Improving Process Safety and Product Quality using Large Databases

Pariyani, Ankur; Seider, Warren D.; Oktem, Ulku; Soroush, Masoud

doi:10.1016/s1570-7946(10)28030-6

Cited by 8 publications

(6 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some examples are Accimap, system-theoretic process analysis (STPA), Blended Hazid (BLHAZID), Dynamic Procedure for Atypical Scenarios Identification (DyPASI), and Resilience-Based Integrated Process Systems Hazard Analysis (RIPSHA). , However, these methods are limited in their application due to their qualitative nature or lack of common terminology and, hence, are good for only screening purposes. Some work has been done in modeling early warning signals or early fault detection in the chemical industry. , One of the major observations in the past decade made by many researchers in the area is that incidents continue to happen, although advancements in risk assessment methods have taken place, but something more is needed . With the application of the method proposed and demonstrated in this manuscript, we will overcome issues such as the limited historical database, missed scenarios in HAZOP, disintegrated social aspects analysis.…”

Section: Predictability Assessment Methodologymentioning

confidence: 99%

“…Some work has been done in modeling early warning signals or early fault detection in the chemical industry. 22,23 One of the major observations in the past decade made by many researchers in the area is that incidents continue to happen, although advancements in risk assessment methods have taken place, but something more is needed. 11 With the application of the method proposed and demonstrated in this manuscript, we will overcome issues such as the limited historical database, missed scenarios in HAZOP, disintegrated social aspects analysis.…”

Section: Predictability Assessment Methodologymentioning

confidence: 99%

See 1 more Smart Citation

Resilience-Based Process Upset Event Prediction Analysis for Uncertainty Management Using Bayesian Deep Learning: Application to a Polyvinyl Chloride Process System

Jain

Chakraborty

Pistikopoulos

et al. 2018

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

There are uncertainties involved in the risk assessment of process systems operations. Also, systems are complex and deteriorate gradually with time or due to exposure to expected or unexpected disturbances and events. Questions such as "what is the frequency of a process upset?" and "can we predict incidents?" have yet to be explored and answered. With the use of the Process Resilience Analysis Framework, this work presents a resilience-based approach to managing uncertainties to better predict process upsets. Prior specification on uncertain parameters is assumed based on historical data. Popular sampling and Bayesian techniques such as Markov chain Monte Carlo simulation and mixture modeling are used for posterior inference on the parameters. The application of the predictability assessment for uncertainty management is demonstrated using a polyvinyl chloride process system. A total of three types of uncertainties (cooling medium temperature, agitator failure, and reactants charging) are considered. It is concluded that with the use of resilience metrics data, the variance of statistical parameters can be updated, leading to high-probability regions of the parameter space responsible for the observed data. This helps the risk assessors make more-accurate and informed process risk decisions.

show abstract

Section: Predictability Assessment Methodologymentioning

confidence: 99%

Section: Predictability Assessment Methodologymentioning

confidence: 99%

Resilience-Based Process Upset Event Prediction Analysis for Uncertainty Management Using Bayesian Deep Learning: Application to a Polyvinyl Chloride Process System

Jain

Chakraborty

Pistikopoulos

et al. 2018

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

show abstract

“…This provides a dynamic modeling of the safety systems. According to Pariyani et al (2010), dynamic data collected from distributed control systems (DCS) and emergency shutdown (ESD) systems are used as ASP data for dynamic risk analysis of an industrial fluid catalytic cracking unit (FCCU). Pariyani et al (2012a,b) used the dynamic failure analysis methodology by using the data recorded in ESD and DCS to increase process safety and product quality.…”

Section: Dra Methodologymentioning

confidence: 99%

“…The near-miss region was identified as the bottom part of the safety pyramid-containing events with no alarming consequences but having the potential to escalate to an accident. Oktem et al (2010) broadened the definition by introducing events such as unobserved problems and unsafe conditions as near-miss situations. Applications of near-miss management in various industries such as CPI and in financial sectors have been illustrated by Kleindorfer et al (2012).…”

Section: Asp Datamentioning

confidence: 99%

Risk and reliability assessment in chemical process industries using Bayesian methods

Roy

Srivastava

Sinha

2014

Reviews in Chemical Engineering

View full text Add to dashboard Cite

Chemical process industries (CPI) are usually home to a large number of complex systems and components required for various operations involving hazardous chemicals. The intense operating conditions and complex interactions between the systems make the chemical plants vulnerable to accidents. Quotidian incidents and mishaps can lead to catastrophic incidents. Thus, the area of risk assessment and reliability analysis in CPI has been of much interest to the research community. The complexity of processes in CPI demands a risk assessment tool that can adapt itself to the dynamic environment and can efficiently model the functional and sequential dependencies between the components and the effects of external factors, component degradation, and variation in operating conditions. The risk assessment tool must have applicability during the operational lifetime of the system to serve as a platform for decision making and risk management. The unavailability of empirical data for some variables is another pertinent issue in risk analysis and reliability assessment in CPI. Analysts often have to work with subjective information such as expert opinion. Bayesian statistical methods based on the Bayes theorem are considered by many to be an effective tool to address the above-mentioned issues. These methods are based on the subjective interpretation of probability that helps to model the epistemic uncertainties and easily propagates them through complex system models. The methods provide a formal systematic way to incorporate subjective information into calculations. The inherent updating property accoutres these methods with the ability to deal with real-time changes. The opponents, however, point out that the methods produce high overconfidence and randomness in computed answers. In the last two decades, an increased interest can be seen in the research community toward the use of Bayesian methods in risk assessment. This paper presents a comprehensive literature review of the application of Bayesian methods through Bayesian parameter estimation techniques and Bayesian updating procedures in process industries. Both these techniques have been extensively used in various aspects of risk analysis, which are very pertinent in CPI. The purpose of the study is to produce an effective reference guide for scholars interested in applying Bayesian techniques to risk and reliability assessment in CPI. AbbreviationsAI artificial intelligence ASP accident sequence precursor BN Bayesian network CCF common cause failure CPI chemical process industry CPT conditional probability table CTBN continuous time Bayesian network DAG directed acyclic graph DBN dynamic Bayesian network DCS distributed control system DDT diagnostic decision tree DIF diagnostic importance factor DOOBN dynamic object-oriented Bayesian network DRA dynamic risk assessment DTBN discrete time Bayesian network EB empirical Bayes ESD emergency shutdown EVT extreme value theorem ET event tree FB fully Bayes FMEA failure mode effects analysis FFT fast Fourier transform FT fa...

show abstract

“…These analyses allow one to detect and quantify risk‐prone spots within a processing plant and then mitigate or eliminate risks to the plant . Tools such as support vector machines, causal dependency, fuzzy logic, event trees, filter‐based methods, improved kernel component analysis, and Bayesian networks (BNs) have been successfully applied to conduct probabilistic inference, sensitivity analysis, and detection and isolation of most probable causes of abnormal events. Methods have also been developed for fault detection and isolation under nonlinear closed‐loop process conditions.…”

Section: Introductionmentioning

confidence: 99%

Maximum‐likelihood maximum‐entropy constrained probability density function estimation for prediction of rare events

Ahooyi¹,

Soroush²,

Arbogast

et al. 2014

AIChE Journal

Self Cite

View full text Add to dashboard Cite

This work addresses the problem of estimating complete probability density functions (PDFs) from historical process data that are incomplete (lack information on rare events), in the framework of Bayesian networks. In particular, this article presents a method of estimating the probabilities of events for which historical process data have no record. The rare-event prediction problem becomes more difficult and interesting, when an accurate first-principles model of the process is not available. To address this problem, a novel method of estimating complete multivariate PDFs is proposed. This method uses the maximum entropy and maximum likelihood principles. It is tested on mathematical and process examples, and the application and satisfactory performance of the method in risk assessment and fault detection are shown. Also, the proposed method is compared with a few copula methods and a nonparametric kernel method, in terms of performance, flexibility, interpretability, and rate of convergence.

show abstract

Improving Process Safety and Product Quality using Large Databases

Cited by 8 publications

References 14 publications

Resilience-Based Process Upset Event Prediction Analysis for Uncertainty Management Using Bayesian Deep Learning: Application to a Polyvinyl Chloride Process System

Resilience-Based Process Upset Event Prediction Analysis for Uncertainty Management Using Bayesian Deep Learning: Application to a Polyvinyl Chloride Process System

Risk and reliability assessment in chemical process industries using Bayesian methods

Maximum‐likelihood maximum‐entropy constrained probability density function estimation for prediction of rare events

Contact Info

Product

Resources

About