REVI-MINER, a KDD-environment for deviation detection and analysis of warranty and goodwill cost statements in automotive industry

Abstract

REVI-MINER is a KDD-environment which supports the detection and analysis of deviations in warranty and goodwill cost statements. The system was developed within the framework of a cooperation between DaimlerChrysler Research & Technology and Global Service and Parts (GSP) and is based upon the CRISP-DM methodology as a widely accepted process model for the solution of Data Mining problems. Also, we have implemented different approaches based on Machine l.earning and statistics which can be utilized for data cleaning in the preprocessing phase. The Data Mining models applied have been developed by using a statistical deviation detection approach. The tool supports controllers in their task of auditing the authorized repair shops. In this paper we describe the development phases which have led to REVI-MINER.

Decision of Abnormal Quality Unit Lists from Claim Database

“…and inference matters like judgment of risk index. There are many studies conducted in terms of this development: Data mining modeling in the automotive industry [4], claim process modeling of the automobile [3], software cost model for quantifying the gain and loss associated with claims [11], software-based reliability modeling [5].…”

AHP-Based Determination of Warning Grade in a Warranty Claims

“…A warranty is a contractual agreement between a manufacturer (seller) and a consumer (buyer) that requires the manufacturer to rectify all the failures occurring within the warranty period (Jack and Schouten, 2000) [11].…”

“…It also needs artificial intelligence in the assessment of system risk. There are many studies in connection with this problem: data mining modeling in the automotive industry (Hotz et al, 1999) [11]; warranty claims process modeling of the automobile (Hipp and Lindner, 1999) [9] attempted to take a fuzzy theory-like approach that can model ambiguous data in that, in relation to product reliability, colloquially expressed malfunction information and partially described defects frequently appear in the real world. However, the interpretation of the approximate inference result is an issue because the decision-making for warranty degree is free to gauge sensational and multidimensional qualitative information.…”