Modelling an Optimized Warranty Analysis Methodology for Fleet Industry Using Data Mining Clustering Methodologies

• Published in: Procedia computer science 2016, Vol.87, p.240-245
• Main Authors: Srinivasan, R., Manivannan, S., Ethiraj, N., Devi, S. Prasanna, Kiran, S. Vinu
• Format: Article
• Language: English
• Subjects: Data Analysis; Fleet Industry; Manufacturing; Optimization; Predictive Analysis; Warranty
• ISSN: 1877-0509; 1877-0509
• DOI: 10.1016/j.procs.2016.05.155

As the industrial revolution started, the complexity of new products in manufacturing as well as fleet industry has improved to meet the ever increasing needs and expectations of successful business. Degradation of Products due to age and/or operational usage and failures when they are unable to carry out their normal functions. The product had a n-year warranty and these warranty data is available for all applicable units in an organization. Data on essentially all failures was available for the initial level of operation on all units. A large set of data on Warranty among operational units contains useful information about product quality and reliability. They are available as coarse data because most often they are aggregated values, delayed reports, filtered, missing or vague and more importantly erroneous due to human mistakes. They are only forms of warranty data an organization has. Analyzing such data is therefore needed and can also be of benefit to organization and in identifying early warnings of abnormalities in their products, providing useful information about failures, nature of failure modes to aid design modification, finding out product reliability for warranty policy and predicting future warranty claims needed for preparing warranty reserves plans.