We propose a risk index naturally suitable in quality-control framework characterized by data often collected on ordinal scale, to measure the risk of failure of a product. A so-called Severity Index is defined on the basis of the relative frequencies of the ordinal variables. We examine the distribution and the statistical properties of its estimator. We apply the index to real data concerning the severity and the occurrence of defectiveness of the products of a multinational corporation manufacturer. Our index may be employed to communicate the level of risk, to compare among different risks and to identify interventions in the production system.
Facchinetti, S., Osmetti, S. A., A risk index to evaluate the criticality of a product defectiveness, in SIS 2017 Statistics and Data Science: new challenges, new generations, (FIRENZE -- ITA, 28-30 June 2017), Università degli Studi di Firenze, FIRENZE -- ITA 2017: 399-403 [http://hdl.handle.net/10807/111103]
A risk index to evaluate the criticality of a product defectiveness
Facchinetti, Silvia
;Osmetti, Silvia Angela
2017
Abstract
We propose a risk index naturally suitable in quality-control framework characterized by data often collected on ordinal scale, to measure the risk of failure of a product. A so-called Severity Index is defined on the basis of the relative frequencies of the ordinal variables. We examine the distribution and the statistical properties of its estimator. We apply the index to real data concerning the severity and the occurrence of defectiveness of the products of a multinational corporation manufacturer. Our index may be employed to communicate the level of risk, to compare among different risks and to identify interventions in the production system.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.