Cost Modeling for Space Based Telescopes

When NASA project leaders decide to send a new high-tech telescope into space, the planning process requires that trade-off decisions be made between design features and cost. As an example of a trade-off decision, note that a telescope requires a large mirror. Sending a large mirror into space can be expensive, so it might be desirable to send several small mirrors instead, and then combine the images in the small mirrors into one large image. The decision then has to be made, will there be one mirror or several, and if there will be several, how many?

A cost model helps decision-makers make these trade-off choices with information about what each decision will add to the total budget of the telescope. These telescopes can cost a lot of money, and often the early design decisions are the main contributors to the cost.

Using statistical methods, a cost model for land-based telescopes was made recently, and NASA project managers want a similar model for space-based telescopes. The space-based telescope cost model will be more difficult than a land-based model, because there are fewer telescopes in space than there are on land. This means that there are very few data points to use to create the model. This is a problem because the accuracy of a model depends on how much information was used to create the model.