Data Mining and Artificial Intelligence

Data mining is the art and science of discovering new and interesting information from data that can put your enterprise ahead of the competition. At PAI, we apply data mining and artificial intelligence techniques to find patterns and subtle relationships in your data and infer rules that allow the prediction of future results. In addition to applying data mining techniques in several industrial applications, our scientists are known contributors to the machine learning and data mining research fields. They have developed innovative techniques and pioneered new algorithm as proven by a solid record of publications in scholarly journals.

Our solutions are automated, tailored to your application, and make use of machine learning, statistical analysis and modeling techniques, and database technologies. Apply our knowledge of data mining and artificial intelligence to your next project and see how PAI makes sense of your data to drive improvements and business growth.

Example Projects


In-house Developed Capability

PAI's KnowledgeXplorer is a software environment that provides a wide range of data analysis and mining capabilities by employing state of the art algorithms. It is capable of performing different types of analysis including classification, regression, association, mining and clustering. The results obtained by applying these techniques to corporate data allows better planning of future business activities by giving managers insights hidden in their data.

KnowledgeXplorer is easy to use and is easily customizable to specific needs. It connects efficiently to most popular database systems including IBM DB2, Microsoft SQL Server and Oracle and it is capable of handling large amounts of business data. By including a number of state of the art analysis techniques, KnowledgeXplorer is capable of solving unique problems effectively. The results obtained from KnowledgeXplorer data mining can be easily integrated with other business applications, hence allowing for data driven decision making process.

ADIS Data Mining and Analysis Project

Lockheed Martin

In this project, PAI delivered to Lockheed Martin a data mining tool to assist planning and operations team in their critical decisions related to CSP operations. This tool was developed to mine Lockheed Martin's part failure data from the ADIS (Automated Depot Information System) database with the objective of managing the inventory of spare parts.  

The tool was used to identify bad actors (individual items that have repeatedly cycled through the repair process), and categorized the items in use and in stock based on usage and number of repairs. Based on that information, we analyzed the in-stock items and categorized them as being in an 'under' or 'over' stocked position. These results were used to control the stocking strategies of the parts. In addition, this information helped Lockheed Martin to make decisions on what item to ship out first, whether to eliminate bad actors from stock, and whether to perform a standard repair, no repair, or detailed reliability analysis of a bad actor which comes back for repair.

By relating operational environment, failure rates, internal, and vendor repair performance, Lockheed Martin management was able to determine which factors are having the most impact on part performance and focus attention on those areas providing the greatest opportunity for process improvement. 

Prediction of Corrosion on Orbiter Parts

The Boeing Company

The project involved the development and delivery of a predictive tool for the expected occurrence and severity of flight hardware corrosion.  Based upon an analysis of maintenance data, the tool provided The Boeing Company with the capability to assess future maintenance requirements based upon the expectation of corrosion events. Various machine learning and artificial intelligence algorithms were developed and incorporated in to the delivered tool.

Data Mining Techniques for the Classification and Prediction of Part Failures

Siemens Corporation

One of the critical issues in the successful operation of a power plant involves the accurate identification of the lifetime of parts within a turbine engine (e.g., blades, baskets, etc.). This project involved utilizing our KnowledgeXplorer data mining toolset to identify which sets of factors affect the health of components within turbine engines in order to optimize operational settings for electric generation equipment. Through our analysis of possible variations in operational settings, the client was able to optimize operational life of equipment and reduce maintenance requirements.