Within the Mars science community, there is growing interest in the role that small spacecraft missions can play in increasing the breadth and frequency of Mars exploration. Fueled by significant advances in miniaturization of science instruments and avionics, innovative concepts for delivery of small spacecraft to Mars, and emerging low-cost capabilities in the fast-growing NewSpace marketplace, small spacecraft missions appear capable of achieving compelling Mars science results at unprecedentedly low mission cost, far below the current NASA Discovery Program cost cap. We report on the development of a cost model for small spacecraft Mars orbiters, providing insight into the dependence of mission cost on key mission parameters, including science payload characteristics (e.g., mass, power), mission design parameters (e.g., ΔV), and mission risk categorization. These results can help guide NASA and others in establishing budget guidelines for a new class of low-cost, small spacecraft missions.

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 technical, programmatic and cost data on dozens of flown CubeSats missions led by NASA, research labs, and universities. The data covers a large range of costs for a variety of different CubeSat designs and mission types. With this collected data, the COMPACT team has completed a prototype cost modeling approach using k-nearest neighbors (KNN). This approach can be used to produce an early ballpark cost estimate for new CubeSat concepts that is grounded by historical actual costs and designs. This paper will demonstrate the use of KNN with the COMPACT data, and will preview our work to develop a full parametric cost model for CubeSats.

What do you do when it is necessary to generate reasonable cost estimates at the earliest Concept Maturity Levels and you have never flown any similar missions before? This paper describes the current and future Team X and A-Team cost processes and methods, how they are being used to expand our data frontiers, cost modeling capabilities and how this enables the ability to estimate early and estimate often.

The NASA Analogy Software Costing Tool Suite (ASCoT) consists of a cluster-based analogy estimator for estimating software development effort, a K-Nearest Neighbors (KNN) analogy estimator for estimating effort and delivered lines of code, a simple regression-based cost estimating relationship (CER) model that estimates cost in dollars, and a probabilistic version of COCOMO II. In this paper we document the analogy algorithms as well as summarize the results of the performance of the KNN and the principle components (PCA) cluster analogy models. KNN performance is assessed by varying the number of inputs and number of neighbors. Four different clustering methods: K-means, Spectral Clustering, Hierarchical Clustering, and Principle Components Analysis (PCA), and their respective evaluation criterion are described in detail. The comparative performance of all four estimation models is assessed using magnitude of relative error (MRE) measurements.

This paper provides an overview of the many new features and algorithm updates in the release of the NASA Analogy Software Cost Tool (ASCoT). ASCoT is a web-based tool that provides a suite of estimation tools to support early lifecycle NASA Flight Software analysis. ASCoT employs advanced statistical methods such as Cluster Analysis to provide an analogy based estimate of software delivered lines of code and development effort, a regression based Cost Estimating Relationships (CER) model that estimates cost (dollars), and a COCOMO II based estimate. The ASCoT algorithms are designed to primarily work with system level inputs such as mission type (earth orbiter vs. planetary vs. rover), the number of instruments, and total mission cost. This allows the user to supply a minimal number of mission-level parameters which are better understood early in the life-cycle, rather than a large number of complex inputs.