1,721,023 research outputs found
Classificazione statistica di frammenti archeologici: la ceramica a vernice nera
No full textL’utilizzo di un sistema di classificazione, fornendo una base di confronto tra singoli oggetti, classi di oggetti o siti, agevola notevolmente il lavoro svolto dagli archeologi e consente di divulgare con facilità i risultati delle ricerche. In questo lavoro si utilizza una tecnica di segmentazione binaria per individuare una “regola di classificazione” dei reperti archeologici, con l’obiettivo di fornire un utile strumento decisionale di supporto all’attività insostituibile svolta dall’archeologo. I dati analizzati sono relativi ad una classe di materiali ampiamente studiata e classificata manualmente: frammenti di ceramica a vernice nera di produzione pestana provenienti dal santuario di Hera alla foce del Sele. La trattazione è condotta in modo da fornire un quadro metodologico della segmentazione binaria facendo riferimento ad uno degli algoritmi maggiormente utilizzati: il CAR
QIC Analysis: a practical tool to manage continuous quality improvement
No full textContinuous quality improvement (CQI) has emerged as an imperative for businesses’ growth in today’s fiercely competitive environment. CQI is a never-ending process that seeks to achieve defect-free, high quality products or services. Different approaches to quality management, like TQM, Business Excellence models and ISO standards, agree on the critical importance of the “continuous improvement” principle for the success of any strategy focused on quality. Notwithstanding this acknowledgement, a review of the extensive literature on quality management reveals that scarce effort is addressed to develop quantitative methods to analyze the quality improvement (QI) process. In fact, the literature on CQI consists mainly of case studies, anecdotal evidence and the prescriptive measures attributed to the recognized experts in the field of quality. Some popular prescriptions of the quality experts are the “14 points” of Deming (1982), the “10 steps” of Juran (1962) and the “14 steps” of Crosby (1979). There is no consensus as to which prescriptive framework should form the basis for CQI and the result is a lack of a practical model that is useful for the monitoring of a firm’s CQI program (Prybutok and Ramasesh, 2005). Since CQI is an ongoing process, it is imperative that firms monitor the CQI program on a regular basis to ensure that it is working well and to continually identify areas for QI. In order to effectively monitor CQI reliable and valid instruments are necessary. The original approach proposed in this paper aims at providing the organization with diagnostic tools enabling it to understand if it is on the right track to the planned QI and to evaluate the walk already covered as well as the residual distance to go up in its journey for QI. The proposed approach relies on the modelling of the quality growth which is obtained by 1) choosing a proper quality index to monitor over time and 2) identifying the Quality Improvement Curve (QIC) that describes the quality growth due to the CQI process. The QIC analysis can be usefully applied to monitor a single QI process over time or to compare several QI processes by means of a suitable indicator. The methodology is illustrated through a case study concerning the QI process of a teaching course in Probability and Statistics held at the Faculty of Engineering of the University of Naples Federico II. Starting from experimental data collected, during the second semester of the academic year 2005-06, by interviewing the attending students about the quality of the course of Probability and Statistics, three QI programs have been hypothesized. The values of the quality index for future semesters (in which QI programs are assumed to be adopted) have been obtained by Monte Carlo simulation
A critical review and further advances in the innovation growth models
No full textIn recent decades, the literature of technology management has proposed the S curves as promising tools for the analysis of the life cycle of technological innovations in order to support the company’s strategies and policies. In this work, the main mathematical and statistical characteristics of the most popular S curve models are analyzed in order to verify their suitability to model technological innovation processes. In particular, the linearity properties of each model have been studied since they are needed to make some useful inferences. The critical comparative analysis has been carried out exploiting some real data sets concerning three different technologies: piston aero-engines, jet aero-engines and digital signal processors, respectively
Valutazione della qualità dei servizi. Applicazione di tecniche statistiche di campionamento e ri-campionamento ad un servizio di pubblica utilità
No full textManaging the technological innovation process via a flexible S curve model
No full textInnovativeness determines, to a large extent, the firm competition level. In order to properly manage technological processes, the company strategists need models for the analysis of the technological performance trend. In recent decades, the literature of technology management has proposed the S curve models as useful tools for the analysis of the life cycle of technological innovations in order to plan the company strategies and policies. As for many other models developed in the field of management, the scarce attention devoted to the analytical foundations has reduced the theoretical consistency of the S curves to model the technological innovation process, making it difficult to use these models as real strategic tools. In this work a flexible S curve model is examined in order to highlight the model suitability to analyse the technological innovation process. Specifically, the linearity properties of the model have been studied and a suitable re-parameterization has been proposed. In order to enable the management to strategically use the S curve model, approximate prediction intervals about the technological innovation level have been built to monitor the evolution of the performance of a given technology. The theoretical considerations are supported by the practical application of the proposed methodology to three data sets concerning three different technologies: piston aero-engines, jet-aero-engines and digital signal processors, respectively
Checking quality of sensory data by assessing intra/inter panelist agreement
No full textThis study aims at checking the quality of sensory data by evaluating and testing both panelist precision and panel reproducibility via an agreement index-based approach which has been already adopted for the assessment of rater reliability but is almost unexplored in the field of sensory analysis. The approach has been applied to a case study concerning the assessment of sensory characteristics induced by different food and beverages
- …
