Tamar Gadrich

Interlaboratory comparison of test results of an ordinal or nominal binary property: analysis of variation

Applications of a new statistical method Ordinal Analysis of Variance (ORDANOVA) for interlaboratory comparisons of measurement or test results of semi-quantitative (ordinal) and qualitative (binary) properties are discussed. ORDANOVA can be helpful for validation of measurement or test methods, proficiency testing of laboratories, development of reference materials with certified semi-quantitative and qualitative properties, that is, probably in every field where ANOVA is applied for quantitative properties. A statistics and criteria are proposed for performance assessment of laboratories active in semi-quantitative and qualitative testing and for other purposes of statistical analysis of such test results.

Evaluating quality measured on a ternary ordinal scale

This paper deals with the evaluation of process/product quality measured via a ternary ordinal scale. Recently, a new approach for dealing with ordinal quality variables, based on constructing a quality ladder using rank and dispersion criterion, was proposed. In this method the quality of a given sample is characterized by its relative position on the constructed quality ladder. The present paper proposes a new mechanism for constructing such a quality ladder. This method is called proportion ratio and dispersion (PR&D). The underlying logic, advantage and applications of the proposed method are discussed and demonstrated using software developed by the authors. Copyright

Engineering of service processes through designing simulation experiments

Service processes encompass a large number of variegated factors. These elements and their interactions have to be considered when engineering a service process. We propose a methodology that allows designers to design simulation experiments through which they can handle various service factors and their interactions in action and thereafter propose process improvements based on a generic analysis scheme. Our methodology deals both with conceptual and detailed designs of service processes and enables the designer to define process factors schematically and simulate possible scenarios based on variations in these factors. We present the application of our methodology to the engineering of a customer order handling process.

Generating and evaluating simulation scenarios to improve emergency department operations

Overcrowding and long patient length of stay, staff shortage, arrival volume increases, and budget constraints are problems hampering ED operations (Sinreich and Marmor, 2005; Maull et al., 2009; NHS, 2010). This paper suggests a framework for schematic generation and evaluation of simulation scenarios to improve ED processes in real-life environments. We illustrate the application of our methodology in a specific ED. We contribute to the area of ED computer simulation by suggesting a methodology that offers the following advantages: (1) Simulation scenarios can be schematically formulated rather than based on trial-and-error experiments. (2) Scenario development can be integrated in the different stages of simulation model development to support designers and management in understanding ED problems, improvement goals, data that should be collected and operational changes that should be applied.

Conceptualizing Emergency Department Scenarios: A DSE-Based Methodology

To handle problems and trends in emergency department (ED) operations, designers and decision makers often simulate and evaluate various case-specific scenarios before testing them in a real-life environment. However, conceptualizing broad possible scenarios for ED operations prior to simulation operationalization is usually neglected. The authors developed a methodology that integrates design of simulation experiments (DSE) as follows: 1) From a literature survey, they culled generic factors whose varying levels determine possible scenarios; 2) the authors drew up a set of generic interactions among these generic factors; 3) a questionnaire was constructed to serve as an instrument to gather the relevant information from management staff about relevant factors, their levels and interactions for a specific ED. Questionnaire responses support a schematic conceptualization of scenarios that should be simulated for a specific ED. They illustrate the application of the authors’ methodology for conceptualization of ED simulation scenarios in two different EDs.

Reliability of Continuous-State Systems in View of Heavy-Tailed Distributed Performance Features

The paper examines the performance distribution for continuous-state systems that may be essentially different from Gaussian. Notable progress in understanding the nature of rare events and the implications for heavy-tailed distributions has been achieved in the last decade. The insights, however, even though they generated robust interest in such distributions, are not reflected strongly enough in reliability engineering practice. The paper presents four simple mechanisms of heavy-tail formation in reliability engineering contexts, their importance and the phenomena where they may appear. It also discusses the implications of this knowledge for optimization of quality controller performance.

A Mechanistic Stochastic Ricker Model: Analytical and Numerical Investigations

The Ricker model is one of the simplest and most widely-used ecological models displaying complex nonlinear dynamics. We study a discrete-time population model, which is derived from simple assumptions concerning individual organisms’ behavior, using the “site-based” approach, developed by Brannstrom, Broomhead, Johansson and Sumpter. In the large-population limit the model converges to the Ricker model, and can thus be considered a mechanistic version of the Ricker model, derived from basic ecological principles, and taking into account the demographic stochasticity inherent to finite populations. We employ several analytical and precise numerical methods to study the model, showing how each approach contributes to understanding the model’s dynamics. Expressing the model as a Markov chain, we employ the concept of quasi-stationary distributions, which are computed numerically, and used to examine the interaction between complex deterministic dynamics and demographic stochasticity, as well as to calculate...

Revealing latent quality information hidden within inspection reports of curtailed tests

The desire to accelerate inspection procedures and reduce inspection costs often results in tests being curtailed. Even though every item is meant to undergo a number of independent/dependent tests, once an item, e.g., item #XXX fails to pass a test, further tests/inspections are terminated. Thus, the empirical data existing at the end of the inspection procedure does not contain information about item #XXX’s ability to pass the cancelled tests. Information mining statistical tools can be used to uncover the latent quality information hidden within these data. This paper proposes altering curtailed testing procedures (e.g., changing the order of the tests or detectability levels), in order to estimate the theoretical joint probabilities (latent quality information) concerning an item’s ability to pass a part of, or the entire sequence of tests. The effectiveness of the proposed procedures is then evaluated using simulated data.