Optimum Risk Estimates, ORE and Aerospace Applications

Optimum Risk Estimates, ORE and Aerospace Applications

Smaller satellites are becoming frequent in the so called era of the “nano satellites”.
Canada, for example, using an Indian launcher placed simultaneously into orbit, a little more than one year ago, four satellites. Indeed, the drive to miniaturize has been a recurring theme in the space industry for size and weight translate directly into the cost of getting a satellite off the ground (setting into orbit one tonne of equipment costs approximately 17M$).

As an illustration of the “race toward miniaturization”, we have compiled the table below displaying the main characteristics of various classes of satellites.

Class	Volume (lt)	Weight (kg)	Cost (M$)	Looks like a
Mini	1000	150	100	dishwasher
Micro	275	75	12	suitcase
Nano	8	7	2	toaster
Pico	1	1.33	<<1	Sits on your palm top

ORE applied to a pico-satellite

ORE (Optimum Risk Estimates) was applied over a year ago to a pico-satellite which typically uses commercial off-the shelf electronic components. The methodologies discussed herein are general in nature and deal with the risk analysis as well as the Weight vs. Redundancy vs. Reliability optimization.

The deployment of ORE is not intended as replacement of sophisticated PRA (Probabilitic Risk Assessments) as they have been developed, for example by NASA for the Space Shuttle or other projects, but to show that, in many cases, a simplified yet systematic and conceptually sound, ISO 31000 compliant, approach can be used to replace common practices like PIG risk matrix (Probability Impact Graph), widely used in various industries world-wide, including aerospace. In other words, this study applies to cases, from various industries, including aerospace and automotive, where the paucity of available data and budget constraints do not allow the deployment of a fully pledged PRA, but there is the need for:

• a risk based decision making optimization of embarked weights, and finally
• an enhanced understanding of the contribution to the overall probability of failure of the various system’s components.

ORE differs from the usual PIG 4×4 or 5×5 risk matrix

In fact, ORE differs from the usual PIG risk matrix as it ranks risk as a function of their intolerable part. That is the “amount of risk” that is deemed intolerable by the clients for a specific project, endeavor, etc. Indeed this is very different from “binning risks” in arbitrarily bounded cells of a matrix, with limitations that have already abundantly been discussed elsewhere. Thus, a fundamental step to implement ORE is the definition of the tolerance threshold, which is client/project specific.

Also, ORE compares the relative value of the intolerable risk among a portfolio of risks potentially impacting a system, leading to a rational, transparent prioritization.
Indeed, ORE systematic approach allows to define drillable hazard and risk registers that, in turn, allow the definition of a clear road map for mitigative decision making.
ORE lead, for example, to the two diagrams displayed below.

1) Bar diagram showing the tolerable as well as intolerable risks.  These relate to complex failures such as failed switch-on, in this example, generated by each component.  In fact, the tolerable/intolerable parts (green/red color) are visible.

2) Pie-diagram showing the relative contribution to the total intolerable risk, In fact these relate to selected complex failure generated by each component.

Finally, the components’ contribution to the overall intolerable risk of a given scenario allows an in depth analysis of redundancies.  As a matter of fact, “False” redundancies are included, because this allows to study the “mitigative value” of an increased or decreased number of redundant components.

Embarked weight drives decisions

As embarked weight is certainly also a driving decision parameter, ORE allows a transparent decision-making process. In fact redundancy and resulting mitigation are on the balance with weight considerations. As each component has a weight, and if that component is introduced in the design it may alter the risks (introduce some additional redundancy), then we can develop an analysis where risk abatement is optimized as a function of the weight of the device.

Thus, at the end of the process it is thus possible to justify all the compounded and additional redundancies, or their absence, based on the ORE-weight results. In fact, this is of great importance and relevance in the optimization of aerospace, or terrestrial high performance vehicles.

Tagged with: ISO 31000, NASA, pico-satellite, PRA (Probabilitic Risk Assessments), Redundancy, Reliability optimization, satellite, tolerable/intolerable parts, “false” redundancies

Category: Consequences, Hazard, Mitigations, Optimum Risk Estimates, Probabilities, Probability Impact Graphs, Risk analysis, Risk management, Tolerance/Acceptability