithms of Viterbi type can be very inefficient due to the large size of state space. Motivated by a recent theoretical result on low complexity maximum likelihood sequence detection, we propose a greedy heuristic algorithm to obtain a candidate source path and search only for state sequences within a constrained Hamming distance from the candidate path. Our approach is applicable to fuzzy hidden Markov decoding problem with a general class of fuzzy measures and fuzzy integrals. From simulation study, we found that our algorithm has plume localization error close to that using fuzzy Viterbi algorithm. It is very effective for state estimation over a long observation sequence when the decoding error probability is small.

Performance Evaluation of Link Prediction Techniques Based on Fuzzy Soft Set and Markov Model

Measuring effectiveness of movement-based three-way decision using fuzzy Markov model