Stefano Barone

Variation Mode and Effect Analysis: a Practical Tool for Quality Improvement

This paper describes a statistically based engineering method, variation mode and effect analysis (VMEA), that facilitates an understanding of variation and highlights the product/process areas in which improvement efforts should be targeted. An industrial application is also described to illustrate how the VMEA can be used for quality improvement purposes. Copyright

A weighted logistic regression for conjoint analysis and Kansei engineering

Customer needs for emotional satisfaction are increasingly being considered by product and service designers. While several existing methods such as conjoint analysis (CA), Kano model and quality function deployment support the translation of customer requirements into technical specifications, researchers are now working to develop methods aimed at integrating affective aspects into product design. Kansei engineering (KE) is a design philosophy that considers customer perceptions and emotions by adopting a multi-disciplinary approach. CA is a useful tool within a KE project. This article presents a methodology for conducting a KE project in early development phases. This methodology is based on two new procedures. The first one is aimed at calculating attribute importance weights by using respondent choice time in controlled interviews. The second procedure allows the exploitation of such weights in an ordinal logistic regression model for analysing the results of CA experiments. By using the proposed methodology, it is possible to identify product/service attributes able to induce specific emotions and feelings in customers and consequently choose the right development strategy. An application of the method for the design of mobile phones is presented. Copyright

A Robustness Approach to Reliability

Reliability of products is here regarded with respect to failure avoidance rather than probability of failure. To avoid failures, we emphasize variation and suggest some powerful tools for handling failures due to variation. Thus, instead of technical calculation of probabilities from data that usually are too weak for correct results, we emphasize the statistical thinking that puts the designers focus on the critical product functions. Making the design insensitive to unavoidable variation is called robust design and is handled by (i) identification and classification of variation, (ii) design of experiments to find robust solutions, and (iii) statistically based estimations of proper safety margins. Extensions of the classical failure mode and effect analysis (FMEA) are presented. The first extension consists of identifying failure modes caused by variation in the traditional bottom–up FMEA analysis. The second variation mode and effect analysis (VMEA) is a top–down analysis, taking the product characteristics as a starting point and analyzing how sensitive these characteristics are to variation. In cases when there is sufficient detailed information of potential failure causes, the VMEA can be applied in its most advanced mode, the probabilistic VMEA. Variation is then measured as statistical standard deviations, and sensitivities are measured as partial derivatives. This method gives the opportunity to dimension tolerances and safety margins to avoid failures caused by both unavoidable variation and lack of knowledge regarding failure processes.

TESF Methodology for Statistics Education Improvement.

The need for universities to achieve excellence in the services they provide has been the subject of research for several decades. The idea of involving students and recognizing the importance of their opinions has led to the creation of various models and tools. This paper focuses on teaching, a central service from which improvement actions of an academic institution should always begin. The article reviews and updates the previously developed Teaching Experiments and Student Feedback methodology. The methodology, which is primarily addressed to statistics teachers, allows practical aspects to be organized and decisions to be made based on data that has been collected from students and scientifically analyzed. The steps for building a student satisfaction index are also described. This index, in its most complete version, takes into account possible correlations between importance of the evaluated aspect and scores, both of which are provided by the students. The paper presents an application of the methodology to a statistics course taught by one of the authors.

Design of a university course quality by Teaching Experiments and Student Feedback (TESF)

Quality of the academic system must be constantly monitored, especially under conditions of growing competition and limited resources. Those who are interested in quality, doing research and teaching on it, should ensure quality of the processes for which they are responsible. Teaching is one of the primary services offered by a university. For it, all fundamental principles of Total Quality Management (TQM) must be valid. In this article the authors propose a methodology for designing the quality of a university course, considered as a basic service of a complex educational system. The methodology is based on the concurrent use of teaching experiments performed by a teacher, and the SERVQUAL model for the evaluation of student feedback. Such methodology can be adopted in any educational context. It is a useful tool for continuous course improvement. The article presents the application of the methodology in two consecutive editions of a statistics course at the University of Palermo, Italy.

A statistical monitoring approach for automotive on‐board diagnostic systems

The current generation of vehicle models are increasingly being equipped with on-board diagnostic (OBD) systems aimed at assessing the ‘state of health’ of important anti-pollution subsystems and components. In order to promptly diagnose and fix quality and reliability problems that may potentially affect such complex diagnostic systems, even during advanced development prior to mass production, some vehicle prototypes undergo a testing phase under realistic conditions of use (a mileage accumulation campaign). The aim of this work is to set up a statistical tool for improving the reliability of the OBD system by monitoring its operation during the mileage accumulation campaign of a new vehicle model. A dedicated software program was developed by the authors to filter the large experimental database recorded during the mileage accumulation campaign and to extract the time series of the diagnostic indices to be analysed. A model-based monitoring approach, using continuous time autoregressive (CAR) models for the time-series structure and traditional control charts for the estimated residuals, is adopted. A Kalman recursion procedure for the estimation of the unknown CAR model parameters is described. An application of the proposed approach is presented for a diagnostic index related to the state of health of the oxygen sensor. Copyright

Service Quality Design through a Smark Use of Conjoint Analysis

In the traditional use of conjoint analysis, in order to evaluate the relative importance of several elements composing a service, interviewed customers are asked to express their judgement about different scenarios (specific combinations of elements). In order to reduce the number of possible scenarios, design of experiments methodology is usually exploited. Previous experiences show that, even a limited number of proposed scenarios cause difficulty in answering for the interviewed customer if the scenarios differ for elements of very low interest to him/her. Consequently, a high rate of abandon of the interview has been observed. In this study it is assumed that a service can be decomposed in several improvable elements and/or enriched with new “optionals”. In both cases, what under study is assumed to be a set of dichotomous attributes. For each of these attributes, its marginal contribution to customer satisfaction has to be modelled and estimated. To obtain the required information, an opportune questionnaire is proposed to a sample of interviewed customers. An interviewing procedure consisting in a customer driven design of scenarios is followed, starting from the full‐optional scenario and eliminating one by one the less satisfying elements. for each interviewed customer, a ranking of attributes is so obtained. Then, by asking the interviewed customer to evaluate on a metric scale the scenarios he previously selected, a rating of attributes can also be obtained. A case study conducted in collaboration with a public transportation company is presented. Contrarily to previous experiences, the abandon rate proved extremely reduced.

Analysis of User Needs for the Redesign of a Postural Seat System

The identification and translation of customer needs early in the design process is a major challenge for product design researchers. Some needs are explicit and customers can state them very clearly. Other needs are implicit, so customers cannot express them, e.g., those pertaining to the affective and emotional sphere. In this work, we describe the methods most commonly used to capture explicit and emotional customer needs, and the traditional ways in which they are used. Moreover, an integration of QFD and Kansei engineering, a simplification of Kano methodology, and a new attribute weighing methodology based on the “choice time” are discussed for the design of an innovative postural seat system for patients affected by mental retardation.

Robust Ergonomic Virtual Design

From the early development phases of a new industrial product, realistic simulations can be performed in a virtual environment to study the human-machine interaction. In a virtual lab, it is possible to perform experiments to assess the ergonomics of the new product using mannequins simulating the human body, and to deal with the problem of anthropometric variation.

Statistics‐driven Development of OBD Systems: An Overview

Automotive on-board diagnostic (OBD) systems are designed to keep critical components under control during vehicle functioning, and to alert the driver in case of severe malfunctions. OBD systems aimed at reducing polluting emissions are mandatory on new motor vehicles. Some research projects conducted in cooperation between universities and the automotive industry have been quite successful in terms of knowledge advancement and industrial gain. An updated overview of the adopted methodologies and results obtained are given in this article. Such results can be valuable for both theorists and practitioners, since they witness the use of statistics as a powerful catalyst of technical progress, and give a possible line of action for further application in several scientific fields. Copyright