Innovations in construction equipment using cognitive and automation techniques, as well as computerized equipment Information Management System (IMS) have greatly simplified the equipment operations and management process. Though the data collection, storage and reporting for equipment management are no longer pressing issues for the company equipment manager, data analysis becomes increasingly difficult with large amounts of data, especially for identifying potential problems in operations and management of a large construction fleet. The proposed decision support system for equipment management uses a resolution-based outlier definition and a nonparametric outlier mining algorithm that can automatically detect inconsistent observations from a large equipment dataset, and rank the records based on their degree of inconsistency. The nonparametric outlier mining algorithm demonstrates ease of use, high flexibility and satisfactory results in construction equipment management.
Bibliographical noteFunding Information:
The research project is partially funded by the Natural Sciences and Engineering Research Council of Canada under Grant # CRD 226956-99.
All Science Journal Classification (ASJC) codes
- Computer Science Applications
- Artificial Intelligence