Early detection of failures from vehicle equipment data using K-means clustering design

Through data mining from vehicle construction equipment, the paper presents a unique approach to the early diagnosis of machinery failures. We used an effective K-means clustering algorithm to discover inconsistencies in field data. Data is stored in a warehouse using the Structural Query Language (SQL). The vehicle equipment data is processed using the K-means algorithm. During K-means clustering, we applied the Waikato Environment for Knowledge Analysis (WEKA) application dataset to represent the correct information, which enables risk detection based on the stored warehouse. The K-means clustering performs early defects identification in a dataset of construction equipment supplied by a Libyan organization. The clustering mechanism is based on the Euclidean distance, which is used to calculate the correctness of files in a large dataset. The tests make use of real-world equipment datasheets spanning five years. The results show that the K-means algorithm produces more accurate results than other methods.