Analysis of Parts Request Portfolio Using Data Mining in order to Improve Forecasting and Supply the needs of the Defense Logistics System

Abstract

Data mining is a powerful technology that has the ability to discover knowledge hidden in a huge range of data. The purpose of this study is to analyze the portfolio of technical parts by data mining in order to improve the forecast and supply the needs of the defense logistics system. Therefore in the stage of discovering patterns and association in terms of purpose is applied, nature and method is exploratory and library information collection, and in the stage of evaluating the usefulness of patterns in terms of purpose, developmental, the nature and method is descriptive- survey and field information collection. The statistical population for data mining is the database of information technical parts requested by the units of a defense organization for related systems and repair centers over a period of 10 years. The statistical population to measure the relationship between data mining results and forecasting and supply the requirements is expert’s logistics systems. In order to analyze the data for data mining, the cross- industry standard process and SQL Server and Rapid Miner software has been used, and to analyze the data collected through a questionnaire, SPSS software has been used. The results of this study showed that method of discovering association rules can be used to analyze the request portfolio of technical parts and achieve the request pattern, and the patterns and the discovered patterns are used to improve the performance of the logistics system in the field of forecasting and supply the requirements of technical parts

Keywords

• forecasting
• request portfolio
• cross- industry standard process (crisp)
• logistics
• data mining