scholarly journals Functional risk assessment as part of the validation in the implementation of chromatography data system

2018 ◽  
Vol 64 (01) ◽  
pp. 39-56
Author(s):  
Mena Ivanoska ◽  
Hristina Babunovska ◽  
Rumenka Petkovska
Keyword(s):  
Risk Assessment ◽  
Business Process ◽  
Laboratory Data ◽  
Data System ◽  
User Requirements ◽  
Automated Manufacturing ◽  
Validation Process ◽  
System Validation ◽  
Software Risk ◽  
Overall Risk

A Chromatography Data System (CDS) is a complex software that can be configured to the specific needs of the user’s business process. As such it falls into the Good Automated Manufacturing Practice (GAMP) 5 Category 4 – Configured Products. The validation process is planned and follows along the phases proposed by GAMP 5 for configured products. The Risk assessment stage of the CDS validation process is to carry out a risk assessment of each function of the User Requirements Specification (URS) determined on if the function is regulatory risk critical or not. The functional risk assessment is made according to the method- Failure Mode and Effects Analysis (FMEA). The Overall Risk resulting from the Risk Assessment has identified all potential failures requiring mitigating actions/controls. Mitigating actions and testing controls during the PQ phase is implemented. The final Overall Risk after implementation of Mitigating actions and testing controls during the PQ phase is not more than Medium. Keywords: chromatography data system, validation of the CDS Software, risk assessment, laboratory data integrity

scholarly journals Diver Risk Assessment Model Considering Trip Characteristics Using Insurance Data System

2016 ◽  
Vol 161 ◽  
pp. 1160-1165 ◽  
Author(s):  
Majid Soltani ◽  
Seyedehsan Seyedabrishami ◽  
Amirreza Mamdoohi ◽  
Vadood Alyari Kordehdeh
Keyword(s):  
Risk Assessment ◽  
Data System ◽  
Assessment Model ◽  
Risk Assessment Model ◽  
Insurance Data

Risk Analysis of Power-Gen Gas Turbines With GADS Outage Data

2017 ◽  
Author(s):  
Bin Zhou
Keyword(s):  
Risk Assessment ◽  
Risk Analysis ◽  
Gas Turbine ◽  
Gas Turbines ◽  
Risk Mitigation ◽  
Data System ◽  
Loss Prevention ◽  
The Past ◽  
The North ◽  
Effective Assessment

According to FM Global proprietary data, power-gen gas turbine losses have consistently represented a dominant share of the overall equipment-based loss value over the past decade. Effective assessment of loss exposure or risk related to gas turbines has become and will continue to be a critical but challenging task for property insurers and their clients. Such systematic gas turbine risk assessment is a necessary step to develop strategies for turbine risk mitigation and loss prevention. This paper presents a study of outage data from the Generating Availability Data System (GADS) by the North American Electric Reliability Corporation (NERC). The risk of forced outages in turbines was evaluated in terms of outage days and number of outages per unit-year. In order to understand the drivers of the forced outages, the influence of variables including turbine age, capacity, type, loading characteristic, and event cause codes were analyzed by grouping the outage events based on the chosen values (or ranges of values) of these variables. A list of major findings related to the effect of these variables on the risk of forced outage is discussed.

Integration of QbD risk assessment tools and overall risk management

Biologicals ◽  
2016 ◽  
Vol 44 (5) ◽  
pp. 341-351 ◽  
Author(s):  
Brian Kelley ◽  
Mary Cromwell ◽  
Joe Jerkins
Keyword(s):  
Risk Assessment ◽  
Risk Management ◽  
Assessment Tools ◽  
Risk Assessment Tools ◽  
Overall Risk

scholarly journals Missing data imputation in cardiometabolic risk assessment: A solution based on artificial neural networks

2020 ◽  
Vol 17 (2) ◽  
pp. 379-401
Author(s):  
Dunja Vrbaski ◽  
Aleksandar Kupusinac ◽  
Rade Doroslovacki ◽  
Edita Stokic ◽  
Dragan Ivetic
Keyword(s):  
Neural Networks ◽  
Risk Assessment ◽  
Missing Data ◽  
Statistical Power ◽  
Cardiometabolic Risk ◽  
Laboratory Data ◽  
Missing Data Imputation ◽  
Common Solution ◽  
Set Up ◽  
Potential Tool

A common problem when working with medical records is that some measurements are missing. The simplest and the most common solution, especially in machine learning domain, is to exclude records with incomplete data. This approach produces datasets with reduced statistical power and can even lead to biased or erroneous final results. There are, however, many proposed imputing methods for missing data. Although some of them, such as multiple imputation, are mature and well researched, they can be prone to misuse and are not always suitable for building complex frameworks. This paper explores neural networks as a potential tool for imputing univariate missing laboratory data during cardiometabolic risk assessment, comparing it to other simple methods that could be easily set up and used further in building predictive models. We have found that neural networks outperform other algorithms for diverse fraction of missing data and different mechanisms causing their missingness.

Sign in / Sign up

Export Citation Format

Share Document