Premium
A poisson regression model for two‐attribute warranty policies
Author(s) -
Moskowitz Herbert,
Chun Young Hak
Publication year - 1994
Publication title -
naval research logistics (nrl)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.665
H-Index - 68
eISSN - 1520-6750
pISSN - 0894-069X
DOI - 10.1002/1520-6750(199404)41:3<355::aid-nav3220410305>3.0.co;2-d
Subject(s) - warranty , product (mathematics) , poisson regression , poisson distribution , computer science , plan (archaeology) , set (abstract data type) , operations research , actuarial science , business , mathematics , statistics , population , geometry , demography , archaeology , sociology , political science , law , history , programming language
We discuss the two‐attribute warranty policy in which two types of warranty characteristics, such as product age and mileage in the y‐year‐ m ‐mile protection plan in the automobile industry, are employed simultaneously as criteria in determining the warranty eligibility of a failed product. Based on the Poisson regression model and von Neumann‐Morgenstern expected utility theory, we develop two kinds of two‐attribute warranty policies: a fixed two‐attribute warranty policy in which all customers are offered the same warranty plan, and a flexible two‐attribute warranty policy in which customers are permitted to choose any warranty plan among the numerous plans furnished by the producer. A numerical example is presented to illustrate the procedure for determining the optimal warranty price in the fixed two‐attribute warranty model and for generating a set of warranty plans for the same warranty price in the flexible warranty model. It appears desirable from both the producers' and consumers' viewpoints to convert the classical single‐attribute warranty policy into a fixed two‐attribute policy or, even further, into a flexible two‐attribute policy. © 1994 John Wiley & Sons. Inc.