Performance Assessment

To quantify performance parameters

(e.g., resolution, timeliness) for a particular approach

More >~ detailed, complex



To quantify how well the system can meet overall mission objectives


3JJ Studies with Limited Scope

The first three types of analyses in Table 3-2 provide methods for undertaking a quick-look assessment They provide limited detail, but can frequently be done quickly and at low cost Consequently, these quick-look assesments are important in any situation which is funding-limited. We will outline these methods very briefly here. However, nearly the entire book is devoted to the process of making initial estimates, which is the basic goal of limited scope studies. We want to be able to understand whether or not a particular project is feasible, and to get some idea of its size, complexity, and cost Doing this requires that we be able to make numerical estimates and undertake limited studies in order to develop insight into the nature of the problem we are trying to solve.

The biggest difficulty with limited scope studies is the tendency to believe that they are more accurate than they really are. Thus it is not uncommon to use a feasibility assessment or point design to establish the requirements for a mission in such detail that in practice the point design becomes the only alternative which can meet them. As long as we recognize the limited scope of these studies, they have a valuable place in the mission analysis activity and represent one of the most important tools that we can use to understand the behavior of the system we are designing.

Feasibility Assessment. The simplest procedure in the mission analysis hierarchy is the feasibility assessment, which we use to establish whether a particular objective is achievable and to place broad bounds on its level of complexity. Frequently, we can do a feasibility assessment simply by comparison with existing systems. Thus, we are reasonably convinced that FireSat is feasible because most FireSat tasks could be performed by existing Earth resources satellites. Similarly, it is feasible to land a man on the Moon and return him safely to Earth because we have done so in the past

We can also determine whether a particular goal is feasible by extrapolating our past experience. Is it feasible to send people to Mars and bring them back safely? Here we need to look at the principal differences between a Mars mission and a lunar mission. These differences include a longer flight time and higher gravity and, therefore, higher lift-off velocity required to leave Mars. These factors make the job more challenging and possibly more expensive than going to the Moon, but there is nothing about the Mars mission which makes it inherently impossible. Getting to Mars is feasible..The problem is being able to do so at modest cost and risk.

The third method of providing a feasibility assessment is to provide a very broad design of how such a mission might be accomplished. For example, in the 1970s, Gerard O'Neill of Princeton University proposed building large space colonies at the Lagrange points between the Earth and the Moon [O'Neill, 1974]. No mission of this scope had ever been undertaken, and it certainly was not a straightforward extrapolation of any of our normal space experience. O'Neill and his colleagues proceeded to establish the feasibility by developing a variety of alternative designs for such space colonies [Richard D. Johnson and Charles Holbrow, 1977]. While the work done was far in excess of a simple feasibility assessment, it clearly established that such colonies were feasible and gave at least an estimate of the scope of the problem.

Sizing Estimate. The purpose of the sizing estimate is to provide an estimate of basic mission parameters such as size, weight, power, or cost We can do sizing estimates in much the same manner as the feasibility assessment: by analogy with existing systems. Thus, if we are aware of an Earth observation system which has resolution and information characteristics comparable to what we believe are needed for FireSat, we can use these parameters to give us an initial estimate of the FireSat parameters.

We can provide a quantitative estimate of key mission parameters by scaling the parameters from existing missions or payloads in order to obtain estimates of the component sizes for our particular mission. This scaling process is described in Sec. 9.5 for space payloads, and in Sec. 10.5 for the spacecraft as a whole. The process of sizing by scaling existing equipment is an extremely powerful approach to estimating what it will take to achieve mission objectives. It is of use not only during the conceptual design process, but throughout the hardware design definition and development phases to evaluate the system design as it evolves. If scaling existing systems leads to the suggestion that a particular component should be twice as heavy as the current design suggests, this gives us reason to look very closely at the current design and to try to determine whether or not any factors have been overlooked. We assume that designers of previous systems did a reasonable job of optimizing their system. If the current design is significantly different, either better or worse, then we would like to understand the reasons for these differences. This is a good way to gain confidence in the design process as we proceed.

As the design proceeds, more and more accurate sizing estimates come from the scaling process. We proceed by breaking down the system into components and sizing individual components based on scaling estimates with prior systems. Thus, we may initially estimate the system as a whole divided into a spacecraft and ground station. As the design becomes more detailed, we will break down the spacecraft into its relative components and estimate the size, weight, and power of each element based upon scaling from prior systems or engineering estimates of the new system to be built. Similarly, we initially size the ground station by comparison with existing systems and eventually by building a list of all the ground system components and undertaking similar sizing estimates for each component As introduced in Chap. 1, this process of creating a list of components and estimating parameters for each is known as budgeting and is described in more detail in Sec. 10.3.

Point Design. A point design is a design, possibly at a top level, for the entire system which is capable of meeting the broad mission objectives. We refer to it as a point design if we have not attempted to optimize the design to either maximize performance or minimize weight, cost, or risk. The point design serves two basic purposes. It demonstrates that the mission is feasible, and it can be used as a baseline for comparison of alternatives. Thus, if we can establish a point design for FireSat that meets mission objectives with a spacecraft that weighs 500 kg and costs $50 million, then we can use this as a comparison for later systems. If other systems cost more, weigh more, and do not perform as well, then we will abandon those alternatives in favor of the original baseline. If we continue to optimize the design so that the cost and risk decrease, then we will let the baseline evolve to take into account the new design approaches.

A point design is valuable because we can do it quickly and easily. There is no need to optimize any of the parameters associated with the design unless it is necessary to do so to meet mission objectives. This gives us a sense of whether it will be easy or hard to meet the mission objectives and what are likely to be the most difficult aspects. One of the biggest problems in a point design is taking it too seriously at a later stage. We are always likely to regard work which we have done as representing the best approach, even though we may not have been aware of alternatives. The key issue here is to make use of point designs but at the same time to recognize their limitations and to continue to do trades to reduce overall cost and risk and to look for alternative approaches to meet mission objectives.

3.23 Trade Studies

Deciding whether to proceed with a mission should be based on a strawman system concept or point design which shows that the mission objectives can be met within the assigned constraints. Of course, the point design may not be the best solution, and we would ordinarily consider a number of alternatives. The system trade process evaluates different broad concepts to establish their viability and impact on performance and cost We then combine the system trade results with the mission utility analysis described in Sec. 3.3 to provide input for concept selection.

System trades consist of analyzing and selecting key parameters, called system drivers, which determine mission performance. We use these parameters to define a mission concept and mission architecture which can then be used for performance analysis and utility analysis as described in Sec. 3.3. The key system trades are those that define how the system works and determine its size, cost and risk. Typically, the key system trades will be in one of the following major areas:

• Critical requirements

• Mission concept

• Type and complexity of payloads

Table 3-3 shows typical examples of areas in which there are key system trades for representative missions. For essentially all missions, specification of the critical requirements will be a key system trade. For the FireS at mission, the subject is probably the heat from the fire itself and the payload is probably an IR sensor. Thus, the principal system trades are probably the mission concept, the resolution and coverage requirements, and the orbit For a mission such as the Space Telescope, the orbit is of marginal importance and the subject is moderately well defined, if only very poorly known. Here the principal trades will be the resolution and pointing requirements, the payload, and the mission concept Communications satellite systems are normally in geosynchronous oibit with a well defined concept of operations. Here the only real trade is with the required traffic load, the subject, and the size and complexity of die payload.

Truly innovative approaches—those that really change how we think about a problem—typically involve finding a new option among these key system trades. Motorola's Iridium program and subsequent low-Earth orbit communications constellations represent a new way of thinking about using satellites for communications. These have a very different concept of operations and different orbit from traditional systems. Similarly, Chap. 22 presents an innovative approach to thinking about FireSat that provides a totally different concept of operations and type of payload. Innovative solutions are never easy to come by. To tiy to find them, a good place to start is with the key system trade areas given in Table 3-3.

TABLE 3-3. Representative Areas for Key System Trades. Although these system trades are critical, we cant expect numerically precise answers to our system design problem.

Trade Area

Where Discussed


Space Telescope

Communications Satellite

Critical Requirements

Chap. 3




Mission Concept

Chap. 2





Chap. 9




Payload Type and Complexity

Chaps. 9,13





Chap. 7




We cannot normally do system trades in a straightforward numerical fashion. Choosing a different concept of operations, for example, will result in changes in most or all of the mission parameters. Consequently, the fact that Option A requires twice the power of Option B may or may not be critical, depending on the orbit and number of satellites for the two options. We need to look at the system as a whole to understand which is better.

The best approach for key system trades is a utility analysis as described in Sec. 3.3. We use the utility analysis to attempt to quantify our ability to meet mission objectives as a function of cost We then select the option which fulfills our objectives at the lowest cost and risk. As described in Sec. 3.4, this is still not a straightforward numerical comparison, but does have quantitative components.

The simplest option for system trades is a list of the options and the reasons for retaining or eliminating them. This allows us to consider the merits and demerits at a high level without undertaking time-consuming trades. This, in turn, allows our list to be challenged at a later date. We should go back to our key system trades on a regular basis and determine whether our assumptions and conclusions are still valid. It is this process of examination and review that allows us to use technical innovation and new ideas. It is a process that must occur if we are to drive down the cost of space systems.

The alternative to simply articulating trade options or conducting a complex mission utility analysis is a system trade in which we make a quantitative comparison of multiple effects. This can be particularly effective in providing insight into the impact of system drivers. For the purpose of trade studies, system drivers generally divide into two categories—those for which more is tetter and those with multiple effects. By far the easier to deal with are the "more is better" drivers, for they simply require us to ask: "What is the cost of achieving more of the commodity in question?^ For example, in a space-based radar, added {rawer improves performance but costs more money. Thus, the designer will want to understand how much performance is available for how much power. A second example is coverage. For virtually any Earth-oriented system, including our FireSat example, more coverage means tetter performance at higher cost. Increasing coverage ordinarily means adding satellites or, perhaps, increasing a single satellite's coverage by increasing its altitude or the range of its sensors. Therefore, we often do a coverage trade considering performance vs. number of satellites, substituting the latter for cost Assessing performance as a function of power or coverage may take considerable work, but it is relatively easy to present the data forjudging by the funding organization, the users, or other decision makers.

System drivers and critical requirements which cause multiple effects demand more complex trade studies. Pushing parameters one way will improve some characteristics and degrade others. In trades of this type, we are looking for a solution which provides the best mix of results. Examples of such trade studies include instrument design, antenna type, and altitude. Each antenna style will have advantages and disadvantages, so we must trade various possible solutions depending upon the end goals and relative importance of different effects.

In trades with multiple effects, selecting the correct independent parameter for each trade is critical. Consider, for example, selecting either a reflector or a phased-array antenna for a space-based radar [Brookner and Mahoney, 1986]. From the radar equation, we know that a principal performance parameter for a radar is the antenna aperture. All other things being equal, larger antennas will provide much better performance. Thus, for our example, we might choose to compare reflector and phased-array antennas of equal aperture. On this basis, we would choose the phased array because its electronic steering makes it more agile than a reflector antenna, which must be mechanically steered. But our choice becomes more complex when we recognize that weight typically limits large space structures more than size does. Generally, we can build a reflector larger than a phased array for a given weight Based on weight, a reflector may have considerably more power efficiency and, therefore, be a better radar than a phased-array system. Thus, we would have to trade the better performance of a larger reflector vs. the better agility of a smaller phased array. Depending upon the application, the results may be the same as for an aperture-based trade or reverse. The important point is the critical nature of selecting the proper independent variable in system trades. To do so, we must find the quantities which inherently limit the system being considered. These could be weight, power, level of technology, cost, or manufacturability, depending on the technology and circumstances.

Table 3-4 summarizes the system trade process for parameters with multiple effects. Typically the trade parameter is one of our system drivers. We begin by identifying what performance areas or requirements affect or are affected by the trade parameter. For example, the altitude of the spacecraft will have a key effect on coverage, resolution, and survivability and will be limited by launchability, payload weight, communications, and radiation. We next assess the effect in each of these areas and document and summarize the results, generally without trying to create a numerical average of different areas. Figure 3-1 shows this step for FireSat. We use the summary to select the parameter value and a possible range. Although the process is complex and may not have a well defined answer, it is not necessarily iterative unless we find that the results require fundamental changes in other system parameters.

TABLE 3-4. System Trade Process for Parameters with Multiple Effects. The example is the altitude trade for the FireSat mission. See also Fig. 3-t.
0 0

Post a comment