The CubeSat Or Microsat Probabilistic and Analogies Cost Tool, or COMPACT, is a NASA Headquarters funded effort to fill the gap in cost estimating capabilities for CubeSats, as well as other microsat spacecraft. The COMPACT team has focused mainly on CubeSats to date, and has collected techni-cal, programmatic and cost data on dozens of flown CubeSats missions led by NASA, research labs, and universities. In late 2019, the team released the first tool prototype which uses a non-parametric regression technique, k-Nearest Neighbors (KNN), on actual data from historical CubeSat missions to produce early ballpark analogy-based cost estimates for new CubeSat concepts. Since the KNN prototype was first released, the COMPACT team has normalized 17 new missions to be added to the model in COMPACT V2. COMPACT V2 also features changes to the KNN tool algorithm including the introduction of Principal Component Analysis (PCA) to the model development process and changes to the input parameters which have made the analogy results more intuitive and have improved model performance. This paper describes the current COMPACT KNN dataset, improvements made to the model in COMPACT V2, an assessment of current model performance, and a forward look at COMPACT's planned future enhancements.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    COMPACT KNN V2: Analogy-Based Cost Estimation Model for CubeSats


    Contributors:


    Publication date :

    2022-03-05


    Size :

    1223816 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    COMPACT KNN: developing an analogy-based cost estimation model for CubeSats

    Johson, James / Lumnah, Alex / Hooke, Melissa et al. | NTRS | 2020



    DANCING CUBESATS

    Debra Werner | Online Contents | 2017


    Design of a Compact Telephoto Objective for 6U CubeSats

    Yazgan, Betul Konakli / Ekinci, Mustafa / Bingul, Ahmet et al. | IEEE | 2023