A modeling framework for assessing the impact of new time/mileage warranty limits on the number and cost of automotive warranty claims

Abstract An automobile with over 7000 parts is a highly complex product. In spite of employing the best quality and reliability practices during product development, manufacturing, and assembly, unexpected failures during warranty period do occur and cost automobile companies billions of dollars annually in warranty alone. Warranty coverage for an automobile is generally stated in terms of mileage (in miles) and time (in months or years). The coverage expires when any of the two limits is crossed. Any change in warranty coverage too, influences warranty cost significantly. However, changes made to warranty coverage are often market driven. In either case, a company needs to plan for maintaining a large cash reserve to pay for the warranty services on their products. In this paper, we present a simple method to assess the impact of new time/mileage warranty limits on the number and cost of warranty claims for components/sub-systems of a new product. We highlight the use of mileage accumulation rates of a population of vehicles to arrive at claims per thousand vehicles, sold with new time/mileage warranty limits. We also discuss the bias in warranty cost estimates that may result in using cumulative cost per repair information. We recommend the use of incremental cost per repair especially when populations with different mileage accumulation rates are under consideration. Application examples are included to illustrate the use of the proposed methodology.

[1]  Kazuyuki Suzuki,et al.  Estimation of Lifetime Parameters From Incomplete Field Data , 1985 .

[2]  Radu Mihaela Elena,et al.  Reliability and life testing handbook Volume 1: Author: Dimitri Kececioglu Publisher: Prentice Hall, Inc. Eaglewood Clifss, New Jersey 07632 xlii+917 pages, includes bibliographical references and index 1993 Price: $79.50 (ISBN 0-13-772377-6) , 1993 .

[3]  J Lawless,et al.  Methods for the estimation of failure distributions and rates from automobile warranty data , 1995, Lifetime data analysis.

[4]  Kazuyuki Suzuki Nonparametric Estimation of Lifetime Distributions from a Record of Failures and Follow-Ups , 1985 .

[5]  J. D. Kalbfleisch,et al.  Some useful statistical methods for truncated data , 1992 .

[6]  Vijayan N. Nair,et al.  Estimation of reliability in field-performance studies , 1988 .

[7]  Nanua Singh,et al.  Hazard rate estimation from incomplete and unclean warranty data , 2003, Reliab. Eng. Syst. Saf..

[8]  Wayne Nelson,et al.  Confidence Limits for Recurrence Data—Applied to Cost or Number of Product Repairs , 1995 .

[9]  Jerald F. Lawless,et al.  Nonparametric estimation of a lifetime distribution when censoring times are missing , 1998 .

[10]  Ratna Babu Chinnam,et al.  A fuzzy logic based approach to reliability improvement estimation during product development , 2003, Reliab. Eng. Syst. Saf..

[11]  Jerald F. Lawless,et al.  Statistical Models and Methods for Lifetime Data. , 1983 .

[12]  Wayne Nelson,et al.  Applied life data analysis , 1983 .

[13]  Jerald F. Lawless,et al.  Estimation of Rate and Mean Functions from Truncated Recurrent Event Data , 1996 .

[14]  W. Nelson Statistical Methods for Reliability Data , 1998 .

[15]  D. Murthy,et al.  New product warranty: A literature review , 2002 .

[16]  Wayne Nelson,et al.  Graphical Analysis of System Repair Data , 1988 .

[17]  Marcos Esterman,et al.  Reinventing warranty at HP: an engineering approach to warranty , 2003 .