Abstract In this paper, an improved approach for space mission risk classification based on text mining is proposed. The method consists of the text mining of extracted bibliographic data with an improved accuracy compared to previous works. Based on these data, a weight is attributed to each of the identified risk parameters and a total mission class score is calculated. A quantification of the interrelation of these parameters, with the use of the t-score metric, together with the appropriate visualization is also provided for further analysis. The proposed method serves as an assistant tool to the appointed panel of experts that is traditionally tasked with the assignment of weights to the parameters. It allows for the consideration of parameters of widely different importance, even by almost an order of magnitude. Eight classification parameters are assessed including: (a) Criticality to the Agency, (b) Objective importance, (c) Cost, (d) Lifetime, (e) Complexity, (f) Manned mission, (g) Destination, and (h) Omega Factor. The most important parameter is shown to be the Complexity of the mission contributing around one fourth of the total weight. Finally, the NASA Mars Perseverance mission is used to demonstrate the application of the classification algorithm.
Highlights Improved approach for space mission risk classification based on text mining. Weight calculation for risk parameters and space mission risk classification. Visualization of risk parameter interrelation based on t-score metric. Automated assistant tool to experts for space mission risk assessment. Risk classification of NASA Mars Perseverance mission.
An improved text mining-based space mission risk classification approach
Acta Astronautica ; 207 ; 353-360
2023-03-21
8 pages
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch