The Naive Bayesian Network (NBN) classifier is an optimal classifier(in the sense of minimal classification error rate) in the case of independent descriptors or variables. The presence of dependencies between variables generally reduce his efficiency. In this article, we are proposing a new classification method named Naive Bayesian Network in the Space of Discriminants Factors (NBNSDF) which is based on the use of the NBN in the space of discriminants factors issue from a discriminant analysis. The discriminants factors are not correlated letting very efficient the use of the NBN. We found on simulated data that the NBNSDF method better detects and isolates faults in multivariate processes than the NBN in the case of strongly correlated variables.
Customer complaints data are usually expressed as counts for a period of time and are governed by a Poisson process. This process is stationary when the number of complaints is constant, while a change in these numbers would indicate a potential change in the product performance. In this paper we describe an approach for establishing the maximum tolerance level for the number of complaints received within a month. Tolerance level is based on a relatively stable period of time when the Poisson process is stationary. A change-point analysis is performed to the complaints data that exhibit large changes to partition the relatively stable period from the problematic period. Examples that illustrate this approach are provided.
Customer satisfaction as the key element for success in business is a major concern for any industry. In this paper we propose a customer satisfaction index using principal component analysis for a software solution company. This index was used as an input to the marketing division to identify their potential customers from their past experience. Since this is a very common problem for any industry, the same approach can be used in similar situations.
The continued rapid worldwide diffusion of clinical hyperbaric facilities has substantially increased interest in clinical quality assessment and service improvement. This paper examines major issues, perspectives, and methods integral to the measurement and improvement of the quality of care provided to hyperbaric patients and their relevance and applicability across different societies. Special focus is directed toward the importance of quality assessment and improvement of clinical hyperbaric care, multiple stakeholder perspectives on improved clinical quality, measurement of clinical outcomes of hyperbaric care, importance of facility accreditation, process improvement methods, and the future importance of quality management in clinical hyperbaric facilities.
Peer Review is commonly applied during the final stage of a project and is used in the science community to determine the soundness of a conclusion and if the quality of information meets the standards of the scientific and technical community. Quality Assurance and Quality Control procedures are most effective when applied to planning, then during the process stage of a project, and during the review of a product. The gold standard for validating the quality of a study's results is comparison with a referenced data set, model, or result (Azouzi, 1999). However, costs can severely limits this approach for many applications. This article proposes an additional means for quality assessment using peer review as a quality assessment tool during the planning, process, and review of a project.
The process of peer review for submissions to scientific journals is a well-established and widely used procedure. However, there may still be room for improvement in the procedural aspects of peer review. Advantages and disadvantages of the current system, against the introduction of systems using either no anonymity, or double anonymity, are assessed. Recommendations to improve the robustness and fairness of the peer review process are proffered for the reader's consideration.
Electronic recordkeeping is increasingly replacing handwritten records in the course of "normal business." As this trend continues, it is important that organizations develop and implement electronic recordkeeping policies and procedures. This is especially true for research and development organizations because of the potential to transform a discovery into a patent, and at times patent application contests are resolved in litigation. This paper provides a basis for the development, implementation, and subsequent assessment of a research and development recordkeeping policy. The approach described in this paper should be tailored by the organization adopting this approach to meet the needs of their organization.
The objective of this paper is to assess the impact of the new ISO 9001:2000 standard on the Quality Management Systems (QMS) of Lebanese firms that were already certified under ISO 9000:1994. To get an accurate feedback of stakeholders in a firm, three different questionnaires were developed and distributed to management, employees, and customers respectively. Empirical results indicate that ISO 9001:2000 has improved the QMS performance of Lebanese firms over that of the 1994 edition. Nevertheless, results showed that the new standard still has some weaknesses when it comes to improving suppliers' relationships and empowering employees.
Tabled sampling schemes such as MIL-STD-105D offer limited flexibility to quality control engineers in designing sampling plans to meet specific needs. We describe a closed form solution to determine the AQL indexed single sampling plan using an artificial neural network (ANN). To determine the sample size and the acceptance number, feed-forward neural networks with sigmoid neural function are trained by a back propagation algorithm for normal, tightened, and reduced inspections. From these trained ANNs, the relevant weight and bias values are obtained. The closed form solutions to determine the sampling plans are obtained using these values. Numerical examples are provided for using these closed form solutions to determine sampling plans for normal, tightened, and reduced inspections. The proposed method does not involve table look-ups or complex calculations. Sampling plan can be determined by using this method, for any required acceptable quality level and lot size. Suggestions are provided to duplicate this idea for applying to other standard sampling table schemes.
Multimedia data from two probability-based exposure studies were investigated in terms of how censoring of nondetects affected estimation of population parameters and associations. Appropriate methods for handling censored below-detection-limit(BDL)values in this context were unclear since sampling weights were involved and since bivariate associations/measures were of interest. Both simple substitution(e.g., using 1/2 or 2/3 of the detection limit(DL)for BDL values)and truncation-based strategies were investigated by creating some artificial DLs and comparing resultant estimates with the original studies'uncensored results. The substitution methods generally outperformed the truncation methods, with the(2/3)DL substitution generally performing best.
Pharmaceutical companies, over a period of time, have attempted to use innovative and modern technologies for quicker and more efficient methods of clinical data capture and analysis. In today's scenario, Electronic Data Capture (EDC) is considered to be the preferred technology that can provide significant benefits over existing manual methods. This article highlights the lacunae of the traditional data capture method and discusses the advantages of using EDC for better data quality, improved performance and productivity, and reduced cost in clinical trial management. It also emphasizes the need for IT infrastructure, training, and 21 CFR Part 11 compliance issues. The authors have also described the challenges to be faced by the investigators and sponsors in implementing EDC. Finally, the article concludes emphasizing the fact that EDC is the future mantra for the clinical trials and all stake holders should face challenges of infrastructure, technology, regulations, and training to make it a success.
Monitoring the emissions flux of contaminant gases from large area sources requires measurement of concentrations from an optical remote sensing device and reconstruction of the plume. Path integrated concentrations are determined using multiple optical beam paths. The spatial distribution of concentrations is generated for a plane perpendicular to the direction of the wind. Estimates of the emission flux are determined by integrating the product of the calculated concentrations and wind speeds over the plane. No standard method exists for the complete process, defensible estimates of the uncertainty of the final emission flux have not yet been developed, and a data validation procedure is needed. Auditors are challenged to configure an adequate performance evaluation standard that is representative of a large area source.
The Internet continues to provide an excellent resource for information on quality assurance concepts, regulations, and practices. A search using just the word "quality" produced over 42 million hits. The combination of "quality" and "assurance" yielded over 2 million hits. Presented here is a sampling of 100 quality assurance sites organized alphabetically by site name, and accompanied by a brief description of the information available at the site. The choice of which sites to include was based on the author's experience and familiarity with the QA profession, and was aimed towards providing examples in active areas of QA including business and manufacturing, good practice regulations (i.e., GxPs), information quality, medical practice, software quality, higher education, and quality of research. The 100 sites provide access to a broad array of documents, services, forums, and opportunities to exchange ideas, and include links to major national regulatory and standard setting bodies around the world.
This paper outlines the development of a CD-ROM training package entitled: The WHO Basic Training Modules on GMP, intended to support the creation of training courses aimed particularly at government compliance officials who inspect pharmaceutical manufacturing facilities. The material was created over a three-year period in collaboration with a team of external experts, WHO regional and local offices, and Drug Regulatory Authorities of participating countries. The nine training workshops and courses that contributed to the development and evaluation processes were attended by approximately 240 participants from 47 countries. To date over 5,800 copies of the CD-ROM have been distributed.
There has been a significant increase in the number of clinical drug trials (particularly phase III) being conducted in developing countries for infectious diseases such as HIV, malaria, and tuberculosis. Laboratory results provided by medical testing laboratories in the region are critical to ensuring the safety of patients and the generation of good quality data. A number of well accepted Good Clinical Practice (GCP) and Good Laboratory Practice (GLP) guidelines govern the conduct of clinical trials internationally. Good Clinical Practice guidelines remain too vague with respect to sample analysis to ensure practical implementation in these laboratories. In their strictest sense, Good Laboratory Practice guidelines refer to the analysis of samples from non-clinical studies. A specific set of minimum standards or requirements for practical implementation of clinical trial requirements in medical testing laboratories in the developing world is urgently required.
This article presents a simple, semi-prescriptive self-assessment model for use in industry as part of a continuous improvement program such as Total Quality Management (TQM). The process by which the model was constructed started with a review of the available literature in order to research TQM success factors. Next, postal surveys were conducted by sending questionnaires to the winning organisations of the Baldrige and European Quality Awards and to a preselected group of enterprising UK organisations. From the analysis of this data, the self-assessment model was constructed to help organisations in their quest for excellence. This work confirmed the findings from the literature, that there are key factors that contribute to the successful implementation of TQM and these have different levels of importance. These key factors, in order of importance, are: effective leadership, the impact of other quality-related programs, measurement systems, organisational culture, education and training, the use of teams, efficient communications, active empowerment of the workforce, and a systems infrastructure to support the business and customer-focused processes. This analysis, in turn, enabled the design of a self-assessment model that can be applied within any business setting. Further work should include the testing and review of this model to ascertain its suitability and effectiveness within industry today.
All scientific disciplines rely to some degree upon existing data to design new studies, test hypotheses, and make decisions. Because existing data can take many forms, a framework for addressing the quality of these data must be general and comprehensive. By nature of this inclusiveness, quality categories for existing data are necessarily broad. Effective employment of existing data requires the development of specific acceptance criteria from broad data quality categories. A framework is presented for collecting and evaluating existing data with examples of Environmental Protection Agency projects employing a tiered data review. The systematic planning inherent in a tiered review is described and attendant data quality considerations are developed in the context of an ecological risk assessment; specifically, the process is illustrated by defining an ecologically protective concentration of a chemical in soil.
This paper assists systematic planning for research projects. It presents planning concepts in terms that have some utility for researchers. For example, measurement quality objectives are more familiar to researchers than data quality objectives because these quality criteria are more closely associated with the measurement systems being used. Because of the diverse nature of research, it is not possible to describe cookbook-style planning procedures to be used in all cases. Instead, several general concepts and techniques are presented and researchers can choose those techniques that best fit their specific projects. Examples are presented to illustrate the techniques.
Multimedia data from two probability-based exposure studies were investigated in terms of how missing data and measurement-error imprecision affected estimation of population parameters and associations. Missing data resulted mainly from individuals'refusing to participate in certain measurement activities, rather than from field or laboratory problems; it suggests that future studies should focus on methods for maximizing participation rates. Measurement error variances computed from duplicate-sample data were small relative to the inherent variation in the populations; consequently, adjustments in nonparametric percentile estimates to account for measurement imprecision were small. Methods of adjustment based on lognormality assumptions, however, appeared to perform poorly.
Laboratory Information Management Systems (LIMS) play a key role in the pharmaceutical industry. Thorough and accurate validation of such systems is critical and is a regulatory requirement. LIMS user acceptance testing is one aspect of this testing and enables the user to make a decision to accept or reject implementation of the system. This paper discusses key elements in facilitating the development and execution of a LIMS User Acceptance Test Plan (UATP).
