Abstract

The study of Flight has anomaly begun 21 years ago. The focus of this research is to automatically resolve conflicts on flights through the flight route. The deciding factor to analyse flight phenomena is a flight data sourced from Automatic Dependent Surveillance-Broadcast (ADS-B) data sets. The main parameters to be referenced position (latitude, longitude); velocity (speed); and traveling time. This study aims to detect flight anomalies based on the call sign in a waypoint area. The proposed method is BSGVD, ie a) Building Segment based on waypoint range; b) Grouping object based on clustering; c) Determining the computing aspect of the clustering; d) Cluster Validity based on index value; and e) Distance measurement on cluster centroid for funding to find a potentially anomalous area. The results achieved in this study were to generate potential anomaly areas in the ADS-B data in the segment region. It is dominantly influenced by time parameters. Because, based on the analysis of the distance between the four aspects, obtained the longest distance is in the aspect (latitude, longitude, traveling time) with distance = 182.01 and aspects (speed, traveling time) with distance = 182.32. Based on the index value criteria, we get the aspect (speed, travel time) that has the cluster analysis. Then, the final are measurement validity results obtained: Dunn index value = 0.645; Silhouette index value = 0.89; and Davies-Bouldin index value = 14.81.

Original languageEnglish
Pages (from-to)516-526
Number of pages11
JournalProcedia Computer Science
Volume161
DOIs
Publication statusPublished - 2019
Event5th Information Systems International Conference, ISICO 2019 - Surabaya, Indonesia
Duration: 23 Jul 201924 Jul 2019

Keywords

  • Four aspects of anomaly detection
  • Internal validity cluster
  • Segment

Fingerprint

Dive into the research topics of 'Cluster phenomenon to determine anomaly detection of flight route'. Together they form a unique fingerprint.

Cite this